So you’ve written the research report: what else do you need to do to ensure people actually read it?

Remember the old days when you wrote a report, published it (perhaps with some kind of executive summary), did a couple of seminars and then declared victory and moved russian-dolls2on? Social media have changed that game almost beyond recognition: to maximize impact, any new report more closely resembles a set of Russian dolls, with multiple ‘products’ (hate that word) required to hit different audiences and get the message out. I’ve tried to list them, but am bound to have missed some – please fill in the gaps:

  • The report: 100 pages of well researched, clearly argued, and insightful thinking. Which (apart from other researchers) hardly anyone reads.
  • The Overview Chapter: 10-15 pages with all the juice from the report
  • The Executive Summary: A two pager for the time-poor
  • The landing page: better be good, or people won’t click through
  • The press release, with killer facts, notes to editors, offers of interviewees, embargo times and all that old media mularkey
  • The blog post: a way of alerting your particular epistemic community to the existence of your masterpiece
  • The tweets, although personally I hate those naff ‘suggested tweets’ you get from comms people.
  • The infographic: if you want to get retweets, these work much better than text. ODI currently the most infographic-tastic of development thinktanks
  • The 4 minute youtube piece, preferably not looking as knackered as Matt Andrews often does

Please add any layers I’ve missed.

If this seems like a huge amount of work, you can always just rely on word of mouth to get your paper out and about. But even when added together, all the comms packaging surely only amounts to a tiny proportion of the work that goes into the paper itself, so it would seem to me to be worth adding every bell and whistle you can. (On the other hand, has anyone got any evidence for the extent to which the packaging improves take up? Time for an RCT perhaps – randomly select some papers for the full treatment, and see what happens?)

Of course, this is only one small part of ‘research for impact’. What you actually say and how you say it, the topic you choose, the rigour of your research, the governance of the project (eg sneakily involving target individuals/institutions from early on) and timing (best to (re)publish after a scandal, crisis or general meltdown, when decision makers are looking for new ideas) are probably much more important in determining whether anyone takes any notice of all your hard work.

Update: lots of good comments below, thanks also to Robert Watt for linking to a great article on the ‘wonkcomms‘ site, which argues that thinking of research in terms of ‘artifacts’ is the start of the problem. Better questions are a) What outcomes do we want from our research? and b) How can we project our research as a thread for continual engagement? Smart.

Subscribe to our Newsletter

You can unsubscribe at any time by clicking the link in the footer of our emails. For information about our privacy practices, please see our .

We use MailChimp as our marketing platform. By subscribing, you acknowledge that your information will be transferred to MailChimp for processing. Learn more about MailChimp's privacy practices here.


23 Responses to “So you’ve written the research report: what else do you need to do to ensure people actually read it?”
  1. Robert Watt

    This is a good menu of options, perhaps even a standard minimum for research communication. But I’m missing three things:

    1. The first question that should always be asked: who can make use of the research? Mapping users (target audiences) will tell you so much about the formats, channels and timing for communication of research.

    2. The process by which communication will engage the users, start a conversation and build long term trust in the applicability of the research. What about a stakeholder dialogue? Or a webinar?

    3. Monitoring success/failure so that the communication activities can be tweaked to ‘what works’.


    Director of Communications
    Stockholm Environment Institute

    • Duncan Green

      Thanks Rob, particularly like the monitoring success/failure point, although presumably applies more to reports of lasting value (like your planetary boundaries work), rather than one off hits where you don’t get a second chance

      • Robert Watt

        Thanks Duncan – perhaps there are a different timescales for montioring. Here are a couple:

        – the almost real time data from online monitoring, which might tell you which format is generating most ‘engagement’ (i.e. what works in connecting with readers) and where to put a bit more effort

        – the longer term monitoring of readership *and* uptake

        I wonder whether your question might generate a richer menu of options if it avoided the framing of ‘reading’. It seems to me that this frame (i) generates, in most cases, a focus on written material (though I recognise you mention video – perhaps 2 minutes rather than 4? – and infographics) and (ii) runs the risk of falling into the information deficit model. These observations get back to my first point: who is it for anyway?

        • Duncan Green

          Excellent points Robert. I wonder who, if anyone has the feedback systems in place to do rapid adjustment of comms work to respond to info on uptake in the way you suggest- does SEI? If so, well impressed

  2. Sarah Jackson

    Great list! There’s a significant side benefit from creating comms products: making research more accessible and digestible for supporters. Some won’t read it even in tweet form, of course, but many will have a real interest in your recommendations and their context. And as we know more engaged supporters make better campaigners and peer fundraisers. Also your comms team will love you 😉

    • Robert Watt

      Excellent points. Translate where possible – SEI experience is wider and deeper engagement. But also consider the cultural context – is a translated executive summary appropriate/accessible? Perhaps users (readers) may prefer to engage with the research through, for example, local radio or community meetings. And bear in mind the question of who has the most credible voice to engage the user too.


      • Duncan Green

        Can’t believe I forgot translations – thanks! I would also add getting the right big names to endorse the report, as well as (if possible) use it/quote its findings once it’s out

  3. Peter McIntyre

    Nice piece – and after the press release actually make yourself available! Alongside this promotion, we also need to re-learn how to read. Forwarding links, re-tweeting and ‘liking’ < actually reading the report. If we don't read any more, how can we expect people to read our own stuff. The Internet – a wonderful resource for finding things we mean to read later.

  4. Patti Petesch

    Duncan, Very useful. Thanks! It just makes me wonder, though, about how to find the beef in the ever growing amounts of flab, and the great divide in capacities for influence and info. uptake — and whither all that. Now the same “highly lauded” gurus and their reports–without really much new in them–come at me in all manner of media. Is this progress? And sure hoping I’m not making myself completely irrelevant by not (yet, anyway) registering for Twitter in the hope that if it percolates elsewhere it might then possibly be worth a click. Cheers, Patti

    • Duncan Green

      Good point, Patti, there has to be a risk that needing all this paraphernalia to cut through the noise and reach people negates the democratising potential of the internet. Instead of gated papers, we now have access controlled by who has the money and staff to churn out top infographics and videos. Hmmmm.

    • Kirsty

      Really good point. I get a bit worried that all these research communication approaches result in a world where the degree to which policy/practice is informed by research is correlated with the personal charm (and ability to make infographics) of the researcher. In some cases, if the research happens to be good/important, this might be fine. But if we support a situation where policy makers get away with being swayed by whoever has the glossiest policy brief, there is a big danger that the next person who comes along can create great comms products but does rubbish research. I’m not against research comms – but think it’s important it is part of an ecosystem of interventions that also includes support for decision makers to find and appraise research – ideally synthesis papers which gives an overview of the body research – based on their relevance & rigour, rather than their visual appeal.

  5. Fraser Reilly-King

    To your great list I would add:
    Get blogged (get Duncan Green, or anyone with a bigger institutional profile than your organization, to write a blog about your report)!
    Beyond your press release and Executive summary, have a few clear messages – which you can convey to different kinds of audiences who can also talk up your report
    Run a (one-hour) webinar (over lunch) for people who don’t like reading!

  6. Joe Miller

    This is a good list of possible products, but I’d submit that it maybe comes embedded with a troubling assumption—namely, that the tried-and-true research report is the real product, and all the other things are just fancy add-ons. But I’m not sure that’s the right way to approach things.

    If it really is the case that no one (aside from other researchers) reads that big 100-page report, then maybe that’s an argument for not producing it in the first place. (Unless, of course, you’re writing for researchers in the first place, in which case, why not publish in an academic journal and be done with it.) But assuming you’re writing for people who aren’t researchers, why not simply start with publishing the results of your research in whatever form your audience actually is engaging with you?

    To put the point another way, maybe the burden of proof should be on showing that a 100-page report that took months to write but got only 43 downloads is actually a worthwhile thing to be producing.

    • Duncan Green

      Interesting, I’m probably not objective, but my view is that the razamatazz stuff needs to be grounded in something solid. A well researched report acts as a comfort blanket to activists and targets alike that you are not just making it up.

  7. Marion Davis

    I would argue that 100-page reports should be extremely rare (1/4 to 1/2 that length is much more viable), and that if you don’t have at least 100-150 downloads of the report itself, you’ve wasted your efforts. As was said above, if you were trying to reach anyone but fellow researchers, why not just write a journal article?
    I also think the idea of 3 layers of summary, but all in the report, is unhelpful. If the ES has to be 3-4 pages to capture the key points, so be it; the second layer may discourage people from going for the long version, since surely anything that matters is in the 10-15 pager?
    I also think stand-alone publications, such as policy briefs and factsheets, need to be on your list, because they’re more portable and easily printable. They’re your ambassadors in the real world.
    On media, the shiniest press release isn’t worth 1/2 as much as a personalized pitch to a trusted and well-known journalist. Also, blogs on institutional sites, while helpful, won’t get the views that a post placed (or cross-posted) externally can generate.
    But most important of all, as my colleague Rob said, the key is to know who you’re writing for, and then tailor to their needs. A good editor can help ensure it’s engaging, has the right tone, and is as succinct as possible.
    @marionsd (SEI climate comms)

  8. Ann K. Emery

    Another successful technique that colleagues and I have used in the past:

    *Mailing full reports to selected stakeholders*

    Example: Conducting a research report about grantees, and mailing 5-10 copies of the report to a dozen different grantmakers, along with a handwritten note. They’ll leave copies in their conference room for colleagues to see.

    With an overwhelming number of resources available online, sometimes the besy way to get their attention is simply to deliver hard copies to their desk.

    • Duncan Green

      Interesting, I’m on the receiving end of this and tend to agree. Although I will bin a large proportion of such reports without reading them, I am more likely to skim them on a bus than if they are sent round as links

  9. Heather Marquette

    Nothing beats emailing links directly to good contacts who think you’re usually sensible and having conversations. Having just read your ‘impact’ post, this is definitely an option for established folks though. Odds are you tend to read stuff on the bus from people who you know will probably give you something interesting. I know that they tend to go to the top of my enormous reading pile.

    • Alison Cohen

      Great comments! If your target audience makes use of Twitter, what about a Twitter chat? Have your author(s) focus specifically on answering “why?” as in: Why did we undertake this research? Why do we think it has value? Why does it matter? Offer opinion in bite size chunks easy to digest – and of course send interested tweeps to the full report.

  10. David Hobbs

    Thanks for the useful article.

    You asked about other layers. I think there are a couple other important layers, the *time* layer and the *context* layer. Actually, in many ways these are more like dimensions to the problem than layers.

    The layer of *time* is really important, but often overlooked. Although at the time of publication (and other activities close in time like the important ones you list), everyone is understandably happy and relieved to get the thing done and out the door. But there were activities leading up to that publication and those that will continue later. In other words, the research output is part of a thread of research. More transparency before publication can be interesting (for instance, releasing preliminary results, key initial graphs, or even requesting feedback on methodology), but from a communications perspective perhaps the most important aspect of the time layer is that if people are interested in the current publication then you want to engage them in the continuing research / investigation / exploration on that topic. Focusing too much on that one artifact (thanks to linking to my Wonkcomms article above) means that it’s easy to overlook this time dimension.

    Related is *context*, that is where does this research fit in with the way the site visitor thinks about the world? What current events or policy decisions does this relate to? Does this concern a particular geography? A particular approach? Especially nowadays when people often wind up deep inside a site due to a Google search, they may not even arrive at the information they are really interested in. Without context, they will leave forever. The site visitor may arrive at what the research team considered a jewel of an output that may be completely irrelevant to the visitor, but there may be something *close* to that research report that is very interesting. For instance, they may not care about that specific sliver of research but be very interested in the general topic — by linking to a rich page on the topic at large you may reach an audience that your would have lost forever otherwise.

    Anyway, I love thinking about this stuff. Let me know if you ever wanted to chat on these issues.

  11. Kyle Freund

    Good listing, we’ve been working to implement something similar in Fairtrade for each report that comes out. People’s attention too divided these days and you have to spill bread crumbs everywhere.

    Two other layers we work with:
    In our monitoring/impact report, we also break chapters out into slideshare presentations ( that people can download and share.

    And don’t forget to revisit info/content regularly. We try to pepper our feeds with a mix of new information, general interest, and revisits to older information when relevant.

  12. Joel Bassuk

    Duncan, good little self-reflective piece. And great comments above too. I would add:

    – Make the report share-able (infographics)
    – Make it find-able – specifically, optimize it for search engines. (Policy people are fond of using puns for report titles. These notoriously don’t include words describing actually what the report is about, and are usually not understood by readers from other language areas.
    – Cross-link the report across your digital platforms.
    – Don’t forget email – still a critical communications tool to reach audiences.
    – Listen – and respond! – to your audience.

    Though a couple years old, this presentation I did at DfID might be relevant: