The use and abuse of research: Some principles to create by


As a former researcher turned agency planner I’ve worked both in research and with it. So as a gamekeeper turned poacher, I’ve always held rather strong views on research. I got a chance to air some of those this week, having been invited by Ipsos ASI to their seminar ‘Research as fertiliser, not weed killer’ in London this week.

It’s fair to say that what this presentation does not cover far exceeds what it does cover. The subject of research and effective creativity after all is vast.  So working on the basis that the essence of strategy is sacrifice, it felt helpful to restate some simple principles to work by. More specifically to share Wieden+Kennedy‘s beliefs about the best use of research.

Of course for some these will seem breathtakingly obvious. After all, the likes of Stephen King, Stanley Pollitt, and Alan Hedges were saying the same things forty years ago. The necessity of restating them might justifiably depress us, but perhaps it should not mystify us entirely. For as Voltaire once had cause to lament, “the trouble with common sense is that it is not very common.”

Good research can and does happen. And it can and does contribute to the development of great, effective ideas. In both the APG Awards and the IPA Effectiveness Awards we have plenty of evidence for that.

But too much research is still done for the wrong reasons, in the wrong way, and arguably by the wrong people.

Too much research has wholly outdated assumptions about the human mind, let alone how advertising works.

And too much research defines itself by its methods, rather than by the relevance and usefulness of its outputs.

My gripe then, is not with research per se (who would not want an understanding of the context for their efforts?) but with bad research. Of which there is still, far too much.

Enough with the preamble. First, here is the work I began with  – a selection of work from Wieden+Kennedy Amsterdam and Portland:

And here is a slightly expanded version of the presentation that followed:

My very sincere thanks go to IPSOS for inviting me to join conversation, and for having the fortitude and broadmindedness to allow me to share some unedited (and occasionally inconvenient) points of view.

And a very special thanks goes to a wonderful gent and a fearsome talent – Ignasi Tudela Calafell here at Wieden+Kennedy Amsterdam, who designed the awesome Salmon vs Lamposts poster.


  1. Alex

    Love this blog, thanks for being honest all of the time, and suffering fools none of the time. There’s a typo on page 92 though.

    • Martin Weigel

      It was not as inflammatory as I’d hoped.
      So either it / I was shit… or the research community has few views on its own future.
      Other than the good people who’ve contributed here, natch.

      • cookie

        Or even worse; they all thought they were already doing what you suggested them to do.

  2. John Dodds

    Do the early slides detailing share gain assume no action from competitors and/or no new entrants?

    • northern

      They’re based on the IPA databang databack, which in turn is based on isolating advertising effect from other contributing factors

  3. Paul Townley-Jones (@plannerpaultj)

    I am also a former researcher turned planner and I could not agree with you more. I have begun to develop a distaste for research as it is often done poorly, with objectives not fully understood and an ‘off-the-shelf’ approach applied to the discussion/interview. There is a market out there for research agencies who really understand the bigger picture

  4. Pingback: Nurturing Great Creative Through Insights « eat:Strategy
  5. saltis99

    (Another) Excellent piece – thanks for sharing. Would love to hear you elaborate on this – any speaking engagements in the foreseeable future in Scandinavia?

    • Martin Weigel

      Hi Olivier,

      To be honest I have often suspected that most of the quotes we bandy around were never actually said!

      Thanks for letting me know though.



      • saltis99

        As Napoleon (supposedly) said: “History is the version of past events we have come to agree upon.” The same probably goes for quotes…

      • olivier

        Let me quote you Abraham Lincoln
        “The things with quotes on The Internet is that you cannot confirm their validity”


        Anyway I guess we use quotes (and historical examples) more for the story we decided to tell using them rather than their exact accuracy.

        My comment was nothing more than a footnote 🙂

  6. Pingback: The use and abuse of research by Martin Weigel | MUJO
  7. Dave

    Thank you so much for sharing your thinking, your work continues to be a light of grounded thinking in this world of shiny and new. Your pieces on relationships have really influenced my thinking and I really liked how you tied feldwick and kahnman together in this one. It might have taken too much time, but Wilson’s poster experiment could have been great to add further support to your section about how unreliable people are when you ask them what they want.

  8. Pingback: La long data va-t-elle infirmer la maxime selon laquelle « le sens commun n’est pas si commun ? | «Notre Lien Quotidien
  9. Jean

    Thanks Martin, your thoughts are always a delight.
    Your next amazing move could be “how to deal with research oriented client when you despise it”.
    This kind of speech is hard to tell to clients who breathe thru research. Even emotionally and rationally proofed. What’s your secret beyond amazing executonial pieces of work people want to buy anyway?

  10. Juliana

    I totally agree that the quality of the research depends on the people doing it (planning, interviewing, answering), and that we should be really careful with research in general. Nevertheless, the John Ford//or not quote just proves me the importance of listening to people. For the world would be a much better and more beautiful place with faster horses instead of cars!!!

  11. Dom Rebello (@DomRebello)

    Really enjoyed reading your presentation Martin. I get asked for ‘nuggets’ daily – doesn’t irritate in the slightest, not one bit! 96-99 resonate. Truisms are better than nuggets I think.

  12. David B

    As always, stimulating and frankly much of it needed to be said. I’m a planner, but have worked with researchers as a consultant and have mostly run my own forms of research in the past.

    I remember a group for a lager brand conducted in Dublin. Upon concluding the group I took them out for a couple of pints to see them in reality. Before leaving the facility (facility!… Bloody hell!) I asked what they thought they would have. All but one said Brand A.

    En route I asked them all why they liked brand A so much, and they were effusive in their praise. When they walked through the door, they all defaulted to Brand B (which you will be able to guess as this tale progresses) I let it go for a round, but when they opted for a mixture of Brand B and Brand C in the subsequent round I asked why this was.

    Apart from simply feeling more comfortable in my company after a couple of hours, being in a more relaxed atmosphere etc, they said that as the bar was bedecked in the liveries of Brand B, they felt that it was almost a subconscious decision. Now I’m not sure if this is correct or not, but it shows that Brand B, whilst maybe not being particularly ‘clever’ in their approach, were damned effective.

    However, when I asked about the mixture of choices in the second round, they gave a variety of answers, ranging from “sticking with what I was drinking at the time” “The pump was off on the one I had before” “He ordered for me, and I just said a lager” it made me realise that there is absolutely no brand loyalty in the category at all. They employ system 2 thinking in their houses, sat in front of a TV, but when they are the point of purchase it is pure system 1.

    By the way, the client expressly banned me from taking the group out for drinks due to the CSR guidelines, but I did anyway on the basis that what they don’t know… If I hadn’t, I would think that the group would have been absolutely useless in isolation.

  13. Pingback: Personality Issues | Unfinished Article
  14. Pingback: “Research Early, Research Late, and Don’t Mess With the Middle” | Consumer and Business Intelligence - Don't let your data just sit there.
  15. Pingback: Nurturing Great Creative Through Insights | Creative Traction
  16. Doug Kleeman

    Martin, thanks for sharing. There are some very useful / smart distinctions in here. I’m writing a piece right now on some of the flaws of creative testing and poor research practices. I’d love to use a few quotes out of this deck and I wanted to see if it was cool to embed this deck of yours into my blog. I want to give credit where credit is due.