UX ROI vs. Intangible Benefits: What You Can Claim

UX ROI compares the financial benefit attributable to a design change with its cost. Other benefits, such as reduced confusion or greater confidence, can be valuable without a defensible dollar estimate. Measure them directly rather than forcing every outcome into a revenue claim.

Key Takeaways

• An improved UX metric is not automatically financial return.
• Use incremental contribution or avoided cost, not gross revenue alone.
• State the baseline, time period and attribution assumptions.
• Keep valuable but unpriced outcomes visible.

Separate three kinds of evidence

Retention and loyalty are not inherently “intangible.” They can be measured, though attributing changes to design and assigning financial value may be difficult. Time horizon alone does not distinguish ROI from other benefits.

NN/g's guidance on calculating design ROI is a useful starting point. The UX measurement guide helps choose a task-level signal before making a business claim.

Work through a transparent example

Assume a redesign costs $8,000. Over six months, comparable support records show 400 fewer contacts about the changed flow. Assume each contact costs $12 to handle, and that $1,000 of additional maintenance is needed.

This is an illustrative calculation, not a client result:

• Avoided handling cost: 400 × $12 = $4,800.
• Total intervention cost: $8,000 + $1,000 = $9,000.
• ROI: ($4,800 − $9,000) ÷ $9,000 = −46.7% over that period.

The UX may be better while the six-month financial return remains negative. That is useful information, not a failure to tell the right story.

The calculation also assumes the contacts were genuinely avoided because of the change and that reduced handling creates a realizable benefit. Spare staff capacity is not necessarily a cash saving. State whether the benefit is time available for other work, reduced outsourcing or an actual budget reduction.

Check what else changed

Record changes in customer volume, acquisition mix, pricing, product capabilities and support classification. A before/after comparison without these checks can assign credit to the wrong intervention.

Where feasible, a well-designed concurrent experiment can help estimate the effect. Where it is not feasible, present the result as observational and explain the limitations. A/B test planning can support sample planning; it cannot repair missing measurement or biased assignment.

Track benefits you cannot price responsibly

Confidence, clarity and perceived control can matter even when no reliable financial link exists. Use defined interview questions, observed task behavior or a repeatable survey measure. Keep the wording and population consistent, and record what changed in the product.

For example: “Participants could locate the cancellation policy without assistance” is a specific observation. “Trust increased” needs a defined measure. “Revenue increased because trust improved” requires a further evidentiary link.

Trustworthy-design research provides a framework for examining disclosure and credibility, not a conversion multiplier to insert into a spreadsheet.

Present a decision, not a decorative return figure

Report project cost, observed outcomes, plausible financial effects, uncertainties and the next decision. If the intervention removes an accessibility barrier, describe that barrier and the verified repair instead of inventing a commercial percentage.

The earlier version's willingness-to-pay and churn percentages could not be traced to a usable study and have been removed. A smaller claim with a visible calculation is more useful than a large return nobody can reproduce.