Adscene.ai — Jorge Vázquez Moreno

01 — Context

A product with strong demand and a pricing model that did not capture its value

Adscene.ai lets performance marketing teams generate UGC-style video ads at scale without coordinating with creators. The product worked — teams were using it to produce dozens of ad variations per week.

But revenue was not growing in line with usage. The pricing model was flat subscription, which meant heavy users and light users paid the same, and we had no way to capture the value we were delivering to our best customers.

02 — Problem

Flat pricing was subsidizing light users and capping revenue from heavy ones

When I mapped usage data against revenue per account, the pattern was clear: our top 20% of users by video output were generating 60% of the measurable value but paying the same as users producing a fraction of that volume.

Churn analysis also showed that light users churned faster — they were paying for a capability they were not using enough to justify. The pricing structure was simultaneously overcharging the wrong segment and undercharging the right one.

◎

The core mismatch

Price was decoupled from value delivered. A team producing 80 videos a month paid the same as one producing 5. That is not a sustainable model — it caps revenue from your best customers and subsidizes your worst-fit ones.

◈

Churn signal

Light users were churning at 2× the rate of heavy users. The flat price was not causing them to engage more — it was just delaying an inevitable cancellation. Lower entry pricing with a clear upgrade path would have been a better fit for that segment.

03 — Key decisions

What I chose to build, and why

01

Usage-based hybrid model I moved from flat subscription to a usage-based hybrid model: a lower base subscription plus per-video credits above a monthly threshold. This aligned price to value delivered and created a natural upsell path.

02

Three tiers anchored to real usage patterns I introduced three tiers — Starter, Growth, and Scale — with credit bundles at each level. The tier names and thresholds were designed around real usage patterns from our existing cohorts, not arbitrary round numbers.

03

60-day grandfathering window Grandfathering existing users for 60 days reduced churn risk during the transition and gave us a clean before/after data window to measure the pricing impact accurately.

04

In-product usage dashboard I added a usage dashboard inside the product so users could see their credit consumption in real time. Transparency reduced support tickets and increased willingness to upgrade when approaching limits.

04 — Results

Compounding month over month from a single structural change

+24%

MoM revenue growth sustained over the three months following the pricing restructure

60-day

Grandfathering window that protected retention during the pricing transition

Top 20%

Of users by volume moved to Growth or Scale tier within 45 days of launch

↓ Tickets

Billing-related support tickets reduced after adding the in-product usage dashboard

05 — Reflections

What I would do differently

01

Run pricing discovery interviews before the restructure We had strong usage data but limited qualitative signal on willingness to pay at different thresholds. The tiers we chose worked, but they were informed guesses more than they should have been. Interviews first would have given us sharper thresholds.

02

Set more aggressive tier thresholds at launch The credit bundle sizes were too conservative. We left upsell revenue on the table by setting the Growth threshold too high. A lower threshold with a more aggressive Scale tier would have accelerated the upgrade path from day one.

A pricing restructure that compounded revenue 24% month over month