What makes a perfect short field trip (and a top village power analysis)?

February 10, 2014

How to build accountability in fragile states? Some lessons (and 2 new jobs) from an innovative governance programme.

February 10, 2014

How to Write a really good Executive Summary? Here are some thoughts, but I need your comments.

February 10, 2014
empty image
empty image

Inspired by your great crowdsourcing on where to do a part time Masters (can someone collect the comments into a single document dead_tired_writerplease?), does anyone fancy helping me draft a short guideline on how to write decent executive summaries? Here’s the draft – over to you for improvements, suggestions for good/bad examples of the art etc. This is to add to Martin Walsh‘s burgeoning series of research guidelines on everything from semi-structured interviews to ethics to how to do a literature review- worth a look.

Picture the scene. An exhausted author grinds to the end of a gruelling process of research, writing and sign off. A 40 page draft lies before them – by now they can hardly bare to even look at it, let alone read it yet again. But the front page has only two words on it ‘Executive Summary’. OMG, better get on with it.

Disaster, because in terms of readership and impact, you should spend a lot of time and care on the Exec Sum, not a begrudging half hour at the end of months of slog.

Why? Because many more people will read the Exec Sum than ever read the whole paper. (Even more people will read the title and then decide not to bother with the report, but that’s another issue). The Exec Sum is the bridge between the often-technical arguments and findings of the main report, and the target audience, whether media, decision makers or aid professionals. Yet all too often, it is dashed off as an afterthought, badly written and missing some of the best nuggets in the main report. What a waste! Here are some ideas on how to get it right, or at least not as wrong.

Who? It may be better (if you can) to ask someone with a media or comms background to draft the Exec Sum. (The author should of course also check and revise the draft.) A fresh pair of eyes will be closer to the mindset of the eventual reader; they will spot what’s interesting and new; they are more likely to avoid jargon and specialist vocabulary; they do not hate the report like the author probably does by now. If you’re a one-person outfit without the luxury of a second pair of eyes, at least get a good night’s sleep and maybe spend a day working on something else to clear your head before starting in on the Exec Sum.

What? Try and keep it to two pages max – one sheet of double sided paper, (and don’t cheat by shrinking the font – if anything it should be larger than for the main report). Above two pages, you will lose readers fast. If you can get it down to a single page, or even a single paragraph (for example if there is one stand-out finding from the report), so much the better.

And now for the Exec Sum...

And now for the Exec Sum…

The first paragraph is the Exec Sum of the Exec Sum – crucial to getting and keeping the readers’ attention (imagine a tired journalist skimming a dozen incoming reports, or someone in front of a computer scanning their RSS feed – what makes them keep reading,  rather than moving onto the others?). It should say who your organization is, why it has written this report, and what is new or interesting in it.

You need a clear statement of the problem the report is addressing and the key findings of the research.

Identify and import any killer facts or particularly striking graphics from the main report.

Add a few of the main recommendations, but not a two page shopping list

How? If you are a new pair of eyes, first interview the author to get their views on what is new or interesting. They will know both the report and the other literature on the topic far better than anyone else in the organization.

Then start with the original full report (not the ToRs or the original Concept Note – they should have been superseded by the report.) Read it carefully, particularly the first and last sentences of each paragraph. Highlight key words or ideas, (even if you’re the original author, this may be a worthwhile exercise) as well as killer facts and good graphics.

Using these prompts, construct the strongest possible narrative. If you have trouble doing so, that may be because the report itself is not clear, either in arguments or structure – writing the Exec Sum can often lead to tweaking the main report, so you may need to go back to the author on this.

Re-read it, test it, get others to read it and comment on what does/doesn’t make sense. If you’re a fresh pair of eyes, ask the author to factcheck it and point out any key omissions. Take your time – this is the most critical part of the report.

Things to Avoid

Do not introduce the structure of the main report: ‘in section one, we discuss the context of X’ – the Exec Sum needs to work as a executive summarystandalone document.

Do not introduce new ideas, references, or arguments.

Don’t oversell – sprinkling hype (‘groundbreaking, breathtaking, extraordinary, outrageous’) over the document will put people off. Substance, not superlatives, is what will convince them to keep reading.

Keep jargon and acronyms to an absolute minimum, and explain their meaning the first time you use them. In particular minimise the use of mind-numbing development speak (ongoing participatory processes etc). Keep in mind George Orwell’s guide to how to write.

Examples of Good/Bad Exec Sums

Please send me your favourites in both categories, (including Oxfam papers of course) along with your reasons for proposing them. Plus a special mention for the person sending the longest/shortest Exec Sum.

And to get you started, here’s a few from the business world


  1. “It should say who your organization is, why it has written this report, and what is new or interesting in it.

    You need a clear statement of the problem the report is addressing and the key findings of the research.”


    I’m afraid – you better put the horses first: Put the conclusions first, then method/credentials.

    What is more effective in stopping a toddler walking into an oncoming car: “STOP – a car is coming”, or “Honey, a car is coming, STOP”? In the second instance, you’ve lost a precious half a second. The toddler (and the Executive – same ilk) trusts the mother implicitly, just like you trust your subconscious mind to pull your hand away from the flame before it visually notifies you of its presence. So go for action.

    In a paper one creates a linear narrative from problem to method/credentials to conclusion. It is a common error to retain this linear structure in Summary – just making it shorter. To catch the interest you state the conclusions first, then justify them (as provide credentials).


  2. If the report has well-defined sections, it might be a good discipline to draft some of the Executive Summary as you complete each section. You can always come back and amend it later, taking the context of the whole report into account. I have not managed this approach myself yet, but I think it could be a good idea.

  3. Spare a thought for your readers. They may be paid to read your report but try not to make it too costly.

    A handy way to help with the writing an executive summary is to imagine an elderly aunt or young (13 year old) nephew asking a question “That thing you said you were writing, what was it about?” You’ve got their attention now come up with an answer that inspires them and leaves them with something memorable. Not an easy task but well worth trying.

    Are you clear, concise and vivid in expressing your argument? Have you a clear problem, cause, solution and consequence of a future with or without your solution?

    A good start is Orwell’s Politics and English Language and heed his advice, especially:

    What am I trying to say?
    What words will express it?
    What image or idiom will make it clearer?
    Is this image fresh enough to have an effect?
    Could I put it more shortly?
    Have I said anything that is avoidably ugly?


    Though not everyone is convinced by Orwell


  4. Aldo is right. Most newspaper writing provides a good guide. The first paragraph – the first 2-3 sentences – should say immediately what is the key finding, to grab the attention. Then say why the research was done and how etc. An effective executive summary should read like a good press release.

  5. Great blog.

    Yes, every writer’s biggest challenge: boiling it all down to a punchy summary — to be performed when energy is at lowest levels.

    TB is quite right: draft a section of the summary at end of every substantive section. This is relatively painless if you have followed advice to conclude each section with a short and succinct, ‘this is what I want you to take away from the chapter you’ve just read’ summary.

    Length of summary?

    Grumpy old men can remember when the pretentious ‘executive’ was added to the simple word ‘summary’. In those days — yes, back in the 1970s, if you must ask — we drafted two summaries. A 6-pager that allowed the reader to get most of the report in a 15 minute read; plus a short summary — supposedly for executives — that would be 1 or 2 pages, to be read in 3 minutes.
    Of course, with time the good habit has been abandoned (when did you last see a report with two summaries?), but the pretentious suffix has persisted …

    … I’ll get me coat …

  6. For non-hard copy documents, I think a good executive summary should have good hyperlinks to either supporting evidence or the longer discussion behind the point. We see this all the time with links on the web, but not so much for links within documents – and I think that’s a shame and a missed opportunity.

    Also, big fan of summary tables and numbered or bulleted lists…

  7. One need only search “plain language guide” to find some great resources on plain language writing. How about starting with the paper itself, before moving on to the exec summary? We have moved on a bit since Orwell, brilliant though he was.

    See – http://www.plainlanguagenetwork.org/resources/





  8. The best summary I ever read was short and medium summary on one fell swoop. There was a single sentence to sumarise the essentials. It was a long sentence, with many clauses. That sentence was then used, clause by clause, as the bold headings for very short explanations of each bit (3-4 lines for each, as I recall). So your could read just the headings, and get the full sentence. or you could read the whole thing and get each nugget briefly explored as you went along. It worked efficiently and elegantly. But did I keep a note of what the document was? now that’s a regret I carry still, as I’ve described it many times to people struggling to write a summary!

  9. Lot’s of good advice here. I have nothing further to add about the writing of executive summaries. Rather, I suggest those involved think of this process as merely one step in getting the ideas from the report into people’s heads. That being the case, while the report author may well need some distance from the content, and some outside help, it’s important to realize that the process of writing the exec sum can also be part of the process of training the author to be an effective communicator. S/he may be fed up with it, but don’t let the author off the hook: involving the author closely, even pushing her/him to do it, can be excellent preparation for summing up the report in conversation, media interviews, etc.

  10. I have (perhaps worryingly) strong views on this. You should write the Executive Summary before you start writing the report. If you can’t write the ES then you haven’t got your argument clear enough to start writing anything longer.

    An old boss of mine used to tell me this all the time and I resisted it for ages but I now find myself insisting on it for my team.

    Don’t knock it until you’ve tried it.

    1. I have strong views the other way Andy. If you think you know exactly what you want to say before even writing the report, what’s the point of doing the thinking, reading and writing involved? I would say set out your initial thesis in the ToRs or whatever other bit of paper you use to kick off with, and that should help get your thinking straight, but you have to be able to develop and change your thinking en route, or we risk heading into policy-based evidence making territory!

  11. An executive summary should not read like a summary. Try putting on a different hat or ask an outsider to write down the essentials for you.

  12. I do think the approach of a short build up to shopping lists of recommendations is one of the most common reasons for uninspiring summaries. It is no surprise the recommendations of policy reports are frequently the things people wanted to influence when they began writing, rather than the most significant or novel findings uncovered during the work. And it is obviously tricky to advise policy wonks not to focus on the list of Dos and Don’ts. But it must be a good idea if engaging readers’ minds rather than name-checking related policies is important.
    Maybe it comes back to something on a previous P2P complexity blog – that we sometimes (and especially if stimulating debate is the aim?) need to put our finger on the right questions, and not pretend to know all the right answers.

    I recently read a report called “Revitalizing Agriculture in Myanmar”. Content aside, the introduction (no summary here) grabbed me both because it acknowledged up-front the recommendations problem, and because it did so by speaking directly and frankly (brusquely, even) to me, the reader. I immediately thought this might actually be worth reading.

    “The worst way to read this report is to turn directly to its ‘Conclusion’ and to look for a set of detailed recommendations from the research team. While the concluding section of this report does indeed offer a series of general recommendations, they are neither so detailed as to permit of immediate implementation nor readily comprehensible without reference to the report as a whole.

  13. Lot’s of good advice here! Executive Summary should not be like summary.

    Summary should be content mastery by including and excluding information based on relevant reader’s objectives.

    For more information visit our website http://bit.ly/exec-

  14. This really helped me, Duncan. I’m working with 30+ Oxfam India research papers for LKET creating descriptions which will appear on iLibrary and the P&P database and website. Although I’ve had experience in a similar role, the art of summarising someone else’s work takes a lot of brain power — and is a lot of responsibility. So I thank you, and to those who commented above, who helped me revise, rethink and re-summarise.

Leave a comment

Translate »