Asking customers explicitly to list their needs is often a fruitless exercise. Customers often have difficulty articulating their thoughts and behaviors out of context, and applying it to the problem you are trying to solve (vs what they are). Instead, your approach should be to observe them in action, and interview them individually, and in groups, to deduce a set of customer needs. In this part we look at the most efficient mechanisms of gathering this feedback via interview and focus groups.
Needs are the backbone of the model
The backbone of the VOC model is the set of identified customer needs. Customer needs are not solutions, rather a description (in their own words) of the benefit to be filled by the product or service. Or better said by Harvard marketing professor Theodore Levitt:
People don’t want to buy a quarter-inch drill. They want a quarter-inch hole.
Well actually, no one really wants a “hole” per se, they want to hang a picture, or put up a shelf — but you get the idea. What we’re trying to ascertain are the core pains the customer wants solved, and focusing prematurely on solutions, can constrain the ultimate product service, such that you can misss creative opportunities to solve the real latent (unspoken) needs.
One-on-one interviews are more efficient than focus groups
Customer interviews are one way to gather the relevant data. While the VOC model does not delve into interviewing techniques, the focus of their research was to look at how to efficiently gather customer needs using both techniques, by understanding the point of diminishing returns.
In their experiments, the researchers conducted a number of two-hour focus groups, as well as one-hour interviews. Regardless of method, the goal was to ask customers about their real world experiences, and to compile a number of distinct “need statements”. The sessions were considered complete when interviewer felt that no new needs could be elicited from the customer(s) in the session.
After analyzing that data, they plotted the number of respondents vs. the % of needs identified, with each curve representing each technique (focus group vs. one on one interview). Here’s what they came up with:
- 2 one-on-one interviews are about as effective as 1 focus group
- 4 one-one-one interviews are about as effective as 2 focus groups
- After that, the marginal yield of a one-one-one interview exceeds that of a focus group
- Since focus groups often require much more coordination and resources, than one-on-one interviews, the interview path is often your best bet
Aim for 15 – 20 interviews
Since we’re often battling resource constraints, of time and money, it’s important to isolate the minimum effective number of interviews to isolate the bulk of the customer needs.
Once again the VOC paper describes an experiment whereby 30 interviews were conducted by 7 analysts, and uncovered 220 needs. In looking at the results, they plotted the number of customer interviews vs. the % of unique needs identified in each session, and arrived at the following plot.
- From this plot, it appears that 15-20 respondents should be sufficient
- Naturally, these results apply to each segment in the market, as well as the product category, so that depending on the degree of variability in either the product or customers this number can vary — but at least you have a rule of thumb
Use at least 3 analysts
Following from the methodology described above, when looking at the number of analysts employed to identify needs vs. the unique needs identified, here’s what they came up with:
- 3-4 analysts looks sufficient, and certainly 5 is a good upper limit
- Having a number of “different ears” on the problem is helpful to get a wider variety of needs due to various biases and perceptions each analyst ay bring — so selecting a more diverse group is likely a better thing
So far in this series, we’ve covered the basics of the VOC model, and how to structure your data collection efforts. In the next post, we’ll look at, how we take all of the needs identified in this step and structure them into a logical framework.