The "TTV Spread": A New Metric for Consistency

If you ordered a pizza and it arrived in 30 minutes on Monday, but took 3 hours on Tuesday, you wouldn't trust that pizza place. Even if the "average" delivery time was acceptable (1hr 45m), the unpredictability would kill your loyalty.

Consistency is the foundation of Trust.

In SaaS onboarding, we focus so much on "Speed" that we ignore "Consistency." We celebrate the 5-minute activation, but we ignore the 5-day struggle.

Measuring Consistency: The Spread

At Tivalio, we use a metric called TTV Spread. It is a simple calculation of the distance between your best-case scenario and your worst-case scenario.

Spread = P90 (Slow Users) - P10 (Fast Users)

Scenario A: The Well-Oiled Machine

P10: 5 minutes.
P90: 10 minutes.
Spread: 5 minutes.
Verdict: High Consistency. Your product works the same for almost everyone. The path is clear.

Scenario B: The Chaos Engine

P10: 5 minutes.
P90: 4 days.
Spread: ~4 days.
Verdict: Low Consistency. Some users get lucky (the "Happy Path"), but many get stuck in edge cases, errors, or confusion.

Why High Spread is a "Code Smell" for UX

A high spread usually indicates that your product is fragile. It breaks under specific conditions:

User Knowledge: "It's easy if you know what an API Key is. If not, good luck."
Technical Environment: "It's fast on Chrome, but broken on Safari."
Data State: "It works if you have clean CSV data, but fails if your headers are wrong."

Compressing the Spread

Your goal as a Product Manager should be to Compress the Spread. You do this by raising the floor (fixing the P90), not just raising the ceiling (optimizing the P10).

Identify the Struggle: Go to Tivalio VS Mode and filter for users in the P90 bucket.
Find the Commonality: "They all failed at the 'invite team' step."
Build the Guardrail: Add a "Skip for now" button, or a tooltip, or a sample data mode.