He then lists a number of ways of doing this. I summarize them a bit for brevity.
- You could talk directly to customers.
- You could talk to a sample of customers (via focus groups or user group meetings or similar).
- You could rely on voluntary feedback (like those ghastly product registration cards so many things seem to come with).
- You could trawl the web looking for blog entries about your product.
- You could solicit feedback from your own sales team.
- You could ask your support staff what problems customers keep coming up against
- You could talk to industry pundits, who will guess what customers think of your product.
- You could ignore the customers and assume that they want what you want.
In another comment, Michael continues:
To pick an example, say you were involved with something like a consumer-grade digital camera (lots on units, lots of customers, scope for plenty of product differentiation). Given the volumes, you can’t talk to all the customers. Given the sales channel, you might not even know who most of the customers are. What can you do for something like this?
For the example above: a consumer-grade digital camera, let’s take a step back and talk about market segmentation. When the camera was being conceived there was an audience in mind for the camera. Today the digital camera market has matured enough that there are clear market segments at which cameras are aimed. For example, a simple segmentation could be:
- toy cameras, intended for use by children
- compact point and shoot, for the casual photographer,
- prosumer cameras for the more serious photographers
- digital single lens reflex (DSLR) for very serious and professional photographers
Now each of these segments defines different types of people, with different budgets, photography experience and photographic intent. When looking to design a camera, you would need to get a clear definition of the segment you’re aiming for, understand the dynamics of that segment (competitors, trends, subsegments etc.) and then start developing the product.
As for gaining insight, once you know the segment and the characteristics of that segment, you know the type of people to talk to: which end users, retailers, distributors etc. Those 8 questions listed earlier (OK, #8 — ignore the customers is probably not a good tactic) are all ways to gain insight, both before, during and after the product is released. Of course, there are many others.
Shifting to software, one of the most well known and oft-cited examples of gaining customer insight was Intuit’s “follow-me -home” program. This program had Intuit staff visit the homes of Quicken users and observe them using the product in their home settings. This program yielded valuable insights into what people did with the product in their home environment and led to several product enhancements such as an electronic paper tape feature in the Quicken calculator.
BTW, if you want to read a great book on Intuit’s history and struggles, I recommend Inside Intuit. It covers, what we can now call, the early years, and if I recall [I read the book a long time ago] has a section covering the follow-me-home program.
The follow-me-home program was an example of ethnographic research. Essentially it is fieldwork; getting out into the real world and truly understanding the environment and frames of reference of those who will be using, recommending, selling or distributing your product.
One thing to keep in mind is that when doing fieldwork, it is best to include several people from your organization from teams other than product management. The reason? Because, regardless of how much primary research you do, you will always interpret what you observe and learn through your own personal frame of reference. The patterns you see when conducting the research will likely be somewhat different than the patterns others see. And the conclusions you draw may be different than the conclusions others draw.
Include members of engineering, market and even sales if possible and feasible in your research. Also take other product managers along. At minimum, you’ll have a control element in your sample of interpreters of the data you collect.
I honestly believe it takes a very skilled person to do ethnographic research well. You have to be able to put your personal biases and views aside, and not only listen to what others are saying and doing, but also ask the right questions (open ended, neutral questions) to solicit the input that is needed.
I’ll probably write about this more in future post, but one thing I see lacking in virtually all product management training is any coverage of the skills needed to do good field research. It is such a critical part of the product manager’s job, but I’ve yet to see that topic covered with any level of substance in any training courses aimed at product managers.
The question: “How does a PM gain insight?” is a good one. The answer, quite simply, is that there is no simple answer. Like most things, it takes ongoing effort, communication, thought and dedication.