Quantitative Research 1

Don't fall asleep! This is the bit of the planning basics when we start to talk about quant. It's something you can't avoid as planner, but as far as I'm concerned, if you treat numbers right, they can be you're friend.

This is not a quant masterclass by the way, it's some useful starts. No more no less.

First up, here are some initial principles......

Starting with WHEN to use it.The accepted way to do research is to start with lots of qualitative, in depth, discursive research. Once you've learned something you can use, you use quant to prove what you think you know, with a big sample answering the same questions. This is fine, but you can do a lot more than that. For a start, it MAY be useful to do more quant up front if you have the kind of client that deals only in fact. That doesn't mean you leave out innovation or inspiration, you just have to know how to use the numbers.

Quant is useful because it helps you get closer to clients - numbers are their key performance measure, so getting to grips with the discipline allows you to join in the debate. If you're REALLY going to show you know their business and customers, it's useful to do it through stats.

But it's pretty useful anyway. Here are the ways it can be used:

FRAMING THE PROBLEM. It's really useful for showing you what the communications TASK is. For example, Skoda wasn't selling enough cars, despite the fact people seemed to like the brand. They found that they were getting on people's initial consideration lists, but the numbers were showing they were not following through to the final shortlist and purchase. The numbers framed the problem - AND the solution. The task was to show why the cars were good AS WELL as the brand

Data is the ONLY way to prove your campaign has been effective. This comes back to the the very real need for you and your clients to know you really are affecting profits... and you won't win an effectiveness award with focus groups.

Now unless you're very lucky, you're not likely to have a quant research in-house, it may be the case that you have the type of c lient who does their research independent of the agency, or you might have the sole responsibility of organising a research company to do the job. In all cases, you need to have input into the research brief and you need to understand the methodology they'll use.

Like any brief, what you get out is directly related to what you put in. The innovation starts here. If the research is your responsibility, that speaks for itself, but if the client has hired someone, you'll want to have input into the brief. Not only because you want to make sure there's something worthwhile coming out the other end, but also it's down to the fact that independently hired research companies make conclusions and recommendations. That's a chance to cut planning out of the process. You don't want that, so have input from the start, get them uncovering what YOU think is important. Otherwise you'll end up with very expensive, very useless information, or you'll have the strategy done for you.

So make the researcher your friend. I can't recommend this enough. And if you have the choice of who to hire, pick someone who's used to working with agencies - you need someone who knows where research end and planning starts.

(Don't forget much of this is from a smaller agency Up North perspective, other experiences may be different. If that's the case please add your point of view in the comments)

Now none of these 'basics posts' can make MAKE any good at planning (assuming they make any sense in the first place), so you won't become a quant wizard over night, that needs learning by doing. But we're going to cover the basics of:

1. How to go about developing a questionnaire and methodology to develop a new strategic direction.

2. How to go about quantitative pre-testing.

3. How to go about testing effectiveness.

It will be from the perspective of a planner working with a research agency, and while it may involve some of the things you'd like to do the least, you can't avoid this cropping up, so you may as well grasp the nettle.

First bit next week.

Comments

[Charles Frith]

Good work NP. I would add that a common fault of young planners and research companies is to research everything just because they can. It dilutes the process, causes confusion and isn't helpful at all. Focus on what you want to find out. Ask yourself if the brief is single minded and if it isn't save the poor respondent from getting bored by asking only the questions that really need to be asked.

Research has a habit of expanding if not controlled.

[Marcus]

Excellent stuff NP. Bloody hell, you're on a bit of a roll at the moment.

I'd also like to build on Charles' comment and say that the young folk should beware of pushing data into a place they need to have it to PROVE something for an idea that they really want to happen.

[Rob Mortimer]

Another great post, and good comments from Charles and Marcus.

This is all really useful stuff. Thanks.

[np]

Charles, you're not wrong!
Marcus, nor are you.
There's a fine art to applying the correct focus and just proving your point.
Rob, thankyou.

[stuart]

Where posible get creatives involved early.

Collating all the info and presenting it on in an easy to understand form is one thing. Coaxing a creative to believe 100 people know more than their finely tuned creative brain is another.

For too long there has been a resentment of research as either the place where good ideas go to die or the ugly mother of painfully dull briefs. But if the quants are to be believed, the tide is changing.

I worked on something recently with NP where I was involved during the research process. The look on someone face or a room full of people nodding moulded what we did instantly. Every group added something and every time we became more sure that what we was doing was right.

I like many creative still have a lot to learn about the role of research. But do you know what the most worrying bit is? I've stopped hating research and actually started to (no, can't bring myself to sat it).

[Rob @ Cynic]

Great post NP ... this really is taking A[P]SOTW to another level.

To put my 2-pennies in ... research is a powerful validator and liberator for creativity but in the wrong hands or with the wrong expectations [from both client and agency alike] it can be a noose around your neck.

Having a 'reseach friend' [or even one in the team] is possibly the best way an agency can get a client to feel 'safe' about investing in pragmatic creative because without quantifiable data, we'd of never had ads like Tango / Pot Noodle / Carlton Draught / Burger King ... etc

[lauren]

this is really interesting to me (is that sad?) and i'm glad people have mentioned that too much research will strangle something, but it's a good reminder that acutally doing the research is still important. in visual arts practice, it's not quite so formulaic, but it's still vital to ask your audience.

[Charles Frith]

I think a post on how quant research fought/defended good creative would be terrific. I've only got regional examples on advertising that I had to adapt.

Anyone else up for this too? Rob, any chance. You've worked on some terrific stuff?

[np]

Stuart, you do realise I'll use this against you don't you.
And I think that's a good ideas Charles, let's see what the Cynical one does.

[Age]

This is so valuable to me, looking forward to more.

Thanks np.

[Lee]

Nice post NP.

Some additional, hopefully helpful, thoughts for newbie planners re: quant:

1) Anything in quant data that looks interesting is usually wrong - you should be able to predict the answers you get back based on your knowledge of the market. If something looks out of kilter with your expectations it is rarely an "insight" and more often a mistake. Get someone to check anything that looks too interesting before you embarrass yourself in front of the client.

2) The answer to the strategic questions you want quant to answer to frame the problem are usually already available in some form within the company. Make friends with the clientside researchers and you will often find that the data was sitting there all along but you just didn't ask in the right way. Some clientside researchers consider themselves to be responsible for the development of strategy and may therefore see themselves as being in competition with agency planners. Help them and they will usually help you.

3) Always ask for the raw data and tabulations. Even if you don't know what to do with them now, always get hold of the raw outputs of the research. A debrief only tells part of the story. A good planner will no doubt have additional questions (perhaps years later on a tangential topic) and the research agency will not always be there to help (they are doing project work and are not usually on a retainer). If you have the tables then you can look for yourself without having to get the research agency involved again (slower and potentially more costly than doing it yourself). Who knows, one day you might become proficient enough to conduct some analysis of the raw data yourself.

4) Get interim results. Fieldwork can be slow, especially if it is a big study or if you are doing face to face research. As such it can be helpful to get an early read of the data to get an indication of the direction that the results may go in. Remember that this will not necessarily be totally representative of the final results due to variations in the sample.

5) Just ask. If you don't understand something about some of the more esoteric quant stuff then there is no harm in asking the research agency (in fact, you are doing your client a disservice if you don't understand something and stay quiet). Planners cannot all be experts at every discipline that is relevant to developing comms strategy (somewhat contrary to the inclusive 2001 APG "definition" of planning).

NP - please try to cover off the basics of sampling in one of your follow up posts.

[np]

Thanks Lee, I always hoped people would build on this stuff.
And thanks for sampling suggestions, hope you help out again.