Just enough research – Why there are no excuses for excluding your customers

By Josh Stehr, Associate

February 3, 2016    7 minutes

John Maeda recently stated in a blog post titled ‘Why design matters more than Moore’ that technology is no longer valuable without design. This got me thinking and I’ve decided that design is no longer valuable without qualitative user research. What’s more there is no excuse for not doing it.

In any design project, it is important to realise that you are making a number of assumptions. This is an unavoidable element of the process, which is why until you involve users, you leave your design to guesswork.

However, since the emergence of the agile process, projects have become more rapid, and businesses and even designers might be forgiven to sacrifice research. But as someone who calls himself a ‘User Experience Designer’, excluding users from the process is simply not an option.

We are often faced with the perception that qualitative research is expensive and time-consuming. It’s often seen as less expensive to fix a mistake post-launch than it is to engage in research before or during the design phase. At WAE, our approach can adapt in situations where speed is of the essence, and I would argue that these added constraints often make user research more effective and more accessible.

A methodology called Guerrilla Research is often described as a fast and cost-effective way to gather insights from users but in many cases it is actually the best approach, even if money and time were no object.

Embrace your inner Guerrilla

Last year, WAE were approached by a UK energy company who wanted to reduce traffic to their call centres. They had identified that one of the most common reasons for contacting the call centre was when people are surprised by the bill they receive (a state of mind, generally called ‘bill shock’). They had prototyped initial screens for a mobile app that responded to three scenarios where bill shock was likely to occur, and we were enlisted to further enhance these visually (essentially — ’make them look pretty’, a brief I’m sure many of you working in the agency or consultancy space have come across before).

In analysing the prototype, we discovered that a big assumption had been made — that the features and information provided would discourage customer calls, yet they hadn’t been validated at all.

So in a project that was to only last a week, myself and a colleague rapidly reinvented the concepts and created paper prototypes (you really can test concepts this early) to evaluate with customers how well they would hold up against its original objective.

We spent an afternoon in a coffee shop putting our new designs in front of people and amongst a breadth of findings we discovered one key insight:

Customers needed to be very confident that their bill was correct and had to know exactly why it was unusually high or low, otherwise they’d still contact the call centre.

This process enabled us to demonstrate that we could encourage people to change their behaviour and reduce call centre traffic with more a direct and text-focused experience.

The sooner you take your concepts ‘into the wild’, the sooner you can iterate towards the next improvement. And one thing is for sure, you’ll always find that some research is better than none.

True Guerrillas don’t always have to ask permission

In another project, WAE were tasked to help an insurance and investment provider to simplify and prioritise the marketing content for their Adviser website and literature. The objective was to propose a new distilled and rationalised structure and hierarchy that made it easier for Advisers to service their clients and of course be persuaded to use this particular provider.

Marketing is all about communicating the value proposition to potential customers (in this case Advisers). But as we often see, what was considered valuable was internally driven and much of the insight about how advisers were consuming this content was fed by out of context quantitative data (analytics) and interviews with the provider’s sales team. It was clear that the insight gained from these sources would only tell half the story. It was vital to understand Advisers’ experiences in order to make a valuable and adviser-driven improvement.

So by adjusting the approach, I rapidly arranged and conducted three phone interviews with Independent Financial Advisers (IFAs) over the course of a day. This led us to some key findings:

IFAs, by law, have to remain impartial and contrary to the provider’s assumptions, an IFA would rarely find themselves simply browsing their website or literature. Usually they would have assessed their client’s needs and have been directed to the particular provider through an independent comparison site. This meant much of the pre-sale marketing was going unnoticed. On the other hand, despite their neutrality, Advisers admitted they were influenced by good customer service and efficient systems.

With this fresh understanding, we produced a new content structure and wireframes that focused on delivering effective content at the time when an Adviser would realistically be consuming it. We prioritised content that would help them to make a decision, and relegated or removed content that didn’t.

Qualitative research is a tool to challenge deep-rooted assumptions that companies don’t think to question, but also to help discover them. Conducting these impromptu interviews with Advisers helped us to discover a vital part of the story and allowed us to pivot quickly without disrupting the project.

Guerrillas are better in the wild

One of the biggest challenges of lab-based research is getting participants to imagine themselves in the context of use. The environment is sterile and users can’t help but feel ‘observed’. This is where guerrilla research really comes into it’s own.

Last year, TfL (Transport for London) approached us with a brief to redesign their services to better meet the needs of cyclists and also to encourage those considering cycling in London that it is a viable and often better travel option.

In this instance we were blessed with time, but with so many features to test, a wide ranging target audience and a need for an iterative approach we went Guerrilla and set ourselves up in Look Mum No Hands (a cafe that also fixes bikes) to put the prototype in front of a range of cyclists and non-cyclists.

With guerrilla research, participants are ‘approached’ rather than recruited, so although you may struggle at first to find willing participants, those that do agree are generally motivated and excited to participate.

The environment and approach is important so that people are relaxed (we deliberately chose a spot where people were likely to be comfortable and also captive). Once you’ve overcome this you’ll find that people are honest, more natural and genuinely willing to discuss in detail.

In this project we found that we could start with a deeper contextual enquiry which not only allowed us to profile participants but also find out realistic scenarios and then tailor the discussion and testing towards them.

Many of the sessions actually overran because people were so engaged and that meant we could spend longer investigating unexpected and interesting paths that you rarely could within the confines of a lab.

The flexibility and unstructured nature of this approach often brings with it rich and detailed insights. It is well-documented that Guerrilla research is mostly suitable for validating ideas rather than creating them, but here are some of the high-level observations that stimulated a lot of design iterations that may never have followed without involving users in this way:

People were confused and outraged at the sight of walking sections in the journey results.

“This is a cycling journey, why am I being told to walk?!”

The unreliability of quiet routes (when quiet routes weren’t actually quiet) could have profound consequences.

“I like to go for bike rides with my two children at the weekend. It would be pretty bad news if the suggested quiet route took me through Elephant & Castle roundabout.”

There was a lack of trust towards estimated cycle journey times, but wide estimates were okay.

“Arrival times are meaningless because as soon as you get stuck at a traffic light it’s wrong. I just want to know the fastest and slowest possible time it could take to cycle so I can plan effectively.”

These are just a small bite of the findings gathered through Guerrilla testing.

All of these early insights enabled us to quickly prioritise which elements of the service to focus purely based upon when pain points were repeatedly observed and the level of negative emotion displayed when participants stumbled across them.

Conclusions

Guerrilla research allows you to quickly gather actionable insights that are ‘just-enough’ to prove or disprove hypotheses, challenge assumptions and design the next iteration of your product or service.

At WAE, our approach is adaptable so that we always include the people that we’re designing for. Whether it is through the rigour of lab-based methods or the quick, flexible and contextual nature of guerrilla methods, we use the best approach to maximise value for our clients and their customers. Contact the team to find out more, we’d love to hear from you — LDN@hellowae.com