HKS Professor Todd Rogers and Lecturer Lauren Brodsky say communicating policy proposals simply and well gives them the best chance to succeed—and not to be branded “too long; didn’t read.”

Featuring Todd Rogers and Lauren Brodsky
October 25, 2023
45 minutes and 12 seconds

Harvard Kennedy School Professor Todd Rogers and Lecturer in Public Policy Lauren Brodsky say that trying too hard to sound intelligent—even when communicating complex or nuanced ideas—isn't a smart strategy. Because today’s overburdened information consumers are as much skimmers as readers, Rogers and Brodsky teach people how to put readers first and use tools like simplification, formatting, and storytelling for maximum engagement. They say you can have the most brilliant, well-researched ideas in the policy world, but if you can’t communicate them, they’ll never reach the ultimate goal—making an impact. Rogers is the faculty chair of the Behavioral Insights Group at the Kennedy School and co-author (with Jessica Lasky-Fink) of “Writing for Busy Readers: The Science of Writing Better.” Brodsky is senior director of the HKS Communications Program and the co-author (with David Chrisinger) of “Because Data Can’t Speak for Itself,” a book about how to more effectively communicate the data that supports groundbreaking research and evidence-based policy proposals. They say millennials may be on to something when they mock your wordy social posts and text messages by replying “TLDR”—"too long; didn’t read”—because that’s how many busy readers feel about a lot of the writing that researchers, academics, and policy wonks do. 

Todd Rogers’ policy recommendations:

  • Do a final round of edits on whatever you write while asking yourself: “How do I make this easier for the reader?” 
  • Try the AI for Busy Readers tool at the Writing for Busy Readers website to learn how to make your emails and other writing more effective.

Lauren Brodsky’s policy recommendations:

  • Use the active voice whenever possible in your writing; it makes your sentences shorter and clearer. 
  • Look at every paragraph you write and ask yourself: “What does this paragraph do? Does another paragraph do the same thing?” 
  • Instead of using many data points, use just one or two supported by narrative and context. 

Episode Notes:

Todd Rogers is a professor of public policy at the Harvard Kennedy School of Government. He is a behavioral scientist who works to improve communication, increase student attendance, and strengthen democracy. At Harvard, he is the faculty director of the Behavioral Insights Group and faculty chair of the executive education program Behavioral Insights and Public Policy. He received a PhD jointly from Harvard's department of Psychology and Harvard Business School, and received a B.A. from Williams College, majoring in both Religion and Psychology. He is also co-founded two social enterprises: the Analyst Institute which focuses on improving voter communications, and EveryDay Labs, which partners with school districts to reduce student absenteeism. He is the co-author (along with Jessica Lasky-Fink) of the book “Writing for Busy Readers: The Science of Writing Better.” 

Lauren Brodsky is the senior director of the HKS Communications Program and a lecturer in public policy who teaches courses on policy writing and persuasive communications. She is also faculty chair of the executive education program “Persuasive Communication: Narrative, Evidence, Impact.” A co-author of the book “Because Data Can’t Speak for Itself” (with co-author David Chrisinger), she also publishes the website Policy Memo Resource at HKS. Brodsky lectures widely on policy communications and the use of evidence in writing for governmental agencies and nonprofit organizations. She is a former Theodore Sorensen Research Fellow at the John F. Kennedy Presidential Library and Museum, where she conducted archival research on public diplomacy programs during the Kennedy administration. She holds a BA from the University of Pennsylvania and an MALD and PhD from the Fletcher School at Tufts University.   

Ralph Ranalli of the HKS Office of  Communications and Public Affairs is the host, producer, and editor of HKS PolicyCast. A former journalist, public television producer, and entrepreneur, he holds an AB in Political Science from UCLA and an MS in Journalism from Columbia University. 

The co-producer of PolicyCast is Susan Hughes. Design and graphics support is provided by Lydia Rosenberg, Delane Meadows, and the OCPA Design Team. Social media promotion and support is provided by Natalie Montaner and the OCPA Digital Team.  

For more information please visit our webpage or contact us at

This episode is available on Apple Podcasts, Spotify, and wherever you get your podcasts.

Preroll: (Ralph Ranalli): PolicyCast explores evidence-based policy solutions to the big problems we’re facing in our society and our world. This podcast is a production of the Kennedy School of Government at Harvard University. 

Intro: (Todd Rogers): When we think about less is more, there are three kinds of less. There's omit needless words, as Strunk and White wrote in "The Elements of Style." Which is cheap, easy, costless—definitely omit needless words. Easy. The next is to omit kind of useful, but not necessary, details or ideas. And that actually requires judgment. And this is why I then want to pass to you as you go much more concrete on it. We always say at the level that we're talking about, it's trade-offs, the more you add, the less likely someone is to read and understand. But the optimal memo or writing of anything is not one word. So there's just trade-offs. But the second kind of less is fewer ideas. And the third is fewer requests. We have these experiments that others have done that I love where when you add additional requests, you decrease the person's likelihood of doing any of them or any one of them. 

Intro: (Lauren Brodsky): So a lot of what I am working on with writing is to contextualize data, meaning one data point with story and narrative and context around it is more vivid than a paragraph of five data points that don't really mean anything. So it's a less-is-more approach, which you're saying. And it's also that using your voice to tell the story of one data point is far more impactful, because it will be more sticky and memorable and understandable than just a string of data. And this is a tip that writers can learn. 

Intro: (Ralph Ranalli): Hi, Ralph Ranalli here. Welcome back to the Harvard Kennedy School PolicyCast. Well, it turns out that the snarky millennials were on to something when they mocked your wordy social posts and text messages by replying “TLDR” — too long; didn’t read. Harvard Kennedy School Professor Todd Rogers and lecturer Lauren Brodsky say that’s how many busy readers feel about a lot of the writing that researchers, academics, and policy wonks do. Rogers is the faculty chair of the Behavioral Insights Group at the Kennedy School and the author of “Writing for Busy Readers: The Science of Writing Better.” Brodsky is senior director of the HKS Communications Program and the author of “Because Data Can’t Speak for Itself,” a book about how to more effectively communicate the data that supports groundbreaking research and evidence-based policy proposals. Because today’s overburdened information consumers are as much skimmers as readers, Rogers and Brodsky say trying to sound smart—even when trying to communicate complex or nuanced ideas—can end up being a dumb strategy. Instead, they teach people how to use tools like simplification, formatting, and storytelling for maximum engagement. Because after all, you can have the most brilliant, well-researched ideas in the policy world, but you can’t communicate them, they’ll never have an impact. They’re here with me today.   

Ralph Ranalli: Todd, Lauren, welcome to PolicyCast. 

Lauren Brodsky: Thank you so much. 

Ralph Ranalli: Policy is our brand here, so we’re not just talking about policy, but actually being bullish on the importance of policy itself as a way to create positive change. And it feels really good to do that. But I've also been at the Kennedy School long enough to not only appreciate the value of good policy, but also to be pretty frustrated at all the good policy ideas that die on the vine for one reason or another. And I thought we'd start out by trying to get a handle on the role that effective communication plays in that. So my first question is, how big of a role does good or poor policy communication play in actually getting policy embraced, adopted and successfully implemented? Lauren, would you want to tackle that one first? 

Lauren Brodsky: Sure. Absolutely. So I like to think about policy communications as that last mile. You can have the most amazing ideas and if you can't communicate them well and get somebody else on board, they won't make it to that last mile. So you asked about implementation. I like to take it a couple steps back and what I like to say is if your goal is to be persuasive, persuasive policy communications would make a change.  

Well, step one is actually, and I think Todd works in the space too, is to get read. If nobody reads your work, you're certainly not going to persuade. And there's some tools and tips to help you be better read. So I think step one is read. Step two is to create care. And then maybe after read and care you're going to get an agreement for the action. Implementation is so far down that policy communication scope, but without a sense of implementation, what it would look like specifically, the audience can't see the vision. So you need to include it, but it's definitely not the lead. The lead is starting with effectively communicating to be read, to get care, to get them to want to do the action and then explain how. But that's so many steps down the road. 

Ralph Ranalli: Right. Todd? 

Todd Rogers: To stay at that level of abstraction, very briefly, a policy is made to enact some kind of change, often a behavioral change. Some human citizen to change some behavior. And very often in order for that to happen, a person needs to encounter some way of understanding what it's about and pay enough attention to understand it. So at the very end, maybe not even the last mile, probably the last 10 yards, or we'll go to meters, mile and yards, the last 10 yards would be how is it written?  

And so an example, the New York City Police Department issues almost 400,000 court summons a year. They are issued to people who are arrested and they're told they need to show up to court on a certain day, on a certain time. It turns out that they're really hard to make sense of, you look at them and there's no way a sensible person reading it would know that the only point of the document is you need to show up to a specific court, at a specific time, on a specific day. High stakes, so they're probably actually engaging with it. Half of people don't show up, they get bench warrants issued for this failure to appear, massively cascading consequences, massively, disproportionately hurts disadvantaged New Yorkers. And so they worked with some researchers at Ideas42 to experiment with different forms. The policy is the policy, whatever the arrest policy or the legal policy is, they need people to show up to court, otherwise really bad consequences for everybody. They redesigned it where they titled it Criminal Court Appearance Ticket. The top of it says, "You need to show up to this court, at this time, on this day." Incredibly effective at reducing failures to appear in court, led to something like 25,000 fewer bench warrants being issued just during the period of experiment.  

And you could call this not the policy—the policy is whatever leads to violations and these things being issued—this is the very end, how do we communicate it to the person? And it's a win for everybody to reduce failures to appear, so these bench warrants aren't issued so inequitably. So when we think about how important is it to communicate it, it's really in the end when we are encountering humans, how do we make it easy for them, given that they're skimming and busy and distracted and under a lot of stress, to understand? 

Ralph Ranalli:  Todd, your book is called "Writing for Busy Readers." You make some core recommendations on writing and communicating more effectively. Could you just very quickly walk us through them and give us an overview?  

Todd Rogers: Sure. I can give you the TLDR version, which is the too-long-didn't-read version. I learned that from my young students. 

Lauren Brodsky: My students told me that. 

Todd Rogers: Yeah. So not everyone will know it. TLDR is I guess an acronym that young people use to say, "You wrote too many words, I didn't read it, please tell me what it means." So the TLDR version is we should all edit everything we write in any practical communication through the lens of how do I make it easier for the reader? And when we make it easier for the reader to read, it's more effective, it's kinder and more respectful and it's more accessible and inclusive. The TLDR version is what we should write so it's easy for the reader and the readers reward us by reading it. 

Ralph Ranalli: So one of your recommendations, Todd, is to tell readers why they should care, in addition simplifying and editing down. And Lauren, you said it's important not to just identify the decision makers you want to communicate with, but also to connect with them. 

Lauren Brodsky: Yeah. 

Ralph Ranalli: So what are the best ways to establish that caring connection that you're both talking about, that emotional connection that gets people engaged enough to read through something? 

Lauren Brodsky: Yeah. So I like to tell writers, practitioners, et cetera, to think about policy writing as creating audience-centric documents. And that is the opposite of writer-centric documents. So in working with different organizations who want to write about their policies, explain different initiatives, whether it's through policy blogs or presentations, reports, et cetera, there is a desire to show what we learned. But that's a lot about what we the writers did. So there's a lot of sharing data and deep diving into the information.  

But when I say connect to the audience, what I'm really talking about is who do you intend to read this? So Todd was talking a lot about the public. I work in a space more that's focused on decision makers or maybe the media. So who is that audience? What do they care about? What do you want from them? And what do they need to know? And some of that information that you have, this really interesting study and lots of data points, that might not be what they need to know to gain some care and maybe action on the problem.  

So I feel that a lot of the work I'm doing with students and practitioners is to help them identify their audience and frankly research that audience. Which seems like a really simple step, but amazingly people don't do that step. So they know who they're writing to or who the minister is, but they don't take a step back and actually read what they wrote, what they care about, what the mission of their organization is to help funnel through all that data and say, "You know what? The one thing they need is this and I'm going to connect with them on that one piece of data." And so it's challenging sometimes for writers because they want to share certain things and that's about the writer, but an audience-centric document is about the reader.  

Ralph Ranalli: There's an old saying in writing, attributed to William Faulkner, which is, "Kill your darlings." 

Lauren Brodsky: Exactly. 

Ralph Ranalli: Which means that there are things that you may be emotionally attached to as a writer, but they are not necessary and even a disincentive to your reader and you have to make that self-sacrifice in order to edit those out. How do you convince people who are writing to do that? 

Todd Rogers: I am going to flip it a little bit … Let's think about it from the reader's perspective. Jessica Lasky-Fink and I wrote this book and have been doing research on this for either two years or 20 years depending on how we think about it. And the big picture takeaway is that everyone is skimming everything we write.  

When you start with that premise, then we need to write so we make it easy for skimmers. And one way to make it easy for skimmers is to make it really clear what aspect of this we think they might care most about. And we use the word care in the same exact way as you do Lauren. And an example of this might be that, when we write—and I think some of your students, Lauren, do this—we write to clarify our own thinking. That's a great function of writing. It's not the writing we should share.  

This next step is, okay, now how do I write for that audience, that reader? And we're really bad at it to begin with just as humans taking the perspective of other people, but we've got to try and what is it that they might care most about? Just because I do randomized experiments, I have to give a quick one: Rock the Vote did an experiment. They recruit young people to register to vote. They were asking their potential volunteers by email to volunteer at concerts to register people to vote. And they did an experiment on: What should the subject line be? In one condition it was, volunteer to register people to vote, because that was their purpose. The other condition was something like, attend free concerts as a volunteer. 

Unsurprisingly the second one had four times more people volunteer. And the idea is that the same content inside, there's just, well, why would a young potential volunteer care about this? Some of them, and bless them, are motivated because they want to increase voter registration. They all are, but it turns out they like going to Beyonce concerts too. 

Lauren Brodsky: Can I connect the dots a little bit on the audience-centric idea and skimmability? I also try to bring out highlights and bold sentences, clear main points that writers can have. And you can use this in a policy blog presentation, et cetera. Interestingly, I think writers really struggle with that skill and I think it connects back to this idea of audience-centric versus writer-centric.  

Most of the time people want to highlight the things they like, little data points in the argument or just parts of sentences that don't stand alone. And it's really fascinating to watch how writers struggle with skimmability. When I ask them to get into the mindset of their audience and I tell them what their audience thinks when they come to the document, they might want to know, why am I here? What's wrong? How bad is that? What do you want me to do about it? There tend to be three or four questions that a policymaker asks when they need to read something. They're trying to figure out if this policy suggestion is for them. And if you can highlight the one sentence answers to those questions, what's wrong? How bad is it? What are you suggesting I do? That's a skimmable document. But the way you do it is you know what the audience wants to know. So skimming is not about what you like, it's about what they need. And once I explain it that way, I see a change in policy communications. 

Todd Rogers: Actually I have recently come to think that people have this mistaken theory that more information and more detail reflects intelligence and thought, whereas I actually think the opposite. I mean obviously no information and no analysis is also not thought, but condensing is this incredibly hard task.  

I learned in Executive Education about three months ago, I taught a class and there was a person who was an assistant secretary in their country, not in the United States, and there was some crisis. The president of the country asked the secretary, the secretary asked this person, "Update me tomorrow." Or it was two days from now. "Update me two days from now." This is straight out of your class. This person wrote a 10-page briefing. The secretary is like, "There is zero chance the president is going to read this, get it to 200 words." And the idea was 200 words on top of it, so if the person wants more detail, it's there.  

And I was like, "That's an incredible story because one of our principles is less is more." And so in the class I said, "Okay, that's interesting. I'd never thought about it as being such a problem. Who else has been in a position where their boss or some person advised them, cut it by 90% because no one's going to read this?" Every single hand went up. I just had no idea how prevalent that problem was in this professional landscape. 

Lauren Brodsky: Absolutely. And actually if you highlighted those three sentences, what's wrong, how bad, what to do, and move them up to a text box at the top of the page—there are your 200 words.  

And the benefit of that as well is you've crystallized those thoughts, so if you come off the page and the minister turns to you and says, "What's this all about?" Rather than speaking all over the place, you actually know the three things to say. So creating skimmable documents actually also helps you communicate verbally. And that comes up a lot with students and practitioners where they'll write something big but they'll be in a meeting, they didn't read it and they turn to you and say the words, "What's this all about? What's happening here?" And you have to be able to communicate that very concisely. So yeah, I think it comes up a lot. 

Todd Rogers: When you teach, what is it? Is it called policy writing? 

Lauren Brodsky: Policy memo writing for decision makers. 

Todd Rogers: Great. No, that's exactly it. So when you teach that, one of the things that we have learned from the research and in writing the book and in talking with folks is skimmability. One of the best ways is to just add structure so that you have section one on this topic, section two on this. So if they want more on section one, they can dive in. If they don't, they can jump past it.  

And it is not linear essay writing. And I wonder how built in or baked in is that because my experience of talking to people about writing, that's a really hard transition, moving from beautiful, excellently written text to text that is easy for someone who's moving very fast to get the gist and dive in where they want and pop out where they want. 

Lauren Brodsky: Yeah. Exactly what you're recommending works really well for reports though. Because if you think each section does one thing and you put the main point of that section at the top maybe in bold. 

Todd Rogers: As a heading or something? 

Lauren Brodsky: As a header, it's not the costs of the problem because that's not skimmable, but it's the actual cost of the problem. You actually describe that. And then you might have four costs in that section, but if the reader doesn't want to learn about those four costs, they know what that section is and they pop right past it. So the way that we like to teach this is tell what the section is with a really explicit header and then have enough white space before the next section so that the reader can find the next section. And then it becomes a bit of a choose your own adventure, but that's a usable document for a reader. 

Todd Rogers: I love that. 

Ralph Ranalli: Do you mind if I flip this around for a second? 

Todd Rogers: Yeah. 

Lauren Brodsky: Yeah. 

Ralph Ranalli: I want to go to the other end of the spectrum. Because when you're talking about communicating about policy—policies can be complex, they can be nuanced, and they can be counterintuitive. So how do you keep from oversimplifying, from the pendulum swinging too far in the simplification direction so that you lose that nuance or that complexity that is important? Is there a time to tap the brakes while you are on the road to simplifying? 

Todd Rogers: Can I take this first? Just because I think I'm going to end up one level more abstract. When we think about less is more, there are three kinds of less. There's omit needless words, as Strunk and White wrote in "The Elements of Style." Which is cheap, easy, costless—definitely omit needless words. Easy.  

The next is to omit kind of useful, but not necessary, details or ideas. And that actually requires judgment. And this is why I then want to pass to you as you go much more concrete on it. We always say at the level that we're talking about, it's trade-offs, the more you add, the less likely someone is to read and understand. But the optimal memo or writing of anything is not one word. So there's just trade-offs. But the second kind of less is fewer ideas.  

And the third is fewer requests. We have these experiments that others have done that I love where when you add additional requests, you decrease the person's likelihood of doing any of them or any one of them. It's incredible. Which completely makes sense, think about how you navigate your to-do list. If you add more, it dilutes the likelihood that the important one is going to get done. And so just like what ideas do you omit, it's also what requests or actions because when you add them, you decrease the chance that the priority ones actually get done. But that doesn't get down to the concrete. So how do you coach people or counsel people? What gets cut? 

Lauren Brodsky: Well, this connects back to what you were saying in my mind about how writers want to sound smart and intellectual. And I work with a lot of economists, they want to share a lot of data and I completely understand that desire. And I think that the academic pathway to policymaking, maybe you're a political science major, you write a 20-page paper, you can keep reiterating the point, which won't work in a short document, isn't the best training to do this work. 

And I think another thing that comes out is what I would call highly abstract solutions, which will be something like: We need a whole of government approach to immigration. We need a whole-of-government approach to climate. And the policy memo or presentation lands there. And then unfortunately the decision maker's left thinking, "Okay, how do I do that?" And it's amazing how moving from really abstract language to one specific request really can make the difference. For example, comma and then just a first step, a next step, meet with so-and-so, do one thing. And so what I have to really help writers do is give us one specific. But I find that writers want to write in an abstract way, I think because they believe it sounds better. 

Todd Rogers: Yeah, I want to build on that a little bit in that they think that fancy language and complex sentences sound better. And in fact, when we were in school, we got rewarded for that. 

Lauren Brodsky: Yes, that's what I mean. Yeah. 

Todd Rogers: And one of the principles that we talk about is make reading easy. And that means shorter words, common words, shorter sentences, less complex sentences. And part of it is because it's cognitively taxing to read complexity. And it may make you feel like you did something smart, but the reader is more likely to just give up or skip over it or not understand it. That's one reason is because it deters engagement. But the other is this statistic that 50% of US adults read at a ninth grade level or lower. 

Lauren Brodsky: Right. 

Todd Rogers: 20% read a fifth grade reading level. And so when we write—and maybe the policy memos are comprehensible to the target decision maker—but when we write to the public, we are very often writing in ways that are completely inaccessible to the majority of possible people that would be encountering it. And then even for people who can understand it and are comfortable reading at that level, it's just cognitively taxing and more likely to deter them. So there's this underlying literacy research that in addition to its... One of the things that I've enjoyed in diving into this area over the last couple of years is there's a world where people have been arguing about this, and I assume you're engaged in these where you're like, "Make it simpler, make it easier for the reader." 

Lauren Brodsky: Yeah. 

Todd Rogers: And people will probably object that it's taste and style. 

Lauren Brodsky: Well, the best way to handle that is to workshop with other writers, to have someone who's never read that argument or knows what you're saying. And frankly, it's best when it's someone not in your field. So when you pair up someone who's interested in national security with someone who's interested in development, they're just doing different work. And the way that you get that information is to say, "Read my piece and tell me what's my point." Just simple workshop.  

And the challenge is the writer wants to say what the point is so badly. And so that's how I bring that out because when the reader responds with, "I think you're saying this or I think you're suggesting that." And then you hear, "Well, not exactly. It's more this." Whatever's spoken needs to come down on the page so people can explain why this matters, what they're suggesting in conversation more easily than they can when they open up a computer and start typing on Word or a Google Doc. So sometimes it's taking the morsels that were spoken and putting them down on the page. 

Sometimes people assume that you're going to take a holistic view of their writing and deduce that main point. It seems obvious, but it's not on the page. And then another thing that happens is they assume I'm writing to the Secretary of Defense, they know exactly what's wrong, so I don't need to say it. Well, that's a really important job. They deal with lots of problems. Your job as a writer is to clarify and coalesce around the point of this piece. It's got to be up top and central: what's wrong. So there's misunderstandings of what a good practice is and then there's a sense that it's clear and it's not. So that's what I try to work towards is creating an open-mindedness to: maybe I need to be more explicit, maybe how I say it is how I should write it. Which would be more simple than academic. 

Todd Rogers: Could you use ChatGPT to do that? Could you feed a paragraph in and ask ChatGPT, what do you think the point of this is in three words or less? Or something like that. 

Ralph Ranalli: Thanks for bringing that up. My next question actually, was about AI and about what you both think about its applications and about training AI to do this. Because AI, once it’s trained, is not going to forget to do this or fall back into bad habits like a human writer might. What do you two think of AI’s potential role here?  

Lauren Brodsky: This will be interesting. I'm curious if we line up in the same place. I would argue that I'm learning as I go what I think about AI, but let me give you an example of how I decided to approach it this fall. 

Because I'm teaching writing, there's a lot of writing in my classes. I work with practitioners on their writing, assuming they're writing. So I assign a policy memo on the first day of class that was written by a student a couple of years ago on why not to be tardy to class. And I didn't assign this memo, this was their memo. So it was this great thing that I have where I can nudge my students by bringing in the voice of a former student. And this memo is great because it sets the scene: you start class, you're trying to settle in, people are coming in, they're drinking a coffee—it's disruptive. And then they go through the costs, very specifically the costs of tardiness. And then as options they give, "Professor X does this and on the other side, professor Y does that. 

And I recommend a compromised approach. "So it's very steeped in the organization, the culture of the place, the details, the look and taste and smell of the place. So out of curiosity, on the first day of class, I asked ChatGPT to write a memo on tardiness. And when it spit it out I thought, "Oh my goodness, this looks amazing." It really did. It had the look and feel, it had it all there. When I dug in, I noticed two things, which you could say you could train it, but we as writers need to know how to do it. There was no personalization, there was no connection to audience because they don't know audience and really good policy writing knows the audience and serves what they need. And there was no organization connection, no narrative, no details. Plus there was some cause and effect. What's wrong and what caused that thing that's wrong? There was some confusion there. So there was some problem with the argument.  

I think you could retrain it and fix it or cut that line out, but I think it would have to be a starting place, you'd have to come back and bring in all that audience knowledge that you create as a person to make that writing better. So that's how I approached it in my class and I told students to cite it and then they didn't use it because they didn't want to cite it, which was interesting. So that's where I am right now. 

Ralph Ranalli: Todd, where are you on generative AI? 

Todd Rogers: I have come to think that the human sets the goals, knows the background information that's not in text and knows the interpersonal norms. And that is unambiguously for now what we are distinctly and uniquely great at. We've tuned GPT-4 to the principles of "Writing for Busy Readers." And on our website there's a Chrome extension and also an actual play area where you can play... We tuned it on the principles and then pre-post lots of examples of emails. It doesn't have to be emails, but this one is. We've done experiments with text messages, with forms, with applications. 

Ralph Ranalli: What's the website? 

Todd Rogers: It's And it's really cool, it's a 24/7 writing coach on how to write to make it skimmable so it's easy for readers. It's really good. So even though I wrote literally with Jessica Lasky-Fink, we wrote a book on "Writing for Busy Readers," I'll write a crappy first draft, I'll put it in and it'll just suggest the structure. It's not the final draft. Then I'll be like, "Oh, I hadn't thought of doing that."  

And I like it as a teaching tool, I use it in class when I teach in Executive Education, when I teach the master’s students and when I give trainings with other agencies. And people like it as a way of ... It's really hard to talk in the abstract about editing. And so it instantly is like, "The six principles would apply in this way." And so this paragraph describes things, you probably could put them in bullet points. And if the bullet points are more than 10 words, I say it should have a three-word summary of the bullets so that otherwise they have to read it. And so it's whatever. So boom, I'm not interested in that. I can skip it.  

And it also moves unessential details below the sign-off. This is just for email, but you could train it for lots of other things. And I love it as a teaching tool and I get feedback now, it's gone out in the world and people are sending me examples all the time of them using it. And the big picture is how do we write so it's easier for readers? And I think of that as one of the really powerful ways of helping to disseminate this approach. 

Ralph Ranalli: So getting back to that notion of emotional engagement for a second, I did want to talk about this notion of storytelling. And I think, Lauren, in your book, the number one recommendation is, use reporting to convey information and use stories to convey significance. I don't know if you know the great book, “The Storytelling Animal” by Jonathan Gottschall. He cites dreams as an example of how how story is part of our DNA. And he says, "Humans are so wired for a story that even when we're asleep, our brains stay up all night telling us stories." 

Lauren Brodsky: Yeah, that's great. 

Ralph Ranalli: So I wanted to ask the both of you what you think the importance of storytelling and narrative is, both in making something engageable and making it sticky in terms of retaining the information that you're conveying? 

Lauren Brodsky: Yeah. Absolutely. Yeah, so in our book, we're looking at how people use numbers to tell a policy story. And there's such an emphasis on evidence-based policy design. And so many organizations that I work with have a lot of studies to share and they struggle to figure out how to do that well. They're attempted just to somewhat, what I call, serve it up. "According to a recent study, we learned X, Y, Z."  

And what we need above that to get interested in that study, that data, that information is the writer's narrative voice. What's happening? What are we noticing? What's wrong? There's a rise in a particular disease, fewer students are going to school, more absenteeism, something like that. What's the point of the piece? And that doesn't have to be story.  

So there's two approaches. You can have your narrative voice and just tell us what's this all about and then bring in the data. So it's always narrative then data. But story's more sticky, to your point, and the story being about people and situations. So if you tell the story of one student who didn't go to school and you humanize that as to why, what's going on for that student? And then you maybe zoom out to a bigger question about absenteeism or something like that. Hooking onto that one individual in many situations will have readers remember the problem more. They might not remember in two days the numbers, but they will certainly remember the problem. So in that way, it definitely is more sticky.  

The other thing that's really positive about that is in numbers, X, Y, Z percentage of students aren't coming to school, we lose humans and policy is about making change that impacts people. So one of the main arguments in our book is to bring the people back because we started by talking about impactful policy, impact for whom? Not the policymaker specifically, but the people they aim to help. So it's more sticky to bring in those people. 

Now, one thing I hear from practitioners is, "I don't have access to the people." Well, that's when your narrative voice of just describing what you're seeing in the world or what's at risk, even if you can't communicate about one individual, can be really impactful. And the way in which you do that is first tell us that story or use your narrative voice and then bring in the data when the data will support that rather than just be served up and standalone and hope that the reader makes some meaning out of it. Which takes work for them and they don't want to do that. So it's really your job as a communicator to do that. 

Ralph Ranalli: And it seems like we have more and more data all the time. We have this fire hose of data, but it doesn't necessarily mean we are getting more or better information, because it’s too much to process, because it's being misinterpreted, and because it's being outright used for misinformation purposes. More data has not ended up being more helpful. 

Lauren Brodsky: Yeah. Absolutely. I think that stems also from a sense of writer-centric behavior. Writers will say something like, "The program costs $1 million, $1,000." And the reader doesn't know if that's expensive or cheap because they don't know what programs cost because they're not in your head.  

So a lot of what I am working on with writing is to contextualize data, meaning one data point with story and narrative and context around it is more vivid than a paragraph of five data points that don't really mean anything. So it's a less-is-more approach, which you're saying. And it's also that using your voice to tell the story of one data point is far more impactful because it will be more sticky and memorable and understandable than just a string of data. And this is a tip that writers can learn. Look at a paragraph, when you see more than 1, 2, 3, is a lot, of numbers. You need to pick one, make your main point and contextualize that one. So you can skim a document and see how you're doing on that and that's a real easy change to make. 

Ralph Ranalli: Todd, how do you integrate narrative into your simplification? And it would seem that it might be difficult when you're talking about formatting and simplifying to also integrate that with narrative. Is it? 

Todd Rogers: First, stories motivate people, engage people, get remembered, all of this. You just need to know what the point of your writing is. It all starts with, what's the point? And then, okay, well what's the context and what's the reader going to be like? And if the story can support your point, then include it. Narrative is really powerful.  

For us, we start with a place like assume they're going to give you a few seconds and as long as whatever you're including is structured in a way that it's not going to distract, derail, send them down a detour and it supports your main point and it's easy to get past when they're ready to skip over it or whatever, it seems really powerful.  

I want to back up a little bit. In one of these ideas that I think is the most radical of our approach, and I predict just because I think it's true that you're going to think it's true. 

Lauren Brodsky: Okay. 

Todd Rogers: But one of the central tenets of design thinking or user-centered design is that if a user or somebody interacts with your object and doesn't know how it works and gives up, it's always the designer's fault, it's never the person's fault.  

We think that we should be writing that way. That if you send something to someone, they look at it and move on. Even if they only spent one second on it and they don't know what it was about, we're often like, "Well, it was all in there. It was complete. It's on you." It's on the reader. The orientation, as long as it's well-written, it's complete, it's clear if you read it linearly.  

But what if we flip it and we're like, if the person looks at it and moves on, it's on us, the writer. Because the reality is the reader is the reader and we can't pretend that the reader is not skimming, the reader is not busy because they're skimming and they're busy. They're jumping all over and then they're moving on to the next thing. And the next thing is so easy for them to get to, click to the next thing or just hit delete. And the next thing just comes up. And so the radical take is if you send something complete, clear, coherent and beautifully written, the person looks at it and moves on, then it is always the writer's fault, it's never the reader's fault for not having read it. 

Lauren Brodsky: Yeah. I actually completely agree. And I end up bumping up against this problem in a really specific way, which will be somebody will have a one-page policy memo to write, but the topic's so important, it doesn't fit. No. That's on you to make very good decisions on what goes in, out. I only have five minutes to talk to a principal about a particular policy and there's no way I can make my point.  

Well, you can do amazing things in one sentence. And we talked earlier about how to bring three sentences up to the top of a policy document in a way that tells the whole story of the piece. So it's incredibly possible, it just takes training. And so I think the way that this happens for me is I watch writers think that they can't do it in the amount of space because the policy is so complex. And a lot of policies really are complex. I certainly don't want to minimize that. But sometimes you're going to get 30 seconds, you're going to only get the elevator pitch, so what is the gist? And that's a communicator's job.  

So it's the same idea. It's on us to figure out, if I only had a minute of time, what would I say? What's the most important thing that they know? And I think that's a skill that takes some training and you have to be willing to come to this place where you see the value in that because it's easier to say, "Well, I put everything there. It's a 40-page policy report and nobody read it." And so I'm seeing organizations use policy blogs as a way to take a 40-page policy report and try to put the blog out into the world, maybe for a hybrid audience of the public and other policymakers, hopefully to get the word out.  

One of the pieces of advice I give is that the main point of the whole report should really be above the fold of the first page of that blog. You shouldn't even have to scroll to figure out what's the main point of this policy proposal. And oh, well, I have to set the context and I can't quite get to the main point in the first three paragraphs. No, really that's on the writer to find a way to bring it up because they might not scroll down. So I'm in agreement with you because I just experienced how important this is. 

Todd Rogers: I've worked with a few federal agencies where they have reports and they have some version of what we might call an executive summary. And they've become very normalized. When it first started, the writer could do whatever they want. And now it's structure and the norms have become it's now three or four pages of executive summary. 

Lauren Brodsky: That's too long. 

Todd Rogers: Exactly. And so some of these agencies are writing summaries on top of the executive summary, on top of the report. And I'm going to ask you to join me on this appeal to the listener. No executive summary should be more than one page. 

Lauren Brodsky: I agree. Actually that's how we're teaching it now at the Kennedy School. 

Ralph Ranalli: This is perfect timing, and I'm glad you were making a recommendation because his is actually the time in PolicyCast where we want your direct and specific recommendations for our listeners. 

Lauren Brodsky: I would take that a step further. I would say the executive summary should be one page, well formatted. You should be able to figure out what is the policy problem, what is a main cost, and what should we do highlighted on page one.  

I think what happens is when folks write an executive summary, they think it's the only page they won't skim. They 100% will skim, and they'll skim to see if they want to read it. And if they don't know what it's about, they're not going to read it. So I would take it even a step further with that executive summary. 

Ralph Ranalli: So the people I’d like you to address now are the members of our audience who are saying, "OK, Todd and Lauren, I'm in. I want to get to this place where I'm a better communicator, where I'm writing more succinctly and getting my ideas across more efficiently and to a wider audience. How do I start? What are the first couple steps I should take?" Todd, do you want to go first? 

Todd Rogers: Sure. I think it makes sense for me to go first also because I’m at a slightly more general level. There should be a final round of edits of everything we write, from a text message, to an email organizing who's coming for Thanksgiving, to a summary report, to an intelligence assessment in the CIA, to a form that we're asking the public to fill out. A final round of edit where it's all there, it's all complete. How do I make this easier for the reader?  

That's the recommendation. Add a round of edit to everything we write. It could be the last round, I think it should be integrated into everything, but how do I make it easier for the reader? Our book is how to do that, what the steps should be to ask, what are the checklist items for what would make this easier for the reader? And then from there, it's context. And only you know there aren't concrete things and so if you're writing policy briefs, very clear context and you're an expert in that and you probably have very concrete advice for that. 

Lauren Brodsky: Yeah. 

Todd Rogers: But one level up, more general, just ask, how do I make it easier? Because it's just kinder, it's more effective. 

Lauren Brodsky: Yeah. Well, what I really like about what Todd is saying is it's a mindset shift and a lot of change that you have to make in becoming a better communicator is changing your mindset. And what we were talking about earlier, when you think about the audience, that's really just a mindset shift. It's not difficult to do, you just have to know to do it.  

Really, on the tactical level, I mean I have a lot. But I'll try to recommend just a few. I mean, I think with a lot of these tips that I would give, I would say often not always. So often it's really helpful if you write in the active voice. Active voice makes your sentences shorter, clearer, et cetera. It's not always possible. So that would be one. Often it's really helpful if you only use one or two data points. Sometimes that's not possible to make your point, but that's the goal. Often it's helpful to bring in people, but I think you were alluding to this earlier, in certain situations like a memo in the DOD, talking about an individual is not the norm. So you need to know the norm of your organization and then that tool doesn't work. I'm trying to think of more. I'm going to stop there. I have so many. 

Ralph Ranalli: Well, those are great. 

Lauren Brodsky: Oh, I have one more. 

Todd Rogers: Okay. 

Lauren Brodsky: I think a good way to edit is to go through your piece and look at every paragraph and ask yourself, what does this paragraph do? And if you start to notice that another paragraph does the same thing, you got repetitive. So if this is about the cost and then this is about the problem and then we've got the cost again, now it's getting repetitive. So a trick is to do what's called a reverse outline, where you look at each paragraph and what is the point of that paragraph? It helps people edit out repetition. So that's something you can easily learn to do. Okay, I'll stop there now. 

Ralph Ranalli: Well, as a professional communicator, I certainly have my homework now, and I’m really looking forward to diving in. And I hope that all the PolicyCast listeners out there really embrace this and see what it can do for them. Thank you both. This was a really enjoyable conversation. 

Lauren Brodsky: Thank you. It was great to be with you. 

Todd Rogers: It was really fun. Yeah, thank you. 

Outro (Ralph Ranalli): Thanks for listening. Please join us for our next episode, when HKS Senior Lecturer Jorrit de Jong and Harvard Business School Professor Amy Edmundson discuss their research on how complex urban challenges like climate readiness, equitable economic opportunity, and affordable housing are best solved through collaboration between multiple stakeholders including city government, nonprofits, business groups and academics.  

If you enjoy PolicyCast, we invite you to check out some of the other podcasts being created by our stellar centers and program. This week, we’re highlighting the Justice Matters podcast from the Carr Center for Human Rights Policy. In a recent episode, Carr Center faculty director Professor Mathias Risse talked to human rights pioneer Kenneth Roth about his long career as the director of Human Rights Watch and as a steadfast voice for justice and rights around the globe. You can find Justice Matters on Apple Podcasts, Spotify, or your favorite podcasting app, or by visiting the Carr Center’s website at 

And please subscribe to PolicyCast on your favorite podcasting app so you don’t miss any our great upcoming episodes. If you have a comment or a suggestion for the team here at PolicyCast, please drop us an email at—we’d love to hear from you. And until next time, remember to speak bravely, and listen generously. 

More from PolicyCast