How to do research on a company?

How can a company conduct user research tests that actually attract and reflect the REAL target audience?

• I noticed there are some trends out there when it comes to user research and the way companies try to attract people to participate to their user tests or research about personas - one being the offering of gift cards or other material incentives for the time and effort put into participating to these tests. But I find this quite limiting, and personally I wouldn't give an hour of my time for a $10 gift card at Starbucks or Amazon (especially if the company is just a start up I never heard of or if what they will do with the data they collect from me is quite uncertain). And I bet there are a lot of people out there like me, for whom time or experiences are the most valuable resource. So how can you attract people that are in your target audience according to each of society's layers to participate? What would you offer? How would you make this experience worth their while? And how would you assure them of the privacy of their data in an age when every corporation tries to collect and store your information and nobody is trusting anymore?

• Answer:

  There are a lot of potential answers to your questions but it all depends on the nature of the company, who is doing the UX research and what the product is. I'll try talk about some of the common situations I've been in but I mainly work in larger corporations as an embedded UX professional and my bias will come from that point of view. Assuming the company is solvent and has a decent budget for the project I'd wonder why the company is only willing to offer a $10 gift card as recompense for a users time. It sounds like they, perhaps, don't really care too much about the research and it's outcomes. Everywhere I've done user research, the company is more than happy to pay $200 - $300 per user - usually around $100 of that goes to a professional market research recruiter to find users that match our user profile (which leads to another question I'll ask further down) the rest is offered as incentive to the user to come in. We try to offer an amount at least equivalent to what they would earn if they just stayed at work. We find usually that we get a strike rate of between 60 - 80% in terms of matching to our profile with recruiters. Could we do better ourselves? Sure, but the time it would take to recruit and schedule is often not worth the cost saving of the $100 per participant - think how much time it takes to recruit one user (plan on at least 10 phone calls or emails - then you need to schedule them and confirm their attendance the day prior etc). One issue I hit consistently is that the organisation actually doesn't have a good idea of who they're users are - or they say "everybody". Both cases tend to cause issues and often you can save time and money by doing a little exploratory research/ethnography into who your users actually are. So when you emphasize REAL, I'm left wondering about what you mean by that and how you determine one user's real-ness over another? Are you taking the company's user attributes as gospel or have your done your own verification? Also, by verification, I mean actually testing who the users are and not just making judgement calls, it's very easy to fall into the trap of mis-identifying your user base. There are other approaches if you are stretched financially and need to do good work cheaply - but they take more time and effort. If you have an existing product then work out how you can contact people that currently use your product - it can be as simple as asking "Are you interested in helping us improve this product?" and then collecting their email address within the product itself. The people who can be bothered to fill out the form and give you their details are interested fans of your product, willing to go out of their way to help out. Once you have their email address, send a follow up email to get extra details and get their agreement on how data from the research will be used. We tend to send a Non-Disclosure Agreement (NDA) for the person to digitally sign (we use docusign for this) that let's the user know we'll be recording the video and audio of the session but that we guarantee that it wont be made public and their identity is protected (in Australia we have pretty good privacy legislation so we don't need to be too specific in our NDA). The NDA also states that the participant wont reveal the contents of the research - this is to cover us and the company. We also send a link to a more detailed profile form where we get some more relevant details about the user. Then use an online scheduling tool like Doodle or NeedToMeet to schedule the user testing. For the most part, the participants you get via this method are willing to help for no charge - they like your product enough that they are happy to go out of their way to make it better. It often means you have to be careful to look out for those ideas and issues that are their personal bug-bear or desire and ensure it doesn't taint your data (ie: don't discount it but also don't leave it in without tagging it as that participants personal bias). There are other less formal approaches too. Check out the book "Running Lean" by Ash Maurya - he has a great example of getting the right people to give him input on his product... it was how he made the book. It's short, informative and easy to read - plus their's a digital version if you don't want to wait for it to be delivered. A couple of things that you mentioned that I may not have answered directly: You talk about all of societies layers, with regard to your target audience. I have no idea what you mean by that, sorry. If you have a clear target user profile then you target that profile with your recruiting method. If you have more than a handful of those then I would say you may not have a clear idea of who your target audience actually is. Do that research first - find out who your audience is then you can recruit them. You expressed concerns about privacy of data - this has never been a huge issue for me, it could be due to the fact that we use recruiters and the participants have signed on with the recruiter for the express purpose of telling us stuff in exchange for money. Part of it also is that as a professional UXer I take a lot of pains to make my research be as ethical as possible. There's a lot of little things you need to do and account for - I have already been verbose enough so I wont go into too much detail. Keep the names of your participant and the company they work for separate from the notes and video/audio you take. Don't start to record during the session till you have all the personal stuff out of the way. I also make it very clear to my employers that they can have the results of my research but that I will not pass on the specifics of the people I speak to. I use generalities when I refer to them (eg: One senior manager from a finance company I spoke to.... or participant number 3 said... or 6 of the 8 participants indicated that they....). You need to be firm about this and you need to assure your participant of the lengths you will go to ahead of you starting the research session. I've really only scratched the surface here - there's a lot of options and ways of going about recruiting the right user, almost all of it, like I said at the start, is driven by who you are, who your working for and who you are interested in targeting for your research.