Don’t do it
Yes, the very first question is whether its really worth it. Are you going to be able to turn any new insights into business gains? No one should do research just to “find things out”.
Look at what you already know
A variant of 1, but maybe you already know the answers. Read the last ten shopper projects your company did
Do it in steps
Don’t blow a big lump of money on any one piece of activity or methodology. Start small, learn a little then go from there
It’s amazing how much “obvious” stuff you can learn from small scale observations, accompanied shops or focus groups. The smallest is, of course, to do it yourself. Just watch what’s going on!
Spend loads of time thinking about your sample
It’s amazing how easy it is to buy lots of samples that turn out not to be needed (too many shoppers who just seem “so- so” in their attitude), or conversely, miss sample on the things that matter ( big enough sample for your “problem” segment)
Get plenty of quotes
Don’t just give the project to the same outfit you used before. Yes, it’s great they are building cumulative knowledge but maybe the new kids on the block have a fresh approach. Look at more than one methodology too.
Reach out carefully to new technology
Things are changing fast in market research, for example, new approaches based on cell phones or social media can be remarkably powerful and inexpensive, but don’t go new for new sake.
Look over the parapet
There are great Industry or shared sources that can tell you an enormous amount without much of the cost and hassle of commissioning your own project or at least allow you to spend your money on a sniper rifle, not a shotgun.
Plan plenty of time
Nothing wastes money faster than projects with inadequate thinking and preparation time. You’ll get penalised on value for money. Don’t spend 10 weeks thinking about it then give the agency 2 weeks to deliver it.
Ask yourself the killer question upfront
What do I expect to find out, and if I find that out (have it confirmed) what would we do differently as a result? If you can’t come up with a half decent answer to this, don’t even pick up the phone.