“We’ve never had a UX person before. You’re the first permanent UX person here. What will you do? How will you do it? There’s only one of you.”
Oh shit? How do I bridge the gap between expectation and optimal practise? There are people involved. With people comes expectation and expectation management.
Having been in this new permanent person a few times, I always try to do my job, on the job. What’s she on about? Well, IA/UX – it’s all about the stakeholder and user and so, when I come into a new place (whether it’s had IA/UX or not) I always conduct 1:1 stakeholder/user (in this case the end is both) interviews across all departments to understand the situation and history, and their motivations, expectations and needs.
Having done this a few times, I’ve seen some similarities in what the end users understand to be IA/UX, the desire for formal processes (not always the solution mind!), their ways of working and immediate needs that need plugging and loving.
Here are my ‘in common’ observations and a culmination of 10 recommendations that need to happen to ensure quality IA, UX, product, service and work:
1. Identify production vs strategic projects
2. Put the original client brief on the table and get your experts (aka resources) input
3. Create amazing creative briefs by Planning and Client Services working together
4. Employ a true UCD approach by having pen portraits/personas attached to every brief
5. Identify and come up with a smart project approach and which UX/IA tools are to be employed (they can flex and change) with the PMs, the project team and Client Services
6. Support IA/UX mentoring and doing via the overall project approach and resourcing if you are the only IA/UX person
7. Put the users first (UCD) by usability testing – 1 day repeat corridor testing as a bare minimum – as part and parcel of projects
8. Integrate tracking into projects at the requirements stage (vital at wireframe stage) and create tracking requirements and specification
9. Enforce collaborative and iterative wireframing as part and parcel of project plans
10. Ensure you never start a project without understanding the business objectives and KPIs
What should the IA/UX function look like?
Before we began to dissect the thing, we all needed to have a shared and common understanding of IA/UX, especially since the term gets used interchangeably.
Given how the craft has morphed over time in terms of label, role, skill and understanding and also given how our own industry wrestles over it’s definition, this would have been difficult had I not been such a firm believer (aka stubborn) that everyone is responsible for UX. IA deliverables and activities have become forms of UX deliverables and activities (this of course excludes the ‘generalist’ UX roles, where the job role seems to advertise for a one-man band).
It is always a challenge to address the widespread misconception that if you ‘do IA you produce wireframes’ which is perhaps best dealt with by explaining what a good wireframe is. Explain what a good wireframe tells you, what it shows and how it works as part of a collaborative process. Also explain how wireframing isn’t just knowing how to use Axure.
Wireframes are an end point deliverable. A result of working out the logic whilst keeping the creative concept in tact. Wireframes articulate the design of user behaviour and psychology, functionality and content, timeliness and relevance, validation and error handling, messaging, information and interaction priority, the nature of copy and instruction. Wireframes tell you how we can enable the user to do what we would hope they would so that we may meet our business objectives.
The wireframes are the end-result of an immerse, thorough and ‘what if’ thought and activity process.
Tactical vs Strategic IA/UX
There is a difference between production work where a PM or CS orchestrates ‘resources’ to get a usually routine, job done versus putting the client brief on the table and getting experts to come up with the best solution. This is a hard thing to do – not because it doesn’t make sense business-wise but because it requires a fundamental mind and perceived power shift – and to make it even more challenging, it needs to come from the top-down.
So what I do is work from the bottom-up. No one can argue with the numbers, so demonstrate ROI by a. increasing the efficiency and efficacy of projects and b. by helping meet KPIs through usability and experience design.
To conduct true strategic UCD, I advocate that every creative brief accompanied by a pen portrait (usually built on secondary quantitative research) or persona (built on primary qualitative research).
The little boxes in the corner on the sketches from hereon in, show not only how difficult/hard the suggestions may be to implement but also the importance of implementing them, to the agency.
Different projects and briefs need different types of IA/UX
There isn’t a one-size fits all approach to servicing a project or brief with the necessary IA/UX attention it deserves. Each should be individually assessed in the context of time, budget, requirements and resource. The activities/deliverables chosen from the IA/UX tool box will also form the back bone of the overall project approach.
Usability testing as part of projects
Usability testing is not UAT or user testing. Usability testing looks at the non-functional, behavioural and psychological actions of a user, which is often unexpected whereas the latter looks at the functional behaviour of the system and whether it is behaving as it ought to.
Usability testing is always the first to go, most neglected and wrongly considered luxury on every project I have come across. Ironically, usability testing is the thing that makes or breaks a system – it is the difference between users using, understanding and enjoying the system – a positive UX and the users walking away with a negative experience.
At the very least, conduct corridor tests with a minimum of 5 users who approximate the target user (not always possible) but at least you’ll uncover fundamental usability, interaction, copy and messaging issues.
This doesn’t have to take long or be arduous. Four hours should cover 5 x 1:1 usability testing with findings & recommendations. Surely that’s not bad ROI for the bigger picture?
Create tracking specs to understand and measure success reflective of KPIs
There is no point implementing tracking tags if you’re not going to do anything with them. That’s not tracking. They need to be monitored, reported, evaluated and optimised on in order to be of value to the agency and importantly to the client who has to justify spend and CPA. Oh and please don’t forget to budget for the time and effort it takes to action any optimisation.