Almost every redesign we are asked to review starts the same way. Someone senior says the product feels dated. A competitor ships something sharper. A board deck lands with a slide comparing screenshots. The conclusion arrives before the diagnosis: we need a redesign.
The trouble is that a redesign is a solution, and nobody has yet said what problem it solves. So the team optimises for the only measurable thing available, which is how it looks. Twelve weeks later the product looks current and behaves identically, because nothing about the underlying experience actually changed.
Interface problems and business problems are not the same
An interface problem is a button in the wrong place, an inconsistent form, a table nobody can scan. Real, worth fixing, and rarely what is costing you money.
A business problem sounds different. Customers sign and then never activate. Support handles the same three questions forever. Enterprise deals stall in implementation. Power users churn at month nine. These are experience problems too, but they live across sessions, across roles, and mostly between screens rather than on them. Repainting the screens does not touch them.
If you cannot name the number a redesign is meant to move, you are not funding a redesign. You are funding a repaint.
Start with the metric someone already owns
Before any design work, we ask a blunt question: which number on somebody's dashboard should look different in two quarters, and who gets asked about it when it does not?
The constraint is deliberate. It has to be a metric that already exists and already has an owner. Inventing a new one at kickoff means nobody has history for it, nobody is judged on it, and it will be quietly dropped the moment priorities shift.
Once the metric is fixed, scope arguments resolve themselves. A settings redesign is interesting, but if the metric is activation, it waits. This is the single biggest reason strategy-first engagements finish faster than they look on paper: far less gets built.
Audit the journey, not the screens
With a metric in hand, the audit gets specific. We reconstruct the actual path a customer takes to reach the outcome that metric measures, then instrument where they fall out of it.
- Where does the drop-off cluster, and what happens immediately before it?
- Which steps require information the user cannot possibly have yet?
- Which handoffs cross a role boundary, and what breaks in the gap?
- What do the top support tickets say people were trying to do?
- Which steps exist to satisfy an internal process rather than the customer?
That last question is uncomfortable and usually the most valuable. A surprising share of onboarding friction exists because a team upstream wanted data at signup. That is a process decision wearing an interface costume, and no amount of visual design will fix it.
What the strategy phase actually produces
Two to three weeks of this yields a short, unglamorous document: the named problem, the friction map with a cost attached to each point, a prioritised list of initiatives, and an explicit list of what is not being done and why.
It is not a mood board and it is not a design. It is the argument for what to build, in what order, written so that a finance lead and an engineering lead can both agree with it. Once that exists, the design work goes faster, because the hard decisions have already been made.
When a redesign genuinely is the answer
Sometimes it is. If the visual language actively undermines trust in a category where trust closes deals, appearance is a business problem. If the design debt is so deep that every feature costs double, a system rebuild pays for itself. If accessibility gaps are locking you out of public sector procurement, that is revenue.
The point is not that redesigns are wrong. It is that they should be the conclusion of an argument rather than the opening bid.
This is the thinking behind our UX strategy and product consulting work. Written by the UXperts team. If this is a problem you are living with, tell us about it.