Why This Matters
"Design is valuable" isn't an argument. "Design reduced support tickets by 30%, saving $240K/year" is.
Design Metrics Framework
Business Metrics (outcomes design influences)
- Revenue: Conversion rate, average order value, lifetime value
- Retention: Churn rate, daily/monthly active users, session length
- Efficiency: Task completion time, support ticket volume, onboarding completion
- Growth: Referral rate, NPS, app store rating
Experience Metrics (design quality indicators)
- Usability: Task success rate, error rate, time on task
- Satisfaction: SUS score, CSAT, ease-of-use rating
- Engagement: Feature adoption, return visits, depth of use
- Accessibility: WCAG compliance %, assistive tech compatibility
System Metrics (design operations health)
- Velocity: Time from design to dev, design system adoption rate
- Consistency: Component reuse %, custom override frequency
- Coverage: % of features with design specs, % tested with users
Building a Business Case
The Formula
Cost of the problem (current state)
- Cost of the solution (design investment)
= Net value of design
Example
Current state: 40% onboarding drop-off
Users lost/month: 2,000
Average LTV: $120
Monthly revenue loss: $240,000
Design investment: 2 designers x 6 weeks = ~$50,000
Target: reduce drop-off to 25% (conservative)
Users saved/month: 750
Monthly revenue recovered: $90,000
Payback period: < 1 month
Annual ROI: 2,060%
What to Measure Before and After
- Pick 2-3 metrics maximum
- Measure baseline for 4+ weeks before changes
- Measure impact for 4+ weeks after launch
- Account for seasonality and other variables
- Be honest about attribution (design is rarely the only factor)
Common Arguments and Responses
"Design is subjective" Response: Usability isn't. Task completion rate isn't. Conversion rate isn't. We can measure whether design achieves its goals objectively.
"We can't afford to invest in design right now" Response: Show the cost of NOT investing. Support tickets, churn, lost conversions.
"Can't we just use a template?" Response: Templates solve the obvious 80%. The remaining 20% is where users struggle, leave, and generate support tickets. That 20% is the design work.
"How do we know it was the design that made the difference?" Response: A/B testing isolates design changes. Before/after measurement with controlled variables. We can't claim 100% attribution, but we can demonstrate strong correlation.