Do you do User Research?
Artem Galenko
10 replies
If yes, please, share your experience and methods you use!
Replies
Kevin@keeev
Great question!
Coming from product design / web development to building my own startup at the moment things changed quite a bit.
Previously I didn't have to start a product from scratch, thus some of the methods you see in SPRINT (the Book) and stuff like that just weren't necessary because we already knew who our customer was. To be honest, I always felt bad when I saw people tweeting pictures and writing medium posts with office walls full of post-its. :D
At the moment while building my own thing from scratch I did it the old fashioned way. Creating personas, assumptions and tried to validate them or at least checking out if they were correct.
Most of the current user research happens in 1:1 conversations with my target audience either via Hangouts or in the co-working space I joined recently (which was one of the best ideas). I'm lucky to have my target audience around me constantly.
One thing that helped me a lot regarding messaging on the website, the product or the target audience was creating a Business Model Canvas. Even though most of the startup hipsters don't seem to be a fan of the traditional way of doing things, but this helped me a lot to manifest my vision down on paper.
Guess what will also help is a value proposition canvas where you have to map the fears, goals and pains of your users which also helps to understand what you are building and most importantly - why you are building it.
Share
I go pretty old school and talk to the people who might buy the product I have in mind. I try to understand the problem, pain point, and what solutions they are already investing in. I usually take notes during the conversations and then collate the notes to try and find common themes and threads.
Yes!
User/Customer research is essential for every startup/product.
Here's how I approach it:
- I create my user persona. Essentially during this session I try to understand for who I'm building my product? for example 30yo Joe with job in finance not married likes to go fishing with friends. I use this tool https://www.hubspot.com/make-my-... it helps to gather everything in one place
- I try to find places that my Joe is hanging out. So if he likes fishing I search fishing groups in Facebook, lists in twitter, maybe I'll find some fishing stores in my target location.
- Connect with my Joe. The Joe is out there I know where to find him so what i do i make a small post in a group that i found and introduce myself maybe hang a poster in that fishing store maybe connect with fishing store owner. People usually are ready to talk about their passions!
- Face to face talk or survey. There is nothing that can be more valuable then talking to you Joe and understanding his needs, so usually someone will reach out from the facebook post or if I meet someone I prepare my set of questions and go after the original content.
A comprehensive guide on how to find your "Joe" :)
Atlassian
@angrigoryan__ thanks a lot)
I've seen some CTO/CEO that thought they know better user needs then users themselves. I suppose, their success story ending is obvious. 🤕
Hi Artem - I suspect the answer from most people will be "yes" in some form or another. To clarify, are you talking about market research (surveys, demographic research, online monitoring, outreach emails) or user/product usage research (e.g. in house user testing sessions, analytics to get insight into user behaviour, community beta programs)?
@unrealartemg excellent - that's good, because I know much less about market research!
For a long time I was part of a digital agency that produced children's games and apps. That required extensive user and product testing. Some of the ways we achieved that:
Analytics in an app or site is a big one: this gives you qualitative, actionable data, but you have to be very careful you don't misinterpret it. Broadly start by defining a funnel that your users will pass through - that might be landing, clicking through to a sign up page, registering, using X feature Y times, and perhaps sharing your product further. Generally you'll want to be measuring and quantifying conversion (what percentage of users that visit sign up, or buy your product) and engagement (do users keep coming back, how long do they use it for, what features do they use). Try to optimize for increasing those metrics, but not in such a short-sighted way that it will ruin your user's UX. Analytics will also let you know if there are features or product areas that aren't being used/need to be promoted more/just aren't useful, etc. Google Analytics works just fine for a large number of cases - Mixpanel used to be an excellent choice if you have more complicated reporting needs, but I haven't used it for a few years now
Qualitative, interactive feedback is just as important too. Find ways to actually interact with your users. You can offer early adopters extra benefits or reward users with some form of community prestige for joining up to a feedback program (e.g. an achievement badge or flair in community forums). When done honestly, lots of people appreciate the chance to give input into your product's direction. I'd be wary of popup survey requests - they can be useful if they're quick to complete and they don't antagonize your users. If you really have a good case for doing them, make sure they're at a natural point in your application's flow (e.g. Skype's call quality rating).
If you can get actual people onsite so you can watch them using your product, that's even better. It can be great to sit by someone and watch them struggle with a piece of UI you never knew was causing problems. Or hear their stories, and "it would be great if your product would let me do this". If you don't have the ability to do that in house, you should be able to pay a local research agency that will be able to source a number of users for you and setup a user testing session.
As an alternative, there's something like https://lookback.io/ if you're shipping an app - I never ended up integrating with it myself, but it records both the screen and the front-facing camera of a user's device. Consent is obviously explicitly required, but that could be a cost effective way of interactively reaching a few more real people than you could get via in-person user testing.
One final point to consider - take on board what users say, but try to look beyond the surface of their words. They're not product managers, so they'll usually be unable to articulate deeper design changes that might be exactly what they need - there will be gold there, but you'll need to dig for it.
Hope that helps!
Atlassian
Atlassian