Getting user experience right needs a holistic approach to customer research so that the right decisions for the design of a product or an interface to be made. In fact decisions about all interactions with a brand need a complete and realistic view of natural behaviours, preferences and day to day activities. Contextual Inquiry is a technique of studying users in their own natural environment, to get under their skin, be it in their place of work or home life, understanding what makes them tick. This means being there for scrum meetings, mothers meetings, understanding details of organisational or social structures and hierarchies, analysing all interaction with customers and colleagues, coffee table interactions, protocols, any systems that have to be used and even looking at the pile of paperwork collected in a day.
The point of Contextual Inquiry is to unearth details and intricacies of work or play discovering parameters, criteria, features, or process flows for design or redesign that you couldn't dream up in a studio without the right level of prompting. There are a million different parameters to take into account and Contextual Inquiry can aid with the collection, collation and analysis of this information.
Before setting off to conduct contextual research, get a team of people together and brainstorm. The aim of this task will be to create an affinity map that can be used to create potential interview questions and most importantly to develop a focus statement to keep the project and the contextual inquiry focused.
An affinity map is built over a number of stages:
Encourage the team to throw in as many ideas as possible and stress that there are no bad ideas. What kind of ideas? It could be anything: questions, assumptions, fears, possible errors in the current system, forward looking views etc. If your team finds it difficult to not analyse as they go along, adopt the 'green hat' from De Bono's 'Six Thinking Hats' theory, a great method for eliminating commentary and allowing only pure, organic ideas to be generated and shared.
Record all of the ideas on sticky notes and then start arranging them and grouping them into themes, remembering that they don't have to be definitive yet. Start looking for trends, patterns and evolving relationships taking out the sticky notes that don't seem to be relevant anymore.
You have now identified affinity groups, and should draw the relationships between these affinity groups to indicate hierarchy and inter-connectivity.
An affinity map could look something like this:
Affinity mapping is a great way to generate initial insights as a starting point for research, and to make associations from seemingly unassociated data to reveal a hypothesis at the start of the project. Having said that, it can also be argued that affinity maps are better used after the inquiry has been conducted, which is also a good idea provided you have the time to do so. Re-producing an affinity map after the contextual inquiry research can help to analyse the findings by showing the range of problems discovered and any patterns that have been uncovered.
Contextual Inquiry can be challenging to conduct so here are 3 key areas that can often go wrong if not considered properly in advance.
The time you decide to spend in the usersï¿½ context, be it their place of work or leisure, should be carefully decided. While you may be tempted to be there with the users for a short, specific amount of time itï¿½s a good idea to spend as much time as possible. Why? Well letï¿½s say youï¿½ve decided to spend 2 hours with customer care executives and your task is to study how users interact with the portal. From those 2 hours deduct the hello time, explaining why you are there, the coffee time and the settling down time. Also, deduct the time you would spend getting to grips with what is happening around you. Then deduct the time that is needed for the participant to get used to someone being around and inquiring about their day to day tasks. How much are you left with? Even though Contextual Inquiry may seem like a lengthy and time consuming, it is very insightful method and actually quite an enjoyable experience.
As with all user research, the participants should be representatives of the deemed user groups. To do this properly, make sure it is discussed in depth using the affinity maps as further guidance. You may think you know which department needs to be researched to improve customer satisfaction or what profile customer you need to research but a well worked-through affinity map could highlight a few other user or customer groups that hadn't been considered. Recruitment needs to be specific and well documented to ensure results can be analysed properly according to the profiles of the participants.
A fully structured plan could be restrictive and concentrate solely on getting answers to the questions written down. The important thing is to go with the users flow, stay focussed on the tasks that the user is doing, and discuss any queries as you go along (without interrupting natural behaviour). Another problem area when conducting Contextual Inquiry is trying to interpret as you listen. Finish the entire conversation and then spend time analysing and coming up with implications for design afterwards. This can be difficult as it is normal human behaviour to process and analyse whilst taking notes. A good tactic is to separate facts from any assumptions or interpretations in your notes by using a different coloured pen or highlighter, a different kind of bullet point, or simply use quotation marks to indicate the direct words used by your participant. When you get back to your notes you know exactly what the participant said and you can interpret and analyse it accurately.
After you've conducted the inquiry and have your notes in place it's time to start analysing. There are 6 main Contextual Inquiry models, which should be used in different combinations depending on the objectives and requirements of the research project. They all sort the information in different manners, producing different flow diagrams that help to build a truly holistic view. It isn't always clear which ones suit best but try building them and the most useful models will become apparent.
A flow model could look something like this showing different roles involved in an organisation and the detailed connections between them:
A basic sequences model could look like this:
A more complex sequence flow diagram of a primary task breaking down the intent, triggers and steps involved could look like this:
A physical model can be as simple as the drawing below:
Creating the models can be very time consuming. The process can be more efficient by making sure the key aspects from the 6 Contextual Inquiry models are included in the affinity maps at the beginning of the project. This ensures that the research or the test plan delivers results already focused on contextual inquiry criteria, making it easier to sort it into models for analysis.
Most systems, products, and customer interface designs are complex. Getting it right will make a huge difference to the customer experience, brand reputation, and ultimately to a company's success. Just as Don Normans book on Living with complexity discusses, complexity can be 'tamed' by understanding the total system, and the best solutions are those that are designed with all aspects fitting perfectly together.