Dana Chisnell recently posted her thoughts on the ineffectiveness of today’s typical research methods when studying the social web. Her perspective on our current methods is spot on and it’s an issue that’s been in the back of my mind as well.
There are four challenges I believe we should address in order to evolve the way we conduct our research:
1. Natural vs artificial environments
We need to get out of the mentality that our users need to come to us and enter our shiny laboratories to encounter our products and provide feedback. Users visit and use our products everyday in their natural environments. They interact with us from their offices, homes and on the go. Speaking to them in these circumstances provides more detail about how our products fit into their lives than we would ever be able to get in a conversation in our labs.
We need to get out of the lab more and take our research into the homes and places where our users are. This can be done with remote recruiting, online meeting tools and other emerging remote usability tools, but it can also be done through home visits, participant diaries and more.
2. Multi-person vs. single person conversations
More and more of the products we develop are used in a social way. Either the user requires the input of someone close to them to use the product or they use the product with others in social settings.
For example, many families watch TV together and share TV sets and DVRs. To build a tool that helps them to manage this entertainment requires that we understand the tool will often not be used in isolation of other household members. The activities a user would perform to record a program or delete a setting effects more than just them, but the household that shares that device as a whole.
In Dana’s example of researching a tool based on financial and retirement planning, it became vary apparent that more than the participant in the room was involved in this activity. Without those other members present, the participant has to pretend thus making the validity of what we learn less reliable and useful.
We should look for ways we can gather data from the people that live in the influence circle of our participant and the area we are studying in addition to the participant themselves.
3. More Active and Generative Participation
Most research studies involve a user’s reaction to a manifestation of the product in question. Few studies actually engage participants as active members of the design process. I’m not lobbying that we eliminate evaluating our creations with users, but rather that we find ways to increase how active our participants are in our studies.
Liz Sanders has been dedicated much of her career to finding ways to increase the participation of our users in the design process. Her site MakeTools.com is full of interesting ideas on how to help participants contribute in a way that makes their latent feelings, dreams and imagination more tangible. This kind of information could really change the direction of our products in ways we never could have imagined with traditional research.
4. Efficiency and Cost Savings
Stepping out of years of tried and true traditional methods can be time-consuming and risky at first. The more qualitative and rich data we collection can really add significant time to our analysis. While we evolve our methods and get closer to our participants and their influence circles we also need to keep an eye on being efficient and cost-effective. Whatever processes we test should be evaluated on their quality output as well as their time to delivery. The best research we do can still be a failure if it can’t get to the products teams in time to absorb it.
I’m really excited about Dana’s thoughts on this subject and helping to rethink how we do user research for the ever evolving social and connected web.