When a Market Research Supplier Must Consider Saying No
The best market research suppliers are service providers who seek not only to satisfy their clients’ research needs, but also to deliver an exceptional project – from the research product itself to the business insights and personal interactions and experiences. Sometimes, in our effort to best serve the client, it is easy to succumb to the pressure to please and say yes to a client request. And in most situations, that is just part and parcel of being a responsive, high-quality supplier.
But in certain situations, market research suppliers need to be better at saying no in the interest of their clients – for example, when the client request may unintentionally have an adverse effect on the quality and integrity of the work.
Here are a few examples of situations where you should consider pushing back on client demands:
- “Ask it this way.”
How to Brand a Clinical Trial
So You Want to Learn About Clinical Trial Branding? Clinical trials, like anything else, can be “branded.” But what we mean by clinical trial bran...Read more
During the design of a discussion guide or survey, clients often share with us a list of questions they want to ask. Frequently our gut reaction is to implement them all without any modifications. We think, “After all, they come from the client, they must be right.” However, these questions are likely to reflect what the client would like to know, which may not necessarily be the right questions to ask. While it’s great for us researchers to hear what type of information clients need, their questions may need to be “translated” into research-oriented questions. Such questions need to fulfill several requirements – e.g., they can’t be leading, may need to be polarizing, may need built-in redundancy for verification, may need proxies when direct questioning would lead to people-pleasing responses, etc. All these factors need to be considered before any questions end up in a discussion guide or survey.
- 3-in-1 Projects
Clients typically have a plethora of information needs but have limited time and financial budgets. This creates the perfect storm of pressure to jam as many objectives into an approved research project as possible, and why we are not surprised when we see the number of questions mushrooming exponentially as we get closer to the final version of a guide or survey. When that happens, it is critical for the sake of the quality of findings that we guide the client through question prioritization aimed at using only those that closely examine key objectives. Please refer to this post for some tips on reining in “everything but the kitchen sink” discussion guides.
- Quantifying the Qualitative
Clients who are more analytical and/or less comfortable with qualitative research will be more likely to ask their market research supplier to quantify specifically how many respondents said the same thing or similar things in qualitative research. And it’s understandable – even in qualitative research, we have to help our client definitively understand what “some respondents” or “many respondents” represent. Still, going so far as quantifying everything is a slippery slope. Despite any assurances that there is no reliable meaning in such numbers, it is human nature to believe that five people saying something is more important than two people saying something, assurances to the contrary notwithstanding. Be deliberate and cautious when putting numbers on qualitative comments.
- “Now, let’s cut the data a few more ways …”
The conceptual equivalent of No. 3 in quantitative research is being asked to run multiple, sequential “cuts” on a given data set in the absence of defined hypotheses. Spurious “effects” (statistically significant effects that are not real and based on chance alone) increase in likelihood the more analyses get performed. Yet because the cost and effort has been sunk and the data are there, the temptation to keep looking is real. We get it – some information is better than no information, so why not see at least if the data are “directional”? So, when it’s not possible to agree upfront on an analytic plan in which particular hypotheses will be tested, then it is incumbent on the research supplier to correct for the inevitable increase in the likelihood of finding an effect that is actually a fluke of sampling (Type I error) by tightening the p-value of any significance testing (e.g., going from p<0.05 to p<0.01 or p<0.001).
- “I need the report … yesterday.”
Given today’s fast-paced business environment, decisions often need to be made even before the fielding is completed. Therefore, it is only natural for clients to rush to conclusions. They are pressured by internal stakeholders to share insights quickly, so they often ask us to turn them around practically overnight. We understand the desire for instantaneous takeaways, but rushing through full report generation may lead to inconsistent or even incorrect conclusions. Most of the time, clients do not need full reports right away anyway, but only select key insights informing their decision-making process. That is why, while still in field, we discuss with clients their immediate needs and determine what findings and accompanying strategic implications need to be analyzed and generated right away and focus on those first.
While clients are experts about the brands they manage, they expect us to be market research specialists who know when to take an initiative and propose alternative ways to handle various aspects of research execution. Therefore, don’t shy away from pushing back when necessary. When it is well-justified, good clients understand and appreciate it.