CPI - Is it meaningful?

CPI [Cost per interview] figure is an interesting concept, originally meant to give guidence on pricing. However, the CPI concept is misused many times, because many researchers and clients are referring to it as an indicator of price level [too expensive or cheap]. The assumption is that total project cost can be broken down to the cost of one particular interview unit [or as if the cost of all project activities could be related to any particular interview] multiplied by the total sample size. There are clients asking for CPI figures and these figures get into long term contracts which is a double edged sword for all involved. Why is that so?

The point, regarding the use of this indicator, is that not all activities can be related to a particular unit of all interviews. For example, project and questionnaire design, data processing, the depth of analysis, reporting and presentation can all be independent from the number of interviews to complete.

Based on the above, if we compare projects with sample sizes of 800 and 1000, it is easy to understand that, in most cases, price of the above mentioned research activities should not account for the price of the additional 200 interviews.

So, to complete 800 interviews, a unit of all interviews may cost 15$ with analysis and all, but with the same project specification to add another 200 interviews that would not add any more work to the project overall, we would get another figure. For 200 interviews we may charge 10$/interview only [because cost of analysis and other activities has been associated with the other 800 interviews already].

So, the point is that for a larger or smaller sample size the CPI figure should remain the same only if change in the sample size results in proportionate change in the amount of work related to all project activities. Now, that's rarely the case..

Why is CPI a double edged sword? If agencies have agreements with clients on fixed CPI figures agencies win if sample size is larger than the one the price was agreed upon and clients win if sample size is smaller than the one the price was agreed upon. The same happens when eg. the complexity of analysis decreases or increases.

In MRCC™, although we use CPI figures due to these figures being somehow required, we provide another figure that is associated with the cost of a unit of additional or failed interviews. This figure shows that if all project parameters remain the same, any change in the sample size would effect price based on selected research activities only.

So, the CPI figure should not be used as an indicator of how expensive or cheap the project is [because it does not add any more information to the total price figure], instead it should be used to describe how much completing one interview cost, excluding the cost of all activities that can not be proportionately related to the size of the sample.

Hourly rate for market researchers?

I find it strange when programmers charge an hourly fee for their work. Can programmers success or effectiveness be guaranteed? I do not think so. They may go on wrong directions as anyone else. Can they plan how many hours it takes to finish their assignment? I am not sure of that.

Still, when it goes to researchers, they can rarely charge for the hours they are likely to spend on a particular project although researchers work on a very tight schedule and they can rarely deviate from the project schedule. Seeing colleagues being overly tired and wearied of their clients is not fun at all.

One of the advantages of our cost calculator is that labor planning can easily be circulated among those concerned. I believe that for researchers or other colleagues it is just sufficient to look at the proposal and the related labor planning to advise if necessary.

The next step would be to enclose labor planning to clients so that they see what is included in the project plan. This could eliminate arguments if clients want more than planned originally. Well, with our product, you could do that, too. 

The cost of market research

How much market research cost? I have seen this title many times with no conclusion in the body. For those who would like to know the real cost of research should have to learn about the operations side of market research. It does take a long time, because there are usually many different activities, skills and knowledge are involved. So the real cost of market research is not easy to calculate.

Prior to developing the cost calculator, I had the task to quote many times. Although I had years of experience in market research, it was way too hard to put a price tag on a project. There were too many factors to consider, let alone knowing clients and what they can absorb. So anyone who asks the question usually gets the answer as a list of items that effect market research cost. And I think that is the best answer one can get in the cyberspace. Unless he or she wants to deploy a product meant to answer the question.

I have developed my product for nearly 2 years. It can only calculate cost for the 5 major market research project types. Countless hours has been put in to reflect market research operations and to come up with something consistent. I think the answer to the question "How much market research cost" has finally been given.