Blog Opinion

Market Research: Is There Just Too Much of It?

In his latest submit, B&T regular Robert Strohfeldt argues that despite all of the market research, is it actually telling adland what it wants to hear?

Change. The one variable that is constant. And often, change means better. Nobody modifications to something that was worse than what that they had before. For many who hanker for the “good old days”, think about what it will have been wish to go to the dentist 100 years ago?


Approach again in 1981, once I first joined a market research agency recent out of college, I was engaged initially as a mathematician to ensure proper random sampling.

Analysis companies had their own area drive again then and most quantitative surveys have been completed door to door, nose to nose. We labored from Assortment Districts from the ABS. A serious problem was value – an interviewer was given a strict set of dwellings and if no one, or the appropriate individual, was not house, they have been to attempt 3 times before alternative a dwelling was allowed.

Respectable sample measurement, quantitative research have been too expensive for all but the giant corporations. Anecdotal findings, coupled with expertise, have been typically all small companies had.

There was also the obscene quantity of paper used – imagine a six-page questionnaire for 1000 respondents.

The quantity of successful interviews achieved in a day was “bugger all” and a pattern of 500 or 1000, took ages just to finish the sector work. And of course, the fee was typically seen as prohibitive for the perceived outcomes. (That the outcomes many occasions strengthened what the advertising group believed they already knew, was deemed a waste of cash).

In the early 80’s there weren’t any PCs – coded questionnaires (coding is a useful art that has been misplaced) have been taken to an organization who produced a pc print- out.  Typically, after working by means of the primary run, further reviews have been produced – “Can we see how the breakdown how of respondents in question four answered question six?”. On prime of this any statistical analysis was executed individually, once more in any case report runs had been accomplished.

Typically, the choice was “Let’s just run a few groups”. Great, an-depth discussions with 30 odd individuals was not likely an accurate indication of anything, bar the shortage of understanding of how research was greatest used.

Many purchasers would baulk at spending $30 to $50,000 to optimise a $5 million campaign.

By 1983 I had graduated to operating groups and have fond reminiscences of operating focus groups in peoples’ kitchens. The respondents have been genuinely thrilled that enormous shopper corporations sought out their opinions and engaged in discussions accordingly.

It was a novelty and as such they gave trustworthy solutions and when the subject was exhausted, that was it, the group was over. (Straightforward to spot) Yes, they have been paid, but many occasions I had respondents state they might have participated even if not paid.

Many golden nuggets of shopper considering (and statements, or in immediately’s terminology, content) have been gleaned from them.


Phone surveys took over from door to door sampling – no- where close to the fee of sending interviewers out into the sector.

For market researchers, their most useful pure useful resource is respondents. Phone interviews quickly turned problematic because of telemarketers.

“Hi, I am doing a survey on house cladding”. A number of question later they knew your sort of dwelling and when you have been a candidate for cladding. And in no- time there was a knock on the door, or typically a telephone name, from an individual making an attempt to promote you cladding.

And as telemarketing grew, many people had skilled having their dinner disturbed by somebody making an attempt to sell them something, camouflaged as market research.

“Piss off” turned a frequent response. Many households grew drained of receiving phone calls, be they from a real research organisation or a telemarketer.

By this time, PCs have been in all analysis corporations and the questionnaires have been on display.  As the interviewer “filled in” the questionnaire on display, the first run of knowledge was being completed concurrently. Not something to marvel at immediately, however when this first turned obtainable, it was revolutionary and analysis corporations tried to push all quantitative surveys into phone surveys.

But all of the whereas, researchers’ most dear natural useful resource, prepared and uncorrupted respondents, was regularly being eaten away.


Quick ahead to 2019 and I doubt there is a individual over the age of 18 who has not been surveyed in some type. As increasingly more businesses transfer on-line, they see it as a chance to tag a number of analysis questions at the finish.

The quality of the research outcomes is completely dependent upon the standard (accuracy) of the pattern – is it consultant of the group you need to survey? “Normally distributed, proportionate to size”, was a phrase used in every quantitative research again when door to door was the norm. The sample can be “reweighted” to the exact proportions of the varied demographic groups in the population being studied.

Research is identical as computer systems i.e. garbage in = garbage out.

I’ve little question that reputable research companies adhere to strict sampling disciplines, but this doesn’t remedy the problem of “too much research”.

I had lunch just lately with two pals who have almost 80 years of market research expertise between. They felt that too much analysis kills our natural resource – the unbiased and trustworthy respondent. Our chat shaped the genesis of this article.

Many giant analysis corporations at the moment run panels – tens of hundreds of individuals who, for a reward (typically money), are prepared to be respondents. By having all of the relevant demographic info of every panel member and by filtering by way of certain behavioural questions, they will supply a ready- made pattern.

However simply how random is that this group of individuals? They will distort focus group discussions – a pair might attend up to two teams every a month for $80 (or extra) per group. And they’re additionally paid to participate in quantitative research. This may give them as much as even $320 a month additional cash in hand, nothing to be sneezed at.

They’re paid to provide an opinion and so they do, however whether that is what they actually consider, or what they assume they should say, nobody can really inform.

“Do you buy jam” – “yes”

“What brand did you last buy?”. The model supervisor for Cottee’s jams is deeply concerned in the model, however the common shopper only provides it a cursory consideration when purchasing. “Last brand? Um, Cottee’s”. It will not be, but as they are paid for an opinion, they higher have one. It might have just as easily been another model that was on particular. But Cottee’s is prime of mind and gets the vote.


After which there’s the basic “How likely would you be to recommend…….?” The Nett Promoter Rating. I read not long ago the new MD of a big personal medical insurance firm was going to link all senior executives’ bonuses to Nett Promoter Scores. Poor bastards!

Mark Ritson wrote a column lately about CEOs’ messing up advertising and advertising- what a colleague refers to as “Seagull Management” – they fly in, shit over every part and fly off.

This was in all probability a superb instance. The MD had in all probability heard of The Nett Promoter score and thought “Bloody brilliant. I will introduce that”.

The place once a survey was something uncommon and genuinely considered being worthwhile, individuals are asked to fee one thing almost each day. And therefore we have now “Seven sounds like a good number”. Nett Promoter Scores might have been worthwhile years in the past before the pure resource of “virgin” or at the least “rarely used respondents” had been used up.

Now, individuals rattle off numbers between 5 and seven, until one thing terribly good or dangerous had occurred to them.

A/B testing is another hoary chestnut. Not the place to detail an in-depth dissertation on the problems with this system. It has turn out to be so commonly used that I would like to take the opinion of an skilled and intuitive artistic individual to finalise a headline or advert idea, over the answer given by some spreadsheet A/B testing method.

Survey Monkey has allowed smaller businesses to conduct their own research. Typically this does throw up useful info, but typically it’s carried out by individuals with no experience in sampling and interviewing. The analogy is of the one that acts as their very own legal counsel has a idiot for a shopper.

We know that because of Google, Social Media and virtually no end of info sources, consumes are much better informed than ever.

I have seen focus teams by which the respondents have been appearing extra as artistic administrators – “It would have been more impactful if you had used morphing”.

Testing communications towards their aims is turning into increasingly testing.

Knowledge appears to have usurped research as a source of insights. In many corporations, the market research and knowledge insights are separate departments. Where as soon as we referred to knowledge as Main Analysis – all out there info (knowledge) was analysed prior to any market analysis being developed – what do we all know, or assume we know, what gaps are there’s our information and so forth. Market analysis and knowledge have been all part of advertising intelligence. There was no demarcation between the 2, they worked collectively seamlessly.


There have been many unbelievable new research methods and programmes developed over the past 40 years. That may be a given. Nevertheless it doesn’t tackle the only largest drawback dealing with any type of market research. As with knowledge, quality ought to all the time take precedence over quantity.

Regardless of how good the method, if the sample is a dud, then the outcomes will end up the identical. Nielsen launched, a quantity of years in the past, a qualitative analysis method they referred to as “Delta Qual”. Derived from forensic psychology, its goal was to determine what individuals truly did when grocery purchasing, not what they thought they did.

Not a easy or low cost method, it required two moderators. Although this strategy is extra probably than many methods to seek out the reality, it’s still heavily dependent upon the standard of the respondent.

“Killing the goose that laid the golden egg”, was how one researcher described the overuse of research.

This can be a matter through which it is far easier to determine the problem than it’s to offer an answer. Cliched as it is, the answer is “each case is different”.

Prior to any research research, remember of the truth that many of the respondents are jagged and provides an answer off the highest of their head. When you’ve got acquired feed back from say 5,000 individuals and four,000 of them are just going by means of the motions, then the results can be rubbish.

Tips on how to determine and entry the respondents who are genuinely fascinated with providing answers? In fact, this then raises the problem of statistical validity. Ensure you know the demographic and related behavioural breakdown of your target group after which reweight the results accordingly.

With qualitative research, converse to the people who find themselves both very joyful, or very pissed off.

Have you ever ever been to a restaurant where the waiter or waitress constantly interrupts to ask, “How is everything”? Perhaps we need to study from this.  Cease asking questions after every buyer contact and study to determine those who actually do have something to say.