Tom Moor

Ask, Listen, Observe, Distill

What People Want

This weeks Startup Edition prompt is “How do you discover what people really want?” - at first glance the answer to this question might seem obvious, you just ask, right?! But the way you ask and how you process the feedback can make the difference between a terrible product and a runaway success.

Here i’m thinking from the perspective of developing and improving an existing early-stage product after you have validated the idea. For me, it seems natural to split the process of soliciting feedback and deciding upon production direction into several steps.

Ask

First, is your company really open to receiving feedback? Including a contact link in the footer of your website is no longer enough - you should allow and encourage users to get in touch using whichever channel they prefer. For example at Sqwiggle we include a contact/feedback button in the main menu of all our apps. You can also simply reply to any email sent from the service to reach support (here’s a great post on this by Joss, Founder of OpenExchangeRates). We also keep an active presence on Twitter and Facebook, making sure to reply to anyone that reaches out here too.

You’ll get the best results by combining and including a variety of channels. As well as email and social - how about blog comments, quora answers, discussion boards, phone calls, surveys and even face-to-face visits (a personal favorite of the Sqwiggle team!).

Listen

In the early days your users will usually be very forthcoming with their thoughts and ideas, even their hopes and dreams for your product. Managing and keeping track of this can be hard, there are a few tricks that I use including:

  • For smaller items and bug reports we find it easiest to simply add these directly into GitHub issues (or your issue tracker of choice!)
  • Tagging of emails so that we can find customers later and let them know when their idea has been addressed.
  • We use our internal discussion tool, to talk about ideas and their potential implementation and priority.
  • One of the key ideas that I’ve embraced since my time at Buffer is from 37 Signals - forget feature requests, If a feature is really wanted by your users you won’t need to keep track as it it will be mentioned repeatedly! For example in our case, features such as screen sharing and file upload both fall under this category.

Observe

Watching customers use your product can be one of the most painful, frustrating, eyeopening and ultimately useful experiences you will ever have as a software developer.

And finding people to observe is easier now than ever before, services such as User Testing let you perform tests at low cost and watch the resulting video back online. Alternatively you can place adverts on craigslist for testers and bring them into your office to record for a more directed/hands-on test.

Distill

I believe this is the most important yet also the most often missed step in the process. After all, nothing creates a poor product quicker than simply adding everything your users ask for! The great entrepreneurs of our time excel at the craftsmanship of distilling and filtering ideas down to their fundamental principles.

For example, Elon Musk regularly talks about his philosophy of reasoning from first principles rather than by analogy (check out this great interview with Kevin Rose). The same principle applies with product feedback. Users will often make many assumptions with their suggested solutions, the key is to figure out the core problem and think from the ground up how to tackle it within the bounds of your product vision.

For example, with Sqwiggle we rethought video communication in this way - and realised that for our use-case the traditional ‘call and answer’ model that other software uses just didn’t make sense.

In my mind these steps are a great start. It would be great to know your processes for finding out what your users really want - i’d love to know in the comments. You can also read more posts on this topic on Startup Edition!

Photo by Giampaolo Macorig