UX Tea Break: What does good Discovery research look like?

[@davidsoriano1186]

Best UX advice in the field, thanks for sharing David, you are helping a lot of people, congrats! 💪🏻

[@DavidTravis]

I appreciate that, thank you! 😁

[@hopezhang8278]

Just came from your segmentation video. This is also very helpful. I was confused about whether I was doing research the right way for qualitative data. Now I think I know what to look for. Thank you again! 😁 Notes: Four criteria - Small samples - Behavior-based (as opposed to opinion-based) - One to one - In context

[@hopezhang8278]

One question though, how can I get my teammates/supervisor on board for good quality research instead of just following their practice? I found it hard to convince them because the method you described clearly indicates more budget and time. Also, they are situated in China, and cannot access your videos here, or I would just send them links. I think what I need is: 1. Convincing arguments to support small-sample, behavior-based, one-on-one, contextual research. e.g. How is it better? Who has succeeded by using this method? How does it apply in China? I somehow felt people are building apps and earning money whichever way they conduct research. Some good research leads to less than satisfactory feature directions, while maybe "bad" research helped people identify some good features too. 2. An expert to tell it. 🤷‍♀️ Please tell me if this makes sense to you, and what do you suggest I do. Thank you very much.

[@hopezhang8278]

Never mind, I found your "Stop Selling UX" video. 😂

[@DavidTravis]

Glad it was helpful and that's a good summary of the key points.

[@DavidTravis]

This depends on their UX maturity. They won't be interested in good quality research until they're burnt by poor quality research. If an app meets a user need, even if it's not that good, it will still make money. But once a competitor enters the fray, the need for research to optimise the product becomes more important.

[@olorin818]

Great video. Thank you. However, how would you account for disparities between quantitive and qualitive data ?

[@DavidTravis]

That depends on the disparity. If my quantitative research contradicted my qualitative research (or vice versa), I would look at my research methodology to discover what I've done wrong. The best way to think of it is that you want to triangulate your findings. Your qualitative research shows that x is a problem so you then use quantitative research to discover how much. Or your quantitative research shows that few people stay on a particular page of your web site so you use qualitative research to discover why.

[@Kl4x4]

What about doing discovery research with users who are working in a wfh context or completely remote?

[@DavidTravis]

Nothing is different. The big mistake is to assume you can do your observation remotely. You can't. You need to be in person. That's because someone's wfh context is not just their computer.

[@claudians7817]

Great advice, thank you David! What if our users are in a different continent and there's no budget to go abroad to see them in context?

[@DavidTravis]

In that case, you will just have to accept that your research quality is compromised. And that's a problem, because if your users are from a different culture this is *exactly* the situation where your own preconceptions could badly mislead you.

[@steveperry6047]

@David Travis continuing this discussion, are there any tips or measures that can be put in place to mitigate compromised research findings? I guess a question here might be "why would you want to use poor research data in the first place", however to me it's a scale concerning the data and quality. As you are well aware, it's not always possible to conduct true UX research, where certain factors seem to impact on one or more of the 4 criteria outlined here.

[@ryonlyon]

What about hiring others to do it?

[@harshgorasia2720]

wow! That was a great insight.

[@DavidTravis]

Glad it was useful.

[@gamma_girl]

Good content 👍

[@DavidTravis]

Thank you.

[@xhotdog100]

Average standard deviation

[@lunatico981]

Ok, not for me. I don’t people.

[@DavidTravis]

Remember it's a team sport. You don't need to do everything. Just find a team member who does do people.