5 Things I Learned about Enterprise UX
What are the biggest challenges a UX professional faces in a large corporate project?
What are the major pitfalls? Which are the best approaches to take?
This article discusses enterprise-level software projects, typically found in companies with several stakeholders representing different departments (i.e. finance, sales, marketing, etc).
This type of (usually legacy) software, suffer by a large quantity of modules and components, along with as much highly complex information as users can cope with.
Company size, along with the sheer magnitude of information that needs processing, make user task efficiency and effectiveness a critical part of the systems’ usability goals. Consequently, decreasing time-on-task for enterprise users saves millions per year for such large companies.
Following, is an outline of some important insights that we have uncovered from research and gained from bitter experience. They have proven vital in many situations, from day-to-day tasks, to major course altering decisions.
#1 — Interview all Stakeholders
This is a task to perform on day one. Spending as much time as possible to listen to everybody involved in the project is pure gold. In most cases, it is of utmost importance to separate decision makers from collaborators, internal clients and the people that just have to be kept in-the-loop (mainly for political reasons).
The hard thing is to identify them all. Be aware that there is a tendency in large corporate projects to have new stakeholders appear in the middle of a sprint, at the worst possible moment. Remember, there is always a decision maker not sitting at the table in the beginning of a project.
When all key stakeholders have been identified, it’s ideal to interview them individually in 1-to-1 sessions. Gathering their unique perspectives and understanding their views, clarifies ambiguities and inconsistencies. Having clarity on the project goals, early on, will help build a good UX strategy. Next step is inviting all stakeholders into a workshop to get them aligned (but this is a topic for another post).
#2 — Test and share everything
Not agreeing on product direction is expected in large-scale projects. Every stakeholder may and will express their own (usually strong) opinion as the project develops. Fighting opinions with opinions is pointless.
Instead of arguing, it is much better to put everything to the test from the beginning and gather data directly from users.
After conducting user research, transparency comes next. Sharing is caring and so is keeping everyone in the team involved. Interviews and their transcriptions should be shared immediately after being processed. There is no reason to keep such information secret.
Every user research iteration provides a new set of knowledge about the users and the domain (e.g. competitors). Regular knowledge-sharing sessions work well to keep everyone in sync. As long as people are properly informed, they feel all the more involved in the UX process. Any level of information depth will do and will be appreciated. That’s the way to faster stakeholder buy-in right there!
“But some secrets are too delicious not to share.”
― Suzanne Collins, Mockingjay
#3 — Keep a ‘meeting minutes’ log
Remember all the stakeholders from the point #1 above? In a large organisation it is very hard to have everyone agree with each other. It is also hard to even keep them on the same page regarding the current process. Unfortunately, in most cases, it is almost impossible to keep everyone in a meeting in-sync with next steps and goals.
Keeping a shared document with written notes during meetings reduces ambiguity among team members and stakeholders. A meeting minutes log typically includes an agenda for the meeting, a list of key points discussed and the next steps to follow after the meeting. It is one of the best communication tools in enterprise projects, especially if requirements tend to change (and they will change eventually).
Having an agenda before each meeting and avoiding meetings without one, saves everyone’s time. Discussion points help to understand the results of the conversation when reviewed in retrospect. Finally, next steps keep everyone in-sync with the agreed strategy, even people that did not attend the meeting.
#4 — Pay attention to IA
Large enterprise systems handle an enormous amount of information. Diverse data is connected or combined in more than a single way and are usually accessible by multiple accounts.
Typically, such entangled information comes in the form of complex tables, large item lists, bottomless forms with infinite input fields and action buttons (e.g. submit, delete) without any kind of error prevention dialogue.
Under these circumstances, working alongside the BAs is ingenious. It takes a while, at first, to recognise the importance of focusing on the taxonomy prior to the layout or navigation of the UI. Creating spreadsheets with buckets of information from the platform, including their variables and parameters, saves a lot of time in designing interface variations later on.
A series of card sorting sessions must be conducted then, in order to validate or invalidate the information architecture with actual users and thus help avoid speculations. Solving this vast information complexity not only simplifies later navigation schemes, but usually makes the 80/20 rule of the job of UX design in such projects.
“Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius — and a lot of courage to move in the opposite direction.”
― Ernst F. Schumacher
#5 — Deliver small changes
Users of the aforementioned systems, struggle with such software a lot. They have no alternative or substitute solutions (apart from Excel of course), so they have to cope with design flaws of enterprise software. Many times, they even stress out and consider themselves as incompetent for not being able to complete successfully certain activities.
Even tiny changes in the platform’s UX can improve the daily workflow of users by removing severe pain-points. To achieve this, it is best to avoid serving brand new wide-scale pieces of software to these users, especially at the beginning of the redesign process. As much as unusable the older system might be, users have got used to it since they have worked with it for years. They will find it very hard to cope with a completely altered workflow all of a sudden.
Think of it this way: Each time Facebook changes its layout, even mildly, users go bananas. That’s Facebook: party pictures, bragging posts and kitten videos! Imagine having the same thing happen in the core software your whole workflow is based on. People will jump out of windows!
In order to reduce resistance to change, it is best to provide small incremental changes that have limited but positive impact to the user workflow. Note the “reduce” aspect of the resistance: it is almost impossible to avoid it completely. However, the discomfort period can be reduced. Needless to say that such improvements have to be based on evidence and be done after a heuristic evaluation or testing with users to discover their major pain-points.
Licking wounds
It is essential to note that during these projects, it is not always possible to follow all of the suggestions mentioned above right from the start. Lacking domain knowledge, prevents foreseeing future surprises of political or managerial nature. So, treat this article as an (exit-)wound list.
Each case and project is different. Trial-and-error will proven an ideal approach here, as long as each trial is kept in small iterations. Observing outcomes and iterating, eventually leads to learning and optimizing. Exactly as the Lean Startup approach suggests, but at the enterprise level.
“A learning experience is one of those things that says, ‘You know that thing you just did? Don’t do that.”
― Douglas Adams, The Salmon of Doubt
I’m sure you will have discovered many more critical issues in your enterprise project. I would very much like to…”compare wounds”, so let me know in the comments below :)
Are you interested to know more about implementing UX in enterprise projects? Interested in UX and Scrum perhaps? Then, contact us at hi@zanshinlabs.io to help you with your specific enterprise product needs or just visit http://zanshinlabs.io/content.html.
