When management asks you to track another quantitative metric on your dashboard
When management asks to see that metrics dashboard on mobile
When your product was last redesigned in the 90s and the designers roll out the new style guide
When your designer gets tired of rectangles and starts trying Something New
When the designer realizes that the front-end framework allows for Interesting typography
When you have to fix the deep system problems in the UI layer
When you have too many "single sources of truth" and instead of fixing the system your manager issues you a second monitor so you can look at them side by side
When your org is undergoing Agile transformation and management asks for a dashboard showing every stage of the SAFe process
When you're told that you don't need to redesign the entire UI for a new feature, just add another button to the menu
When your org releases the "more intuitive" redesigned product and the users try to get some work done
When the manager watched too many Apple keynotes and asks to make the user interaction more tactile
When the super-modern front end requires integration with the existing back-end, and you open the legacy code base
• • •
Missing some Tweet in this thread? You can try to
force a refresh
"Don't worry about edge cases or errors, this is an MVP" is a very dangerous line of thought that I usually hear from people used to designing for stakeholder demos, rather than production releases.
A demo for a stakeholder, especially a "user proxy", is a very different path from normal use. Any scenario has to contrive a path through every single one of the stakeholder's pet features, so usually it's a checklist approach instead. "We added XYZ like you asked, see?"
In reality, the "as a user I want to use X feature" user story doesn't cut the mustard - and without designing for plausible scenarios, the product ends up being entirely unusable.
If you're not designing for recovery from exceptions, you're not designing a viable product.
The business: hey we need you to draw some wireframes for this engagement
Me: Do we have a scope for the problem? How are the user's existing tools failing them? Are we resourced to answer these questions?
The business: nah it's ok they can be low fidelity wireframes
Design artifacts are not the deliverable. They just document the design decisions, and design decisions must be rooted in user research.
If you can't draw a direct line from the artifact to how the decisions it documents will be informed, you won't produce anything of value.
Before someone sets pencil to paper, it's really easy to think you have something while remaining pleasantly vague. Only when you need to render the artifact do you realize that there's no there there. It's up to the designer to say "hang on, I can't do anything with this."
Product ownership is an interesting challenge - you want to limit WIP and avoid context switching, but you can't just laser focus on building and neglect caretaking.
This is why trying to develop many new things in parallel doesn't work, and why single threaded leadership is so effective. Without consciously attending to all the dimensions of the product, you'll build an unstable tower that is constantly collapsing.
This model also shows why focusing on outcomes is more important than outputs. The second approach looks tempting from a feature factory perspective, but if you have visibility into outcomes, the work in the first approach becomes just as meaningful.
One of the most important Enterprise design tools you can learn is PowerPoint
This is not (just) my usual spiel about most visual tools being merely documentation - if you don't know how to make your decks look good, it'll be rough to get your ideas across.
PowerPoint can do almost everything InDesign can do. Take advantage of it.