Conducting a “B to B Survey”

Many DIY research tools are available to give a feeling of empowerment as far as research is concerned. These tools make it look easy to design a Q’re, but often these DIY Q’res do not fulfil the objective of the study. It surely creates Q’res and answers. Enough data points to make us feel empowered, but not providing the crucial business intelligence to really act on. Business-to-business target market is small and relationships are critical. If a valued customer or prospect receives a poorly designed survey, they’ll quite rightly question how much you really value them and their opinion.

My suggestion for the business-to-business survey is using professionals, who can build the right Q’re with a minimum number of questions providing clear objective oriented data. So let me share a few tips for anyone looking to conduct business-to-business survey themselves. But if you decide to really take the DIY approach, here are my suggestions.

An EIGHT step method is what I suggest for your B to B survey:

  • Objectives of the survey must be drawn out clearly with clear priorities assigned to all
  • Assign different priority to all of them. If you have 30 questions prioritise each with unique values between 1 and 30
  • Mark jargon-filled and ambiguous questions. Rewrite them with jargon-free text, keeping the objectives in mind
  • Remember to have very few questions in the final Q’re. My suggestion is between 5 and 20
  • Decide whether telephone, online or face-to-face is the best method of delivery
  • Incentivise it appropriately by offering a future discount or something even better!
  • Pilot the survey among a small group of friendly clients to understand if your Q’re is fulfilling your objectives
  • Validate the findings and act on them

Few more pro-tips for writing the Q’re:

  • More than 20 question survey – If a question doesn’t directly relate to your objectives or if you can’t say exactly what you’d do in response to the answer, then don’t ask it. My advice would be to aim for a survey length of around 5 – 20 questions max as business-to-business survey respondents are always short of time
  • The leading survey question – Be careful that survey questions aren’t deliberately or inadvertently leading the respondent to a particular answer. Read and re-read them to check they’re not biased in any way
  • The double-barrelled survey question – A question should only ask the respondent to comment on one thing so that it’s clear what their answer relates to. Check that you’re not asking respondents to feedback on multiple things with one answer option

Finally, once you have the response, check if the number is enough for making the data statistically robust and representative. By representative, I mean the distribution of the data reflects all segment of your market. Once you are satisfied with these you can start taking action on your newly found business intelligence.

Let me include a very well-designed sample Q’re. It has only 5 questions and does not include information like the name of the respondent etc. This is deliberate, as this Q’re should be delivered with identity tokens, which will record the identity of the respondent in the data automatically. Here is the link for the Q’re

Let me know what you think about this article!