Do I DIY
Photo credit: stock_colors/iStock / Getty Images
I read this post from Peter Merholz earlier on and it really resonated with me. I'm a real advocate for designers and product managers doing research. It was what my talk for Industry Conf centred around a few years ago. Absolutely yes, designers and product managers should be talking to customers, users and the real people who use their services to build empathy and understanding for their needs. It's fundamental stuff. It's also just the way it is in a lot of organisations.
A few years ago I did a quick survey for an article I wrote for Net Magazine. 226 people responded. I asked ‘who is responsible for conducting audience/user research in your company?'
51% said designers, 42% said the UX team, 35% project or studio manager and 22% developers.
Only 16% of the people who responded said that a researcher or the research team typically did the user/audience research in their company.
This rises slightly to 19% for those people who said they worked in-house.
So having a research team or a dedicated researcher in a team sounds like a bit of a luxury. You should definitely be nice to us!
Does this mean that you can do away with researchers? Absolutely no. As Peter highlights, this is particularly true for in-house teams but I can also think back to times at the design studio I ran - Mark Boulton Design - when our clients needed my research skills and not those of one of our designers.
With the rise of movements in UX such as Agile and Lean and the Big Data industry, research is happening in new ways and traditional Market Research budgets are being squeezed. Access to oodles of data and the use of guerrilla user research methods means traditional Market research budgets are being slashed and researchers of all flavours can literally be squeezed out of a job in some organisations. Russell Parsons cautions against this in a recent article for Marketing Week:
"Scrimping means marketing will be less the voice of the customer more the voice, face and whim of the people sitting next to you in the office."
So when should you consult a research expert rather than do it yourself? I think it depends on few things - where you are on the research funnel, the desired business outcomes and the proposed methodology. Let me remind you of the research funnel I wrote about a few years ago.
Research Funnel
The funnel categories describe research as something that can and should happen at every stage of a product/project lifecycle. They describe how close you are to the problem or solution. At the top of the funnel you are furthest from the problem and at the bottom you are closest to the problem.
It also describes how broad the data collection should be - how wide or deep to go. The funnel isn't one way and research doesn't necessarily need to move through the stages in order. For example, insights from the bottom of the funnel can be dripped back into the top or middle of the funnel to inspire creativity, inform strategy or provide direction and tactics.
So how can this inform when to consult with an expert? Here are my tips:
Operational research
Often a place where you can ask the help of an expert to set up tools, analytics or dashboards and then let them run themselves. For example, A/B tests, NPS surveys, analytics. The key here is interpretation and analysis. As a product manager or designer you may get pretty good at looking at results and interpreting them but it's good practice to talk through results with the broader team to avoid your own biases. Likewise you may need a data analysis expert to help you crunch all that data you've collected.
Tip: Make sure you know some quant researchers or data analysts. They are usually pretty smart high achieving folks and like to be asked for their input and ideas (not just treated as a pair of hands). They can sometimes be bribed with nice snacks. Nuts and dark chocolate are good.
Tactical research
Often a place where user research comes into it's own and can be done by multiple people in the team. Semi structured interviews and usability testing are king here. Of course, most people can conduct interviews, set up analytics funnels and given some practice make a pretty good job of it. The risks of misinterpretation and mistakes are often lower if you make changes based on what you learn. That's not to say that a small UX change doesn't affect the bottom line but you are generally not talking about a whole new business model or offer with this kind of research. Don't forget that everyone has inherent bias however and the closer you are to the solution the less likely you will be to want to throw it out. If you can, research someone else's work or at least collaborate with a colleague to try to avoid 'proving a hypothesis' - your job is to understand and investigate, not prove your solution is right.
Tip: Make sure research is a team sport at this stage of the funnel. You should definitely invest in a wide range of snacks, don't just rely on doughnuts. Everyone needs to work together to get the possible outcome here so you need to cover all the bases.
Strategic research
The key place where you should probably call in an expert. Often you're fairly far away from the problem/solution and often this is big picture stuff like customer segmentation or market sizing. You will likely need more complex multi-stage projects using mixed methodology. You can't throw semi-structured interviews at every research problem and you may need some specialist quant analysis combined with qual. Every successful organisation will have an experienced, senior researcher who you should really talk to before undertaking any strategic research yourself.
Tip: Get the kettle on, you need to get friendly with your researchers for this stage of the funnel (they may have market or user experience in their job title). This is a long-term relationship as you will need to involve them all along the way and they will need a lot of tea and cake to make a good job of things.
Exploratory research
An interesting one. Again this is a great time to use a research expert. It's typically a fairly broad kind of research but it can also go deep. Generally you need someone who is dedicated to the project who can spend the time exploring random avenues of exploration and isn't too focused on getting something to market or designing a specific feature. This takes someone whose job it is to work on research or at least someone who has been given the remit and time to undertake this kind of work. In my experience, product managers are simply too busy or focused on short-term business outcomes to lead this kind of work.
Tip: It's time to go and find your R&D researchers or design researchers. Get to know them and understand the conditions they need to do their best work. They often like collaborating with designers or engineers but they are equally happy noodling through problems in isolation. This might involve late nights, alcoholic beverages and Haribo (the fizzy ones).
Getting to know the strengths of your research team, where they might live in the organisation, their strengths and skills is key. This will help you to understand when to call upon them. What are you waiting for? Go and get the kettle on!
