The best ways for publishers to build credibility through transparency

As so much in the world of news and information changes, the fundamental bond of trust between journalists and the communities they serve is one of the few things that doesn’t. In fact, its importance has grown.

One of the most important ways journalists and news organizations earn the trust of the public trust is by being transparent about who we are and the work we do.

Every newsroom, and basically also every single story must show why they deserve more trust than dozens or even hundreds of others on the same topic.

We attribute information to the source to show provenance. We have bylines and credits to provide a sense of ownership and accountability. We offer opportunities for people to respond to what they read, hear and see. We invite the public to report errors and request corrections, and we publicly admit our errors.

Today, journalists are rapidly adapting and enhancing these basic frameworks for transparency and accountability, both to suit digital platforms — and for a world of abundant sources of information.

This paper is part of the American Press Institute’s ongoing series of Strategy Studies, deep examinations of how publishers can build new revenue models and grow audiences. The studies draw insights from multiple examples, focusing less on the examples themselves and more on the lessons and actionable insights for others to borrow. They are designed to be pragmatic and realistic, noting potential obstacles and emphasizing the how-to elements.

For this Strategy Study we talked with experts and newsroom leaders, and reviewed academic literature and research findings, to examine how practices related to transparency and credibility are evolving in five key areas:

  1. Show the reporting and sources that support your work
  2. Collaborate with the audience
  3. Curate and attribute information responsibly
  4. Offer disclosures and statements of values
  5. Correct website and social media errors effectively

What is transparency, really?

Transparency has emerged as one of the most-discussed and evangelized aspects of practicing ethical journalism in the networked age.

As traditional notions of journalistic objectivity are challenged and also criticized, transparency has emerged as an ideal that newsrooms and individual journalists strive for. But it can also be overhyped and offered as a panacea.

If journalists are truth seekers, it must follow that they be honest and truthful with their audiences, too.

“Transparency is more than a buzzword,” wrote journalism professor and ethicist Stephen Ward in an article, “Why Hyping Transparency Distorts Journalism Ethics.” “Too often it is a magical idea — a norm with seemingly magical powers to restore democracy. It is a ‘god’ of institutional ethics.”

Rather than exploring abstract — or “magical” — notions of transparency, this study focuses on tangible actions and practices. For example, journalists can make a greater effort to link to original source material, to offer ways for the audience to participate in the newsgathering process, or to be responsive to requests for correction.

The digital environment provides journalists new and more effective ways to practice transparency. Well-established transparency practices can be updated and adapted to a multi-platform world, and new, digitally native ones can also be applied.

Transparency adds value to the work we do, but it’s also a fundamental part of how we do our work, according to Bill Kovach and Tom Rosenstiel in their book, “The Elements of Journalism.” (Rosenstiel is the executive director of the American Press Institute.)

“If journalists are truth seekers, it must follow that they be honest and truthful with their audiences, too — that they be truth presenters,” they wrote. “If nothing else, this responsibility requires that journalists be as open and honest with audiences as they can about what they know and what they don’t. How can you claim to be seeking the truth when you’re not truthful with the audience in the first place?”

Transparency also has other virtues, according to the authors. “It signals the journalist’s respect for the audience” and “… also helps establish that the journalist has a public interest motive, the key to credibility.”

It’s important to be realistic about the impact of transparency. It will not necessarily result in immediate changes in traffic, or engagement, though some initiatives will certainly have positive effects. It is not a cure-all, but it does have value.

Here’s how an editor at The Washington Post put it in an interview for a recently published academic paper, “Newsrooms and Transparency in the Digital Age”:

We are really aware that people have very high levels of suspicion about the media generally. Rightly or wrongly, the public does not trust us and so we have to make an effort to “show” readers that we are professional in the way we do our job … also there’s definitely the aspect of competition … what we are selling in effect is our credibility compared to say a blog or a smaller outlet and so being more forthcoming is one way of doing it.

In the same paper, an editor at the Los Angeles Times said that “transparency is telling people how we got the story … reporting why we included what we did. What was left out and why, who our sources are.”

Implementing processes, policies and tools that enhance accountability and credibility are not just good journalism — they can be a competitive advantage, a differentiator.

“[I]n the digital world, where information is infinite and infinitely replicable, being transparent about provenance and sourcing helps distinguish journalism from other content on the web,” wrote Martin Moore, the executive director of the Media Standards Trust, in a blog post that listed the arguments in favor of transparency.

Mathew Ingram, who covers the media for the website Gigaom and was formerly the first community editor of Canada’s Globe And Mail, has compared aspects of practicing transparency to a disaster preparedness plan.

“Its value only becomes obvious in extreme circumstances,” he said in an interview for this study. “… 99 percent of the of time no one will look, but then something does happen and they can go and look.”

Implementing processes, policies and tools that enhance accountability and credibility are not just good journalism — they can be a competitive advantage, a differentiator.

Transparency must therefore be linked with consistency: determine how you will practice transparency and the areas in which you will apply it — and then keep doing it. By making it a habit, it has more of an effect both internally and externally. And it provides the kind of worst-case scenario protection cited by Ingram.

In a 2009 paper, “Transparency in Journalism: Credibility and trustworthiness in the digital future,” Dr. Klaus Meier of Catholic University Eichstaett-Ingolstadt argued that “evidence of trustworthiness must be given repeatedly” in a competitive digital environment.

“Every newsroom, and basically also every single story must show why they deserve more trust than dozens or even hundreds of others on the same topic,” he wrote.

Kovach and Rosenstiel highlighted two key transparency questions journalists should ask themselves in the course of their work: “What does my audience need to know to evaluate this information for itself? And is there anything in our treatment of it that requires explanation?”

These go to the heart of transparency.

In the digital world, there is a third question to add: What elements of this story can I share that will help the audience contribute valuable information or perspective to improve the reporting?

Chapter 2

Show your sources

One important way to build trust in reporting is to show the audience the sources it relies upon. There are several ways to do this.

I. Share source material

Digital platforms don’t suffer the same time and space constraints as print, television and radio. This freedom offers new opportunities to incorporate and disclose source material, and to be clear about content changes.

One core feature of the web — the hyperlink — offers an easy and powerful way to point to the sources and information that have played a role in telling a story. Links are way to offer proof, in effect footnotes, without breaking the flow of narrative work or adding unwanted narrative to shorter or more visual forms. Because of this enormous advantage, anyone producing journalism in the digital age must make linking out to sources a fundamental practice that applies to all work.

Use workflow checkpoints to force linking

Articles, multimedia projects, video, apps, and other pieces should incorporate links that point to sources and additional information. This is both good journalism and a service to the community.

A simple rule for linking might be this: if a claim or fact was gathered and verified online, it should be supported by a link.

The Register Citizen in Connecticut has evangelized the use of links as part of its effort to shift to a digital workflow, explained Matt DeRienzo, the state’s group editor for Digital First Media.

A simple rule for linking might be this: if a claim or fact was gathered and verified online, it should be supported by a link.

“We audited our site in terms of how much we were linking out to sources and showing the studies we were quoting,” he said.

They saw work to be done, so the paper’s managers implemented a new practice: Copy editors send reporters back any story that didn’t include links. (The editors also make an effort to add links.)

“That has improved things tremendously,” DeRienzo said. “We’re trying to make that part of the DNA.”

List all the sources

Since its earliest days, PolitiFact, the Pulitzer Prize-winning fact-checking website, has listed all of the sources it consulted in putting together each fact-check. This list is published in addition to including links within the text of fact-check articles.

“We include the interviews we do, the documentation we look at, any studies we reviewed, any news reports we looked at,” said PolitiFact editor Angie Holan.

Each fact-check must have an accompanying source list, or the reporter is told to go back and show their sources, she said. There is also a field in the PolitiFact CMS where the sources are entered, and a fact check can’t be published unless it has been filled in, according to Holan.

“I think it would be hard to get people to do if we had to nag them to do it and they could literally publish without doing it,” she said.

This practice provides a roadmap to curious readers who want to know how PolitiFact arrived at its verdict. It helps build credibility.

As shown in this example, PolitiFact's source list can be extensive. It includes the publication date (down to the minute) of the fact-check, as well as the publication information and hyperlink for each item. Finally, it includes a full list of the journalists responsible for the fact-check. It in effect says,

As shown in this example, PolitiFact’s source list can be extensive. It includes the publication date (down to the minute) of the fact-check, as well as the publication information and hyperlink for each item. Finally, it includes a full list of the journalists responsible for the fact-check. It in effect says, “This is who we are, and how we did this work. See for yourself.”

“In a skeptical age people do not take even respected media reports at face value,” Holan said. “They are very cautious about accepting what the media says, and I think they want to investigate for themselves … in other cases it merely shows people that we looked at the relevant evidence.”

Another benefit of making a list of sources part of every fact check is that it provides PolitiFact with a repository of useful sources.

Holan said the newsroom’s internal source list where reporters are supposed to enter contact information for sources is rarely updated. Making the sources public ensures they have a place to revisit useful information.

“If [the source list] was not public, it would not get done,” she said. “We use the source list ourselves just as much [as the public].”

Show the source documents

ProPublica’s Explore Sources goes one step further in disclosing sources.

It’s a tool that integrates source document excerpts into a story, so readers can view related documentation as they go. The documents must first be uploaded to DocumentCloud, a free service. (As of now ProPublica has not open-sourced the code for Explore Sources)

Explore Sources debuted in a 2011 story that incorporated excerpts from 64 documents that reporter Marshall Allen gathered. He used Explore Sources to link specific excerpts to relevant passages in his story.

Explore Sources shows the source document as a pop-up window when a reader clicks on a highlighted passage in the story.

Explore Sources shows the source document as a pop-up window when a reader clicks on a highlighted passage in the story.

II. Show changes to content

An oft-repeated maxim today is that “journalism is a process, not a product.”

The stories we report don’t stop just because we’ve hit publish. They are updated to incorporate new information, and these additions should be disclosed.

One model for capturing the changes to content over time is Wikipedia. Every article on its site includes a “History” tab at the top that shows all of the changes made to the article, and by whom.

For example, the history of the Wikipedia article about Joseph Pulitzer can be accessed here. This level of total transparency is great; but the page is difficult for the average user to understand.


A more user-friendly way to show a revision history would be welcome, but as of now this is not something news organizations are putting effort into. (BuzzFeed’s deputy editor-in-chief, Shani Hilton, told the Poynter Institute the web publisher is considering it.)

For now most news organizations handle updates in one of two simple ways:

  1. Add a note at the top or bottom of the piece of content that begins, “Update: This article was updated to … ” and that goes on to list the basic nature of the update made. Then the updated information itself is incorporated into the content by rewriting/adding as needed.
  2. Add an update note at the top or bottom of the piece that details all of the new information, thus leaving the original content intact.

The first option places all of the updated information within the context of the content itself and simply adds a note of disclosure.

The second places all of the new information within the update note itself, preserving the rest of the content as it was originally published. In both cases, it’s best practice to add the date (and even the time) of the update, so a reader can see how the story evolved.

Both options note clearly that content has changed from when it was first published. This is a critical difference with old tradition of a “writethru” that would see a story rewritten again and again without any record of the changes being made. Similarly, TV and radio reports would often be updated or re-cut with new information, and not include a note about previous versions.

Just because we can update something online without it being evident to the audience doesn’t mean we should. The principle of transparency suggests that online it’s essential to note when and how content is changing because it continues to exist in the same place, and because people may be linking to it. Adding updates also gives a reason for people to come back and read the content again, thereby increasing traffic and engagement.

Online it’s essential to note when and how content is changing because it continues to exist in the same place, and people may be linking to it.

For particularly fast-moving stories, many organizations are moving away from static article pages toward using live blog tools. Live blogs automatically capture the chronological timeline with full attribution and time stamps, offering automated transparency.

Core principles

When adding updates to content, consider these five best practices:

  1. Be clear about the nature of the update. Why are you updating the content, and what information has been added? An update note that only reads, “This article was updated to reflect new information” does not serve the reader.
  2. Note when and why was it added. Make it clear why you added new information by citing the source and the addition. Also indicate the date and time of the update. This communicates to the reader a sense of time and place.
  3. Think about readability. Clarity is the first priority in an update. But the placement and overall readability is also important. If you have a major update that renders much of the previous content obsolete, the best option may be to add the update at the top and include all of the new information. If you expect a story to require multiple updates, flag that for the reader and think about how you can set up the page so that they can easily understand how the story is progressing. (Or think about a live blog.)
  4. Have a review process. Who signs off on the update? It’s unwise to have reporters add information or updates without it being approved by an editor or producer.
  5. Updates are not corrections. If you’re making changes due to a factual error, critical omission or other mistake, don’t label it an update. If you are fixing an error, then it should be labeled a correction. (See section five of this study for advice on making digital corrections.)

Obstacles to expect and overcome: If your organization has been producing content for legacy platforms, including links or publishing source lists may not be standard operating procedure. As shown by The Register Citizen and PolitiFact, the way to change the culture and make linking a regular practice is to require reporters or producers to add them to all work, and, like PolitiFact, to even build this into the CMS. It has to become a habit. Some journalists may also push back on noting updates to content. A first step is to create boilerplate update text that can be used as the basis for adding updates, making the process easy to do. A second step is to create a clear policy that fast moving stories go to a live blog or another format better suited to updates, thereby preventing journalists from having to make repeated updates.

Chapter 3

Collaborate with the audience

Journalist Dan Gillmor likes to say, “my readers know more than I do.”

Implicit in Gillmor’s axiom is a reminder that journalists shouldn’t think of people only as consumers. The people we serve have collective knowledge and expertise that can vastly improve the work we do. Digital platforms make this easier.

It’s not necessary to be transparent about everything. Nor is it necessary or advisable to look for community collaboration on every story.

The key is to find productive, mutually beneficial and meaningful ways to include your community in the reporting and publishing process.

This new more mutual relationship starts with a willingness to share elements of what you’re working on before you’re ready to publish. This can be a painful process in some newsrooms. People will ask, “Aren’t we just going to tip off our competitors to what we’re doing? Won’t we be giving advance notice to any targets of an investigation if we talk about it too soon?”

These are reasonable concerns. It’s not necessary to be transparent about everything. Nor is it necessary or advisable to look for community collaboration on every story.

Joy Mayer, director of community outreach at the Columbia Missourian, said she has at times heard pushback from journalists who think they will give something away by publicly stating what they’re working on. Or they worry people will be upset if a story they talk about publicly never comes to fruition.

“Those are really valid concerns, but they apply only in a surprisingly small number of cases,” Mayer said.

Based on the experience of Mayer and others, journalists should think carefully about the details you share if:

  • It could turn out to be false.
  • It could put sources or other people in harm’s way.
  • It could compromise people’s privacy.
  • It will undo your work by tipping off people or entities to what’s coming.

Mayer said those circumstances don’t apply to most stories. Even if one of them does, the story could still be a candidate for transparent collaboration, so long as you focus on the specific aspect where community collaboration can add value and won’t result in unwanted consequences.

This speaks to one of the key things to keep in mind: focus. People are busy and are not seeking out ways to spend time helping you and your organization.

“People aren’t obsessed with your process unless it really affects them,” said DeRienzo of Digital First Media. “They’re not necessarily going to stick their hands there in the sausage and help you make it.”

You need to think about where there is mutual value in collaboration, and how you execute in a way that realizes this value.

As a starting point, Mayer developed a set of questions she provides to reporters who are thinking about how the community might be able to participate with a story:

  1. Who’s already talking about what you’re covering? Where or how (offline and online) are those conversations taking place?
  2. Whose experience or expertise could help you in your reporting? What sources are you looking for, and how could we get creative about finding them? (This could be specific people, or communities of people.) Or should we invite someone to contribute their own voice as a companion to your story?
  3. Is there an opportunity for — and would there be benefit from — letting the community know what you’re working on as you’re still reporting? Is there any danger in doing that?
  4. What do you hope your story will accomplish? Is there conversation that might (or should) follow? If so, what should we do to facilitate or be a part of that?
  5. Who’s your target audience? Who do you think most needs — or would most enjoy — the story you’re telling and information you’re providing? How can you make sure they’re invited to see what you produce, and interact with it?
  6. What can the audience DO with your story, or in response to it?

With that in mind, here’s a look at three common ways news organizations are inviting collaboration from the community.

I. Invite input on the story and assignment process

It starts with ideas.

Newsrooms put effort into generating story and project ideas, and this early stage is one place where organizations can bring in the public. Digital First Media’s DeRienzo has experimented with putting the day’s story budget online every morning so members of the community can see what the staff is working on.

Similarly, the Columbia Missourian often posts a list of assignments to its Facebook page.

The goal is to encourage community members to share their thoughts on the planned coverage, and to add any information they might have. Starting in 2011, The Guardian also experimented with making its daily story budget available online. One takeaway from that experiment was that it’s helpful to pair each story with a specific call to action for the audience.

“When we ask people to just suggest a news story, it’s such a huge subject people don’t really know where to start or what we might mean,” Dan Roberts, who was at the time the national editor of British-based Guardian, told Journalism.co.uk. (He is now the Guardian’s Washington bureau chief.) “But when we say we’re writing about the [National Health Service] reforms tomorrow and we’re looking for people who have experience of it first hand because they work in a hospital, for example, we get much more practical responses.”

A simple list of stories may elicit a few comments, but a specific request tied to each (or at least some) can focus people’s responses. They can clearly see how to add value.

In some cases, a broad call out for story ideas can be successful — provided it’s properly supported and managed. One example of this is the Curious City project run by WBEX public radio in Chicago.

Its mission is “to include the public in editorial decision-making, make journalism more transparent and strengthen multimedia coverage about Chicago, the surrounding region and its people (past or present),” according to the project’s website.

Curious City does this mainly by inviting members of the community to suggest story ideas – in the form of questions — that other people can then vote on. The questions that are selected are then reported out by the project team, who also share the progress of their reporting.

It’s a five-step process, according to the station’s website:

  1. Questions come from the community.
  2. You vote for your favorites.
  3. WBEZ investigates, posting updates in real time.
  4. Followers help shape the investigation.
  5. We discover the answers together

To make this work, the project has three committed team members who work hard to generate questions. They also to recognize the community members who come up with questions that turn into stories, as in this Facebook post:


Curious City illustrates the importance of matching calls for collaboration with consistency, adequate resources, and a commitment to recognize and reward those who participate.

II. Encourage participation

Crafting the right prompt or request is something the team at De Correspondent in the Netherlands works on regularly with its team of correspondents.

The online Dutch publication set a world record for journalism crowdfunding after bringing in more than one million Euros in eight days in 2013. At the core of its concept is the idea that its journalists own and manage their own beats, and that there is “a permanent, two-way relationship between you and your audience,” said editor in chief Rob Wijnberg.

One of the items listed on Wijnberg’s 10-part manifesto for De Correspondent was, “From readers to participants.”

Transparency is part of De Correspondent’s business model and product offering.

Each journalist has her own “garden” on the site, where paid members can see what the journalist is working on, share ideas, respond to prompts and invitations for information, and read the finished work. Only paid members can see all of the journalism-in-process content and participate in conversations. (Wijnberg said finished stories “can be freely shared by members and correspondents with anybody.”)

In this respect, De Correspondent has made the elements of transparency and participation a value-add for paying subscribers. They receive additional access to the journalists and their works-in-progress. Transparency is therefore part of De Correspondent’s business model and product offering.

Wijnberg said their correspondents are encouraged to include specific prompts for information and discussion as a way to help them develop their reporting.

“If you ask more direct questions then people respond to that,” Wijnberg said.

How to handle contributions responsibly

By asking others to share and open up, we must do the same.

In putting out a call for participation, you take on responsibilities: to use the information provided as you said you would, to accurately credit those who participate, to secure all necessary permissions, and to be responsive and communicative during the process.

Wijnberg said that rather than being seen as “extra” work, journalists must understand that this kind of transparency and collaboration is now part of their job.

Mayer, of the Missourian, encourages her reporters to phrase their invitations for collaboration in a way that doesn’t leave some potential participants feeling left out.

“We work on what that prompt is,” she said. “You want most people to be able to see themselves in it.”

The key, she said, is to appeal to the “universal particular” — things that can be as universal as possible, but that open the door for people to share their particular experience or expertise.

Mayer also said there are times when the goal is just to elicit a few specific sources or responses, rather than a flood of contributions. This requires the right expectations, and also an approach that considers their privacy.

“If you’re looking for low-income families who use services for special needs kids, you’re not going to get Facebook comments saying ‘Me, me!'” she said.

III. Tweet your beat

Another way reporters are inviting the community in is by tweeting their beat.

This can take various forms — from looking for experts to soliciting questions, or sharing elements of the reporting process as a journalist goes about it.

It’s common to see reporters tweet (or post to Facebook) that they are planning to interview someone, and to invite any questions from their followers. This is also popular with television outlets. They will solicit questions for interviews and ask people to use a hashtag to submit them.

Journalists involved in this work identify three things to keep in mind with this approach:

  1. It’s not as helpful to put out a request for suggested questions when you’re about to start the interview. Give people enough notice to think and submit.
  2. Soliciting questions and feedback brings with it a responsibility to respond to people.
  3. If someone’s question does get asked, it’s good practice to reply and pass along the answer. This encourages them to participate in the future, and shows others that the reporter and news organization are serious about collaboration.

IV. Invite mass collaboration

In some cases, news organizations can only deliver a specific piece of reporting if they secure broad participation from the community. This is often the case when trying to process large amounts of data (that can’t be done effectively by a machine), or when the community must generate the data itself.

When ProPublica decided to investigate political ad buys in swing states during the 2012 presidential election, for instance, editors knew they couldn’t process all the data on their own. The available data consisted of thousands of unstructured PDFs containing information from different TV stations.

The only way to combine the data to get a picture of who was spending what, where, would be to get people to review the documents and enter the necessary information into a database. The resulting project was called Free the Files. An effort of this size required ProPublica to not only think about how to encourage participation, but also about how they can make the process as open and transparent as possible.

One key decision those involved in the project at ProPublica made right away was to narrow the scope of the project and only focus on specific swing states, rather than the whole country. They also picked a small set of important questions they wanted answered about each document. This gave focus to the task, and limited the time and effort required of participants.

Here’s what the interface looked like for a participant, with the document on the left and the questions on the right:


They also made sure to design a site that immediately showed new and returning visitors the progress that was being made. Disclosing the statistics helped encourage participation.

“When a visitor arrives at Free the Files, one of the first things they see are project metrics: the number of files ‘freed’ and the total ad spending logged by our volunteers,” wrote Amanda Zamora, ProPublica’s senior engagement editor, in a blog post. “These were our success metrics, touted each day to keep the momentum going and spur new participation.”

In an interview, Zamora emphasized the importance of constantly communicating to participants that their work was making a difference.

“Being able to see what your review rate was and where you fit into the overall picture, based on how many documents you reviewed related to your peers, gave people a reason to come back,” she said.

They also built in ways for people to broadcast their participation to social networks, and ProPublica organized “sprint” events to inspire people in a specific geographic area to help free their local files, according to Zamora.

When it was all over, 880 people had reviewed at least one file, logging over $650 million in swing state ad buys. At the conclusion, Zamora identified the top 10 contributors and wrote a blog post to celebrate their effort, and to give each person a chance to talk about why they participated and what they learned.

This reinforces the importance of recognizing, rewarding, and encouraging people as part of the open collaboration process.

V. Encourage people to report errors

One of the ways the public has always collaborated in the reporting process is by speaking up when they see a mistake.

At one time, they would write a letter, or pick up the phone. Today, these requests for correction can flow in by email, on Twitter and Facebook, or via specialized error report buttons or forms offered by news organizations.

Some people are already in the error-spotting habit, but there are easy ways to encourage others to help you eliminate mistakes from your work, while reinforcing that you take accuracy and accountability seriously. (Section five of this study offers advice on doing digital corrections.)

The error report button or form

The Toronto Star is one of several organizations to have integrated a “Report an Error” button on all content. It appears in a drop-down drawer that’s part of the Star’s article tools menu, as shown in this example:


A reader clicks on that text and is shown a pop-up menu where they can fill in the details of what they saw:


“The idea was to give readers a quick way to point out a factual error to help us act more quickly in catching mistakes, and a simple way for readers to register a complaint about an article,” said Star public editor Kathy English, who added that the form has been part of the site since 2006. “Giving readers such easy means to report errors in effect makes every reader a fact checker working with us to ensure accuracy in the Star.”

English said completed forms are sent directly to her office, enabling her to quickly investigate and correct.

Error report forms have been used by the Chicago Tribune, Columbia Missourian, Washington Post, Huffington Post and Global News (Canada), among other organizations. The forms have a dual benefit of providing evidence of an organization’s commitment to being accountable for mistakes, while also nudging readers to report errors in a consistent and therefore more manageable way.

A form makes it easy for people to understand where to send their requests for corrections, and it also enables the organization to include the necessary fields that help it evaluate the requests.

When the Washington Post launched its error report form in 2011, its then-Managing Editor Raju Narisetti told Poynter, “This increases engagement because we’re being responsive to readers, and there’s significant value in that.”

Canada’s Global News launched an error report button on all content in 2014. Its national director of editorial and online news, Rob Waksman, described the experience for readers:

When you click on ‘Report an Error,’ a form comes up providing visitors with the opportunity to offer a correction or provide additional information and story ideas to enhance our coverage. The feature immediately connects users with the reporters and producers who worked on the story and allows them to make quick corrections.

Crediting and rewarding participation

Each month, Maggie Walter contacts a lucky reader of the Columbia Missourian’s website to let them know that they’ll soon be receiving a prize package from the paper.

Walter, an interactive news editor, runs the paper’s monthly Show Me The Errors contest that rewards one eagle-eyed reader for helping point out typos and factual errors.

“Show Me the Errors invites readers to join the editing staff of ColumbiaMissourian.com and have an opportunity to win prizes,” is how the contest page describes the effort.

To encourage participation, the paper has an error report button at the foot of all content. It encourages readers to submit what they find, and it also explains the contest.

Readers receive one entry in a monthly drawing for each correction request they submit. The contest used to automatically give the prizes — a T-shirt and the copy of a book — to the person who identified the most errors. But that had to change thanks to one prolific error-spotter.

“We consistently had the same winner, and he didn’t have need of so many copies of the same book,” Walter said.

Walter emphasized that it can take internal effort to get staffers to encourage people to point out errors and that consistency is key.

“You need to keep pushing it as much as you can and you have to be positive about it,” she said in 2012. “It feels like people are picking on you when you first start, but you have to remember the people sending these [error reports] are helping.”

She helps reinforce that point by announcing the name of the winner each month in the paper’s “Dear Reader”column.

While the Missourian’s contest is rare, it’s increasingly common for news organizations to credit a name, or Twitter or commenter handle, when adding a correction. The San Francisco Chronicle’s pop culture critic did this when a reader helped him fix an error about an old video game:

CORRECTION: An earlier version of this post suggested that a singular being named Yar was getting his revenge in the Atari 2600 game Yars’ Revenge. In fact, the Yarians were a race of aliens, and were collectively seeking revenge. The Big Event apologizes for the error. (Thanks to TBE reader Marty for the e-mail pointing this out.)

This echoes an approach advocated by Wijnberg of De Correspondent. He said corrections should be expressed with an element of gratitude for the reader who spotted the mistake, rather than with defensiveness.

“The more defensive your attitude, the worse it gets,” he said. “We get a lot of responses from readers when we correct mistakes … We really see people reacting in a very positive way. They actually feel like we listen.”

Establishing a correction workflow

One area where news organizations often struggle is in establishing a corrections workflow that gathers error reports, efficiently reviews them, and results in the proper response. Without one, it’s impossible for an operation to deliver the promise to be consistently transparent or collaborative.

A correction workflow has five core components. In creating a workflow, the goal is to establish a clear path that enables everyone in a newsroom to know their role, and to play it effectively.

1. Collection of requests. Prepare for all of the ways in which people will submit requests for corrections, and think about how you can effectively collect these into a relatively frictionless or streamlined processing queue. Typically, requests for correction will be sent by email to reporters, editors and to general email accounts; via tweets and Facebook comment on a post; by phone; and by people commenting on the piece of content on your website. They will also come in via a form if you set one up.

2. Evaluation of validity. Once received, how will it be evaluated? The New York Times’ Ethical Journalism handbook [PDF] directs reporters to pass requests for correction “promptly to a supervisor.” The more systematic that process can be made, the more consistent and transparent the process can be. This might begin with categorizing the nature of the request. Is this a factual error, or more of a difference of opinion? If it is a claim of error, determine who is responsible for checking it? (This could be done by a corrections guru, an editor or by the reporter.) What guidelines do you have for determining whether a correction is required? Finally, regardless of who does the evaluation, the conclusion must be communicated to the original editor and reporter, in case there is disagreement.

3. Drafting and approval. Many organizations have one unit responsible for writing corrections. Who sees those? How are they edited? As with any content, only more so in the case of one generated by a claim of error, it should be reviewed for accuracy and grammar, and someone should have the final approval before it’s published.The Denver Post directs reporters to speak to a supervisor to discuss and approve corrections:

When an error is discovered – whether it is detected by a member of the public or a staff member – it should be discussed immediately with your supervisor and corrected as soon as possible.

If there is a dispute over whether something is incorrect, a supervisor should be consulted to resolve it. Correction forms should be filled out and turned in to your supervisor.

BuzzFeed’s correction policy is also specific that an editor or team leader must approve corrections:

DON’T add a correction without first running the proposed correction by your editor or team leader. Vertical editors can run corrections by management. And when you approve a correction, please cc management so we can keep track.

4. Publication. The correction is placed on the offending piece of content, in the location designated by the policy. (See section five of this study for advice about correction placement.)

5. Promotion. This is an increasingly important step with digital corrections, and an important consideration when thinking about the intellectual honesty implicit in trying to be transparent. If a fundamental goal of every correction is to help spread the truth, then there is a duty to promote the correction. This can be done by the reporter herself on social media, and also via organization accounts. Part of this process can also include placing the correction on a dedicated online corrections page. This page is where organizations list recent corrections. One way to make this page useful for the public is to have it linked clearly from the homepage. It’s yet another way to demonstrate your commitment to being transparent and accountable. Here, for example, is The New York Times’ online corrections page.

The Register Citizen in Connecticut maintains a blog dedicated to bringing attention to its recent corrections.

“It’s not just about trust, it breeds more accuracy,” DeRienzo said. “If everyone sees that on a regular basis that you are correcting stuff then they will point out things in the future because they know you are correcting stuff.”

Obstacle to Expect and Overcome: Pushback from journalists about the time it takes to interact with the audience. This is a common concern in today’s overworked newsrooms. Wijnberg said the first thing is to communicate that this isn’t “additional work.” It’s part of the journalist’s job. “Basically, they have to find a new balance between what they thought was their full-time job and these new elements,” he said. He suggested that reporters think about when it’s the right time to step back from the interaction and let the conversation carry on (or conclude) on its own.

Chapter 4

Practice ethical curation and attribution

One of the oldest and most hallowed forms of transparency in journalism is attribution. You link what was said to the person who said it. You cite your sources. You provide a roadmap that leads back to all the people you spoke to, the documents you read, the other articles and research that helped form your work.

As noted in the first section of this study, the digital environment enables us to take attribution to the next level by linking out to sources. This takes on even more importance when we aggregate/curate (I’ll use those terms interchangeably for this section) the work of others. This can include utilizing user-generated content from social media, or otherwise creating new works based on the gathering and combining of previously published work.

“There’s often a thin line between aggregation and theft,” cautioned Bill Keller, the former executive editor of The New York Times, in a 2011 blog post.

Steve Buttry, a longtime newspaper editor, journalism teacher and recently the digital transformation editor for Digital First Media, argued in response that in fact “aggregation has a long, proud and ethical history in journalism.”

He continued:

The New York Times and Washington Post also have long histories of aggregation. In my years at various Midwestern newspapers, we reported big local and regional stories that attracted the attention of the Times, Post and other national news organizations. Facts we had reported first invariably turned up in the Times and Post stories without attribution or with vague attribution such as “local media reports.”

Certainly there is a line between responsible and transparent curation and whole appropriation. The digital age and the concept of transparency make walking that line more important than the era in which the Associated Press picked up and rewrote stories from local newspapers for redistribution to member TV, radio and print clients.

The starting point for ethical aggregation is to practice attribution. The NPR Ethics Handbook provides clear direction on the standard for attributing information:

When in doubt, err on the side of attributing — that is, make it very clear where we’ve gotten our information (or where the organization we give credit to has gotten its information). Every NPR reporter and editor should be able to immediately identify the source of any facts in our stories — and why we consider them credible. And every reader or listener should know where we got our information. ‘Media reports’ or ‘sources say’ is not good enough. Be specific.

The principle and practice of showing our work builds trust and credibility, and it’s therefore essential to clearly indicate when information comes from elsewhere. This can be done by using quotes or the blockquote function in a CMS, by citing a source by name, and by linking out whenever possible.

There is another element of common sense and respect that must be mentioned: In the process of aggregating a piece of work it’s important to ensure you don’t extract all of the value of the original piece, or quote so much of it that there’s nothing left for a person to see at the original source.

If you extract all of the key passages and information and simply rewrite them, then there is no reason for a person to follow on to the original source. Or if they do, they may question your ethics.

This was a point made by Keller in his post about the potential dangers of aggregation:

Sending readers to savor the work of others at the sites where they publish — that’s one thing. Excerpting or paraphrasing at length, so the original sources doesn’t get the traffic or the revenue, that’s something else.

Aggregation works best when you pick out something from elsewhere and add value with new insights, additional reporting or salient background and context.

I. Send them away to come back

Some organizations fail with their aggregation when they try to avoid linking out to other sources. This obscures where the information came from, and opens them up to accusations not only of a lack of transparency, but also of plagiarism or the downstream activity that some have come to call “patchwriting.”

A post by Poynter’s Kelly McBride described patchwriting: “Rather than copying a statement word for word, the writer is rearranging phrases and changing tenses, but is relying too heavily on the vocabulary and syntax of the source material.”

By linking out to useful and credible sources, you provide a valuable service. You enable readers to learn more and dig deeper. Even if people click away from your site, the thinking goes, they are more likely to return because you provided them with what they were looking for.

One believer in send-them-away-and–they-will-come-back is Quartz, the global business website from Atlantic Media. When it launched in 2012, an article in Columbia Journalism Review criticized it for sending readers away from its site via so many outside links. Quartz senior editor Zach Seward responded in a comment on the article:

Our goals are just to cite our sources, acknowledge that there’s a whole wide world of great business reporting, and point our readers to material they should see … We’re thrilled if readers leave Quartz because we’ve pointed them to great material elsewhere because we know they’ll love us for it and come back for more.

Steve Buttry wrote that there are four key ways to provide value when aggregating content from elsewhere. Here are Buttry’s four ways to add value (reprinted with his permission):

Summarize. The best curation will provide a good overview of a story or issue, so the reader gets a basic understanding whether or not they click through to the various content you have collected.

Organize. You add value by grouping related content together: You gather news reports, blog posts, tweets, videos and other content on related issues, giving the compilation value beyond the sum of its parts. The organization within the curation adds further value: grouping the pro arguments and the con arguments or telling a story chronologically from multiple sources.

Original reporting. While curation is by its nature derived from the work of others, it doesn’t have to be just a compilation of external content. Our curators will at times fill gaps with their own reporting or by including in the curation summaries from and links to original reporting by Digital First newsrooms.

Context. Curators should place news in context, linking to background materials and to related content. Much of curators’ work focuses on the news, but you should always remember the value you can add from linking to your own archives and archives of other news organizations, Wikipedia and other reference sites. Topic pages are a helpful way to provide context, giving an overview of a running issue or a person frequently in the news, with links to earlier content.

II. Secure permissions

Too often journalists overlook the need to secure permission from people on social media to use their content.

User-generated content is not free for the taking just because it didn’t come from another news organization. Newsrooms also need to be aware of privacy and security issues related to user-generated content.

The best practice is still to reach out and secure permission to use others’ content before you embed or republish.

“We talk a lot about people’s expectation of privacy and the culture of different platforms,” said Mayer of the Missourian. “We would include a tweet in a Storify before Facebook posts, for example … We also talk about vulnerable populations, like quoting from a teenager’s tweet.”

One rule in Mayer’s newsroom is they always ask permission if they plan to take something from one social platform and move it to another. “If we’re going to make a Facebook photo album and we find an Instagram or Twitter pic [that we want to use], we will say, ‘Do you mind if we use this on Facebook?'” she said.

Permissions come down to three issues:

  1. Upholding the ethic of attributing the information and being transparent about where it came from.
  2. Dealing with members of the community in a respectful way that builds a better relationship between them and your organization.
  3. This is an emerging area of law and practice, and journalists need to be mindful of how they may expose themselves and their organization to future legal action. A study of newsroom practices related to user-generated content by the Tow Center for Digital Journalism at Columbia University warned that “it will not be long before an uploader takes a news organization to court for using content without permission or for failing to attribute due credit. The result of any such case would have wide-reaching implications for the news industry.”

One way to ensure proper attribution of content from Twitter, Facebook, Google Plus, Instagram, or YouTube is to embed it using the official code snippets provided by those networks. The terms of use for many of these platform specify that by users consent to allowing their content to be embedded.

But the best practice is still to reach out and secure permission to use others’ content before you embed or republish. This is especially important when the subject matter is sensitive, or there may be privacy or security concerns.

For example, when BuzzFeed collected tweets for a story about the #yesallwomen hashtag that sprang up in the wake of a shooting at UC Santa Barbara, the writer reached out personally to each person in order to secure permission to use their tweet. The story noted this for readers at the bottom:

Note: BuzzFeed has received permission from all those featured in this post to use their tweets.

Obstacle to Expect and Overcome: People may not know they need to ask for permission to use UGC. Pointing them to the summary of the Tow Center report is a good starting point. As with anything, this needs to be clearly and consistently communicated by leaders in the organization. To that end, create a team to develop your organization’s UGC policy for use and permissions, and hold training session to ensure everyone knows the correct procedures. It’s also useful to create a simple one-page checklist that reporters, producers and others can consult.

Chapter 5

Offer disclosures and statements of values

When an earthquake struck Los Angeles in March 2014, the Los Angeles Times was able to quickly publish an article that contained basic data about the quake. At the bottom of that story was an interesting disclosure:

This information comes from the USGS Earthquake Notification Service and this post was created by an algorithm written by the author.

The story byline belonged to Ken Schwencke, a journalist on the Times Data Desk. But as the above line explained, the story itself was automatically generated by an algorithm Schwencke wrote to grab the USGS earthquake data and quickly assemble it in a basic news story.

This was such an unusual scenario that the Times deemed it necessary to include a disclosure of the method within the story. It was the right call.

Looking across the landscape, we conclude there are typically five situations when a disclosure needs to be added to a piece of content:

  • Personal connections to sources, organizations or events mentioned in reporting.
  • Organizational links to sources, organizations or events mentioned in reporting.
  • Limitations of any data or material used. (Meaning: Here’s what we don’t know, or some mitigating factors.)
  • When unusual methods are used to gather information (such as in the above example).
  • When there is an unusual arrangement with sources (such as enabling them to remain anonymous or use a pseudonym).

I. Make disclosures within content

Disclosures are typically placed within the context of the content itself, or at the bottom of the page.

When disclosing an arrangement with a source, it’s important to insert it right after the first mention of the source, rather than make people go all the way through without understanding important details about that person. This is particularly true with anonymous or pseudonymous sources.

The NPR Ethics Handbook section on transparency requires its journalists to reveal as much as possible about an anonymous source without risking identification:

When a decision is made to use information that we have obtained from a source that must remain anonymous, we describe in as much detail as we can (without revealing so much that we effectively identify that person) how they know this information, their motivations (if any) and any other biographical details that will help a listener or reader evaluate the source’s credibility.

It’s also best practice to also offer an explanation of why someone is being granted anonymity or pseudonymity, and to explain as much about them as possible. For example a New York Times story about Libya included this disclosure paragraph:

More than half a dozen American diplomatic, military, law enforcement, intelligence and other administration officials were contacted for this article. All spoke on the condition of anonymity because of the diplomatic delicacy of the matter and the prospect of future military operations.

Philip Corbett, the Times’ standards editor, outlined guidelines for the description of anonymous sources in a memo. He warned that “Pat, formulaic expressions of why an anonymous source wants to be anonymous are probably worse than no explanation at all. They are uninformative and give readers the impression that our anonymity rules are on autopilot.”

Corbett offered these guidelines for disclosing information about why sources were granted anonymity:

Reporters and editors should in all cases discuss why the source wants anonymity, and consider seriously whether we can say something informative or interesting.

When warranted, a thoughtful sentence or paragraph, describing the pressures or concerns of the people involved in a situation, may give readers greater insight than a terse phrase. In other cases, a shorter explanation may be useful, but only if it conveys some real information.

For issues that relate to a reporter or outlet’s connection to someone or something mentioned in the story, the standard practice is to note it in a “disclosure.” For example, a Gigaom story about the parent company of WordPress includes this:

Disclosure: Automattic is backed by True Ventures, a venture capital firm that is an investor in the parent company of Gigaom.

Or this example from a post on Poynter by writer Andrew Beaujon:

Disclosure: My wife recently began working on a freelance basis for Quartz’s sister publication The Atlantic as a researcher.

Ideally, the disclosure should be placed as close to the relevant passage as possible. This connects the relevant story content to the disclosure, as with the source example outlined above.

II. Make disclosures about the editing process

A reporter’s name, email, phone number and even Twitter handle are now common information that accompanies their content. But some news organizations are now also disclosing the names and contact information of editors. Reuters, Bloomberg and smaller organizations such as the Missourian list the editors of an article at the foot of the story.

On Bloomberg stories, such as this one, the editors are listed at the bottom as part of a call to action for people to get in contact:


The Missourian discloses the supervising editor responsible for a given story. This is done in part, according to Mayer, because the paper’s staff is primarily made up of students from the University of Missouri’s journalism school, which means there is constant turnover.

“It’s important for us because the face of the Missourian is transient…” she said. “So we’ve seen some positive relationships with sources and communication with readers since we started pointing out who edited [the story], and linking to that person’s bio, just so there’s a more permanent source for contact.”

An instructive example of an editor bio page is this one for public safety and health editor Katherine Reed. Along with basic bio and contact information, it lists her most recent articles and comments on the site, and it includes a brief Q&A with her about the work she does. Mayer says this helps signal to people the kind of information Reed might be interested in, and it provides a human face.

“We definitely are hearing from more people, and a couple of our city editors have said they are getting more tips,” Mayer said.

III. Journalist disclosures

Beyond disclosures about specific reporting, more reporters and organizations are offering pages that are updated over time to disclose any personal ties, investments or relationships.

Re/Code, the technology news and reviews site run by Walt Mossberg and Kara Swisher, requires each reporter to maintain an ethics page, a practice they also did for years when they ran All Things D under the umbrella of the Wall Street Journal. This is where they list any relevant financial holdings, relationships with companies, or other items that require disclosure.

Swisher said she and Mossberg originally wanted to highlight reporter disclosures in a way that stood out from everyone else in the tech press.

“One of the things we decided early on was that it was important,” Swisher said.

The inspiration was the fact that her previous employer, The Wall Street Journal, required reporters to sign a standard ethics statement. But that statement was never shared with readers. And any specific disclosure that needed to be added to an article would be restricted by space constraints.

“So we felt, why not have a full public disclosure?” she said.

The Re/Code Ethics pages maintained on an ongoing basis, they are given good visibility by being linked to from every reporter’s byline:


Swisher’s ethics page, which she describes as “epic,” is written in plain language and explains, among other things, that her wife is a senior executive at Google. It describes how Swisher approaches this in her coverage of Silicon Valley.

Her ethics page begins, “Here is a statement of my ethics and coverage policies. It is more than most of you want to know, but, in the age of suspicion of the media, I am laying it all out.”

Swisher’s page details her approach to reporting, her financial holdings, her policy for accepting speaking fees, the structure of the parent company of which she is a co-founder, and much more.

Swisher also offers specific information about the financial arrangements she and her wife made to preserve her independence: “I have signed legal documents that disallow me from future rights to own [her wife’s Google shares] and, in the event of her death, her wealth will pass directly to our two children.”


As Swisher notes at the top of the ethics statement, it’s more exhaustive than would be necessary for many journalists, but the detail is there purposively, to make a point about her motivation to be transparent and therefore trustworthy.

“I have a much more complicated personal life [than the average reporter] and … I wanted to show what you can do,” she said. “I’m trying to numb you into submission about my personal life, but most of [the ethics statements] are short.”

Swisher says readers love the ethics pages.

“It’s so easy [to do] and I have to say readers really, really love it,” she said. “We’ve gotten dozen and dozens of letters over the years. They are shocked by it, and thrilled.

IV. Organizational disclosures

PunditFact, an offshoot of PolitiFact, offers a paragraph of disclosure about its sources of funding right on its homepage.


“When we launched PunditFact we made sure to include how it was being funded,” said PolitiFact editor Holan. She said the persistent disclosure is a result of her paying attention to a question that she was is asked when speaking publicly about PolitiFact.

“I do a lot of public speaking and people often ask who owns PolitiFact,” she says. So the homepage disclosure is a way to answer a question they know people often ask. It’s as much a service to people as it is an ethical effort.

Distilled from the work of others, here are four keys to doing organizational disclosures:

  • Whenever possible, place disclosures within the context of the content. Ensure people get the information about a source, organization etc. when they first encounter it.
  • Determine the questions that people ask about your organization’s operations/funding/ties? What questions are reporters always asked about how they invest/vote/volunteer etc.? These questions are your roadmap for knowing what to proactively disclose.
  • Error on the side of disclosure. It’s better to share the information early and often, rather than have something come out later that raises questions about your independence and credibility.
  • Create a dedicated place for reporters and your organization to make disclosures. This helps create a habit for staffers, which means people will think more about any disclosures they need to make, and therefore be less likely to forget to share something.

Obstacle to Expect and Overcome: Journalists saying that maintaining a disclosures page is too much work. Swisher says the ethics statements take very little time to write, and require infrequent updates. “We just hired someone from Mashable and he sent it to me yesterday and he wrote it in like four seconds,” she said. She also makes it clear to all editorial staffers that it’s not an optional activity. “All our reporters get it immediately and they don’t have a problem with it at all,” she said. “And if they do, too bad.

Chapter 6

How to correct website and social media errors

One elemental concept of transparency, which is a conceptual shift for some organizations, is that corrections are a good thing.

Errors are of course unfortunate and bad. We must make as much an effort as possible to prevent errors before they make their way into the world. The ease with which mistaken claims and false facts can pick up speed and spread requires that we be diligent in verifying facts before publication. News organizations should track the error rates of individual reporters, and offer training and, when necessary, discipline to help ensure overall quality.

However, some amount of errors will inevitably occur, especially in a fast paced digital environment. That’s why we correct our errors — and do so publicly and transparently.

Rather than destroying trust, corrections are a powerful tool to reinforce how accountable and transparent we are.

“If you’re willing to admit you’re wrong, people will trust you more,” said Mathew Ingram of Gigaom. “If I said to someone ‘You know, I’m never wrong’ they would think I was a psychopath or a liar, so they would trust me less. That’s versus if I said ‘I screw up all the time.’ They trust you more because you’re more human.”

That’s the paradox of trust: admitting our mistakes and failings make us more deserving of trust.

After a decade spent researching, writing about, and offering training and guidance to newsrooms about corrections, I’ve arrived at some corrections fundamentals that don’t change in the digital world:

  • Feel and write like a human. A correction is a very human thing, in that it admits a mistake. But too many corrections read as if written by a robot, rather than a human. If someone has been wronged, acknowledge it. And if the error is admittedly funny and the context allows for it, don’t be afraid to let the humor shine through.
  • Be clear about what was incorrect, and the correct information. Don’t make things worse by writing a foggy correction. There is a natural tendency to downplay the mistake, but this will only make things worse by leaving the reader confused or suspicious. Clarity is paramount when writing a correction.
  • Due prominence. This is a concept that exists in the British press regulation system. It holds that a correction must be given the same prominence as the original article. It’s admittedly an artifact of the print era, when the correction would appear on a separate day from the original. But it applies to the digital world because it’s still important to think about how you can give the correction a similar level of prominence and promotion as the original story.
  • Help the truth spread. A central goal of a correction is to communicate the correct information. This has always been the case. In a networked world, we must now also think about how we help the correction spread just as the incorrect information may have.

I. Create the right culture

In some quarters, a culture of shame traditionally surrounds corrections. Reporters are so stricken at the idea of making mistakes, and perhaps being punished for admitting them, that the tendency may be to avoid a public correction. They may also fear being disciplined for admitting their errors, which causes them to try to hide their mistakes.

This is the worst possible cultural scenario: it leads to an atmosphere of denial, of errors not being corrected, of audiences being frustrated — and it leaves journalists less likely to learn from mistakes, or from readers. The result, ironically, is more errors.

The New York Times’ Guidelines on Integrity send a clear message that journalists need to report their errors.

“The paper regrets every error, but it applauds the integrity of a writer who volunteers a correction of his or her own published story,” it reads.

Similarly, the San Francisco Chronicle’s corrections policy states that, “It will be considered unprofessional conduct and a breach of duty if employees are notified of possible errors but fail to respond. Correcting errors and clarifying ambiguous information is a virtue and an admirable practice.”

Based on extensive research of newsrooms that do corrections well, these are the four ingredients that make a positive corrections culture:

  1. Model behavior from the top. Newsroom leaders must emphasize that the worst (unintentional) errors are ones that go uncorrected. Make it clear that errors are teachable moments, and the focus is on learning why mistakes occur and finding ways to stop them from happening again. Reporters, producers and others must also do their part by committing themselves to working to prevent mistakes.
  2. Encourage people to report their mistakes, and to respond to requests. This must be reinforced consistently. One way to communicate its importance is to instill a policy that anyone caught hiding errors will be disciplined.
  3. Make it easy to do corrections. Create a workflow (as previously outlined) that makes it easy to have corrections quickly approved and added to content.
  4. Offer praise and examples. Celebrate people who act quickly to add corrections, and who interact with the community to respond to requests for correction. Show examples of well-executed corrections. Have regular meetings for teams or departments to discuss their mistakes in an open and constructive way, and talk about ways to prevent them from happening again.

II. Create a corrections policy

A corrections policy is the basic roadmap of why and how to do corrections. It need not be a long, wordy document. The more clear, concise and digestible, the better.

News organizations are also increasingly publishing this policy publicly. The Reuters Handbook of Journalism [PDF] is publicly available, and includes a detailed section about correcting and preventing errors

“Reuters is transparent about errors. We rectify them promptly and clearly, whether in a story, a caption, a graphic or a script,” it reads. “We do not disguise or bury corrections in subsequent leads or stories.”

Other detailed corrections policies worth looking at are from The Washington Post and BuzzFeed.

Public editors and others responsible for creating and maintaining a corrections policy cite five key elements to include:

  • Communicate the value of corrections. Why are corrections important to your organization? Why is it essential that journalists admit their errors and speak up when they think something is incorrect? When do you need to add a correction? Articulate your organization’s view of corrections.
  • Set your style; offer examples. The Washington Post, for example includes this basic template in its policy: “We generally revise the story to make it accurate AND append a correction to the file. Typically, online corrections read like this: ‘Correction: An earlier version of this story incorrectly reported …'”
  • Designate an internal lead. Who will be your corrections guru? This person is the keeper and updater of the policy, and should make the final call whenever there is a debate about whether to deliver a correction. In the end someone must be responsible for corrections, otherwise a policy will fall out of date, and no one will be there to offer reminders and positive reinforcement.
  • Design a workflow. How will you gather and process requests from the public and from sources? Who writes the correction, and who approves it? These and other elements of the correction workflow are essential to establish; otherwise there can be a backlog and confusion.
  • Determine if any tools are needed. As noted previously, more and more news organizations are offering forms and other ways for members of the public to submit requests for corrections. Think about what you may need to create to help the flow of corrections match your policy’s goals.

III. Write good corrections

Clarity is paramount in writing corrections. People must understand the initial error, and the correct information.

Attempts to minimize or obfuscate the mistake can render a correction confusing, or meaningless. It also compounds the damage. Along with the goal of being clear, a well-executed correction includes these elements:

  • It’s labeled a correction. The term correction has been used in the press for a long time. The public understands what it means. So if you are correcting a factual error, label the correction a correction. It’s not an “Update” or “Note,” etc.
  • It’s of the same quality as other content. Corrections must be written and edited with the same care as other content. This protects against having to issue the dreaded correction-to-a-correction.
  • The original error has been fixed. Along with adding the correction to the offending piece of content, it’s also important to fix the factual error in the content, if possible. This eliminates the cognitive dissonance of someone reading the incorrect information and then also seeing a correction that contradicts what they just consumed.
  • The mistake, and the correct information. As noted above, a good correction makes it clear what was incorrect, and what is the true information.

A few other elements to consider including:

  • An expression of regret. Adding a “we regret the error” or “apologies” to specific corrections where someone may have been offended or wronged can be a sincere way to communicate a sense of empathy and accountability.
  • When appropriate, reflect the tone of the story. A genuinely amusing error made in a lighthearted piece of content could open the door to a playful correction. For example, The Economist mistakenly reported that staffers at Bloomberg Businessweek are forbidden to enjoy casual drinks in the office. That resulted in this correction:
    “Correction: An earlier version of this article claimed that journalists at Bloomberg Businessweek could be disciplined for sipping a spritzer at work. This is not true. Sorry. We must have been drunk on the job.”But a word of caution: being funny can be hard, and risky. Ensure you only strike a lighthearted tone when it’s appropriate. Never compound the damage with a tone-deaf correction.
  • Indicate when the fix was made. News organizations will often include the date or even a time stamp with a correction, to show when the content was updated.
  • Credit the person who spotted the error. This is a great way to turn a correction into an opportunity for community engagement. As previously noted, corrections today sometimes include a link to a comment that pointed out the error or a thank you to a specific error spotter on Twitter or elsewhere. This rewards them for their participation, and communicates that your organization welcomes reports of errors.

IV. How to do social media corrections

Different social platforms have different best practices for corrections. Here’s a look at the right way to issue corrections on three major platforms.

Twitter corrections

Whether it was a tweet with an error in it, or a piece of content you tweeted that included an error, it’s important to flow corrections to Twitter.

As shown below, the ideal way to issue a Twitter correction is to send it as a reply to your original, mistaken tweet. In this example, the smaller tweet at the top is the original, which included an incorrect photo. (It showed Javier Bardem instead of Vladimir Putin.)

After realizing the mistake, Slate sent a second tweet as a reply to the first that included the correction:


Sending it as a reply does two key things:

  • It permanently links the correction tweet to the original one, thereby showing why the correction was issued. This also ensures that anyone looking at the mistaken tweet later will see the correction below it.
  • It ensures that all of Slate’s followers can still see the correction.

Note also that the correction text is very clear, and that it’s labeled (in all caps) as a correction. So the fundamentals of writing corrections still apply.

“In the past year, I’d say, I’ve started correcting mistakes within tweets as replies,” said Slate social media editor Jeremy Stahl in a 2014 interview. “Prior to that, I was running separate correction tweets after making errors, but then I realized that the original incorrect tweets were still out there and if people didn’t see both tweets they’d miss the correction. This seems to be an effective way of not only issuing a correction, but indicating the error on the mistaken copy.”

Retweets are another important aspect of doing Twitter corrections. If your incorrect tweet — or tweet pointing people to a piece of content with an error in it — was retweeted, then it’s good practice to make sure those users see the correction, and to encourage them to retweet it to their followers.

In cases where there may be too many retweeters to reach each of them, pick out the people who are most influential, in terms of the number of followers and overall profile. Then reach out. This may sound labor intensive, but it’s actually a very fast process (you can send the same reply to people), and you’ll receive a lot of positive responses.

As a general rule, you want to avoid having to delete a tweet. In cases where there may be legal issues, or when the erroneous information continues to be retweeted at a significantly faster pace than the correction, deleting may be necessary. But it’s to be avoided whenever possible — on any social platform.

“It’s rare for us to delete a post, and we would not delete just to prevent embarrassment,” said Mayer of the Missourian.

Another aspect for individual journalists to keep in mind is that the more personal and human you can be in offering a correction, the more sincere and effective it will be. Your social media profile is an extension of you. Think about offering a personal tweet to point people to a correction.

An example is this very genuine duo of correction tweets sent by Brian Stelter when he worked for the New York Times:


Stelter explained in an interview with Poynter that he wanted to be clear about the nature of his mistake.

“I had received a few tweets and emails from Bill O’Reilly’s viewers accusing me of lying, so I replied and explained that it was a mistake, not an intentional lie,” Stelter said at the time. “Then I decided to tweet it out widely.”

Facebook corrections

Facebook offers a feature that’s very useful for doing corrections: you can edit a post after it’s been published. So, as with a blog post or other type of online content, you can go in after the fact, fix your error and add a correction.

On top of that, however, it’s also best practice to add a comment on the post itself to notify people that you have edited and corrected the content. This element of disclosure is useful because people who previously commented on the post may be notified of your new comment, thereby drawing their attention to the correction.

Facebook also offers the ability to “hide” a post from your wall. This is a way to take an erroneous post off of the main page of your wall, without having to delete it.

It’s a useful feature when, for example, your original post featured an incorrect image that dominates the post itself. After you add the correction, you may want to hide the post, given the fact that the image itself was the main piece of content and it will be far more prominent than the correction.

Instagram corrections

As in the below example from NowThis News, you can add a correction to an Instagram post by adding it in a comment.

In this instance, the news organization responded to a person’s comment by adding a new comment, and noting that it was a correction. (It would have been better to call it a “Correction” rather than the abbreviated “Correx,” just for the sake of clarity.) NowThis also chose to give a “hat tip” the user who pointed out the mistake.


Obstacles to Expect and Overcome: Pushing corrections to all channels. Journalists feel awful when they make a mistake, and this leads to a natural tendency to want to stop thinking about the error as soon as possible. The idea of proactively reaching out to people on Twitter, or elsewhere, to alert them to a mistake is uncomfortable. However, by getting just one colleague to do this, you’ll be able to show two things: 1. It doesn’t take much time at all to reply to a few people. 2. The people you contact will be receptive and even grateful. This positive reinforcement can go a long way to getting journalists to embrace the discomfort of actively promoting corrections for their own mistakes.

Chapter 7

Worksheet: How to invite audience participation

We developed a worksheet with a series of planning questions to get you started with inviting your audience to participate in the reporting of a story.

Consider sitting down with a few creative collaborators from across your organization to tackle these questions together.

Download the strategy worksheet here.

Chapter 8

Study background and sources cited

People interviewed for this study either by phone or email:

  • Matt DeRienzo, Digital First Media
  • Kathy English, Toronto Star
  • Angie Holan, PolitiFact
  • Mathew Ingram, Gigaom
  • Joy Mayer, The Columbia Missourian
  • Martin Moore, Media Standards Trust
  • Kara Swisher, Re/Code
  • Maggie Walter, The Columbia Missourian
  • Rob Wijnberg, De Correspondent
  • Amanda Zamora, ProPublica

Reporting Errors

If you see a factual error or other mistake in this study, please contact us.