In commercial marketing research, there is much competition. This leads to ‘negotiating games’, whereby clients want the best deal and agencies may ‘buy business’ in order for contracts to be awarded.
Sometimes budgets are stated on a brief; sometimes they are not. Sometimes the stated budget is well below a real amount available. Clients sometimes argue that if a budget is given, the agency will make sure they spend it all. If budgets are not stated, agencies may argue that they may be wasting their time to create an elaborate solution to the problem. This may be fully justified if the response is that it is a good solution, but there are no funds. Another ‘negotiating game’ is when a client asks for several proposals, then chooses the best ideas and ‘suggests’ them to one of the agencies with whom that client wants to work. All of these problems can be avoided by developing honest and sincere working relationships. Asking for no more than three proposals is reasonable.
Terms of business
The terms of business include legalities and details such as ownership of materials and intellectual property. Some companies incorporate a confi dentiality agreement, whereby personnel in a research supplier company agree to avoid discussing any details of a project with anyone not directly involved in the project. Some agreements contain a ‘non-disclosure agreement’ to reassure a client that the agency will not share or disclose any of the data provided by the client. The agreement may state that, on completion of the project, the agency must delete the customer data after a certain time period.
Credentials
The credentials are there to reassure a client that the personnel being used are capable and experienced. If previous clients have agreed for their names to be used in promotional materials, they may be mentioned in this section. Newcomers to research are often surprised
by the fact that this device is used so little by researchers; advertising agencies are very active
in mentioning previous clients. The main reason for avoiding mention of previous clients by researchers, however, is a desire to give prospective clients assurances of discretion. Credentials vary from two lines of mini-biography through to full résumé (CV) forms.
Selecting an agency
Research agencies are sometimes considered in the same way as advertising agencies. They both actively seek new work and in both sectors this is called making a ‘pitch’ for business; they will both be given a brief by the client, they will both make proposals, and they will be commissioned
to carry out their jobs. During and after work, debriefs will communicate progress.
Part 1 Research preparation
56
However, there are diff erences that sometimes leave newcomers confused. One diff erence
concerns credentials: it is obvious if an ad agency works for a particular company—they work
hard to communicate on the company’s behalf. This then means that the client name can be
used in the agency’s own promotional materials. Generally speaking, ad agencies will not work
for competing companies—there may be a confl ict in commercial interest. In contrast, market
research agencies pride themselves on protecting commercial secrets and this often means
that client names do not appear on promotional materials. This can have a useful benefi t to the
agency because they can accept work from competitors simultaneously. The downside is that the
usefulness of ‘big name’ endorsement is lost and clients need to fi nd other ways to select agencies.
The decision to select a marketing research agency should be well considered. The
prospective buyer of research will follow the steps outlined in Table 2.8. The emphasis is on
fi nding a supplier that understands the research needs well; it should have experience in
the sector or in the particular methodology anticipated. There is also the matter of personal
Selecting a research agency
Step Activity Description
1 Internal
discussions
Discussions internally about the need for research—what is the purpose? What action will be taken?
2 Create brief The creation of a written document to outline the nature of
the problem and instructions to a research agency
3 ‘Long list’
agencies
Create a ‘long list of agencies’ from various sources. This will inevitably use published directories, but will also draw on the experience of colleagues and recommendations
4 Shortlist
agencies
It is normal practice to brief just three agencies, so the long list needs to be reduced, keeping some in reserve in case any of the shortlisted agencies are unable to provide a proposal
5 Briefi ng A briefi ng may be a meeting, telephone call, or just sending
a briefi ng document by email or by post
6 Respond to
queries
Agencies will now prepare a proposal, but will inevitably need clarifi cation on specifi c points, usually concerning the company background and sampling
7 Receive
proposals
The three proposals are received and considered
8 Reject and
accept
The ‘winning’ agency is informed about the decision and any fi ne details are agreed. The proposal now forms the agreement between buyer and seller. A formal acceptance is made and this is a legally binding contract. The other agencies have invested time and energy in their proposal,
so it is reasonable to give full feedback on why they were unsuccessful. Besides being good practice, you are likely to
be working with them in future, so it is important to foster the relationship.
Table
2.8
Chapter 2 Planning research
57
chemistry: the relationship is a partnership and if the respective teams are unable to work together the project may fail. Clearly, this becomes more important with continuous, ongoing projects. In larger organisations, a ‘roster’ is in place. This means that there will be a list of preferred suppliers that have been screened for suitability; the list is continually refreshed with feedback on past performance. At any of the stages, an agency may be invited to visit the prospective client in order to present its credentials.
Evaluating proposals
Having written a clear brief, it is usual to receive proposals from three diff erent research agencies; this allows the research buyer to make a considered judgement about the fi nal choice. It is then a matter for the ‘buying team’ to decide which agency to use. It is wise to create criteria specifi c to the project, which will embrace:
1. Resources of the agency (fieldforce, regional strengths, etc.)
2. Expertise in the sector (use of correct terminology, understanding of market)
3. Reputation (well-known references, advice from clients)
4. Understanding (originality of solutions, understanding of objectives and brief).
Although there is much simplifi cation, the matrix in Figure 2.11 will help you to visualise the strengths and weaknesses of diff erent proposals.
A = Misleading and poor value for money
B = Fair
C = Study is misleading
D = Good value for money.
Beyond the design and cost of research, there are other factors worthy of consideration. What quality control procedures are outlined in the proposal (for example, ‘callbacks’)? Are
Low cost
Inaccurate Accurate
High cost A B
C D
Figure 2.11 Proposal evaluation matrix
Part 1 Research preparation
58
respondents recontacted to confi rm that they actually participated in an interview? What
measures are in place to ensure errors are minimised in the transfer of data from paper to
computer? If the research team move to another project, what provision is made? Is the
agency large enough to carry out all tasks? It is useful to create a checklist of items to evaluate
the proposals—the checklist will vary from project to project.
Validity and reliability
Successful researchers must consider two key concepts: validity and reliability. If a study is valid,
it means that it measures what we think it should measure. If it is reliable, it means that if we
repeated the study, we would get the same results. Each time we come across a new study, we
should ask ourselves how valid and reliable that study is. This constant ‘criticism’ is essential for
the market researcher. Questions about validity and reliability need to be asked about every
part of the study procedure: the questionnaire, the sample, the data capture method, the
interviewer, etc.
Market researchers know that anything and everything they do can aff ect the outcome of
their study. The diff erent components of research, such as sampling, the wording of questions,
the method used, and hundreds of other things, can change results.
One example might concern radio listening. A question may be created as follows: ‘Q3
Yesterday, for how many hours was your radio switched on?’. If the intention was to measure radio
listening hours, the results would be incorrect. Many people leave the radio on when going
shopping; they do this as a deterrent to burglars. The validity is in question here.
Consider another example: if you interview a representative sample of the nation’s
population every week, you can detect whether more (or less) people read a particular
newspaper in comparison with another. If you decided to off er a prize draw as an incentive to
do the interview, it may aff ect the results. It might appeal to young people, thereby increasing
the sample size of youngsters, who may not be readers of that newspaper. The results may
then indicate a decrease in that newspaper’s readership. Here, the reliability of the study would
have been aff ected by the incentive.
Is research really needed?
Research may not be necessary—there is always a choice not to listen to customers. There are
many reasons why market research may not be used; there are problems with those decisions
that are ‘data driven’. The fundamental question is whether marketing decisions are informed
by data or driven entirely by human judgement and experience. It is possible that experience
can give confi dence and this prior knowledge can act as a proxy for new facts. If data-driven
marketing decisions are made, there are several risks: using inaccurate data, misinterpreting
data, or even misusing data.
There are some other reasons why research may not be used, including:
• Study needs are misunderstood
• Proposals may not be convincing, or the overall comfort level is low
• High cost may mean the research does not fi t a budget
• Low costs may evoke doubt about the seriousness of the supplier
Chapter 2 Planning research
59