How a Series A AI video app grew MRR from $180K to $640K

A Series A AI video generation app that turns photos and prompts into short clips grew MRR from $180K to $640K in 8 months, February to September 2026, while subscribers grew from 10,000 to 32,000. The decision that drove it: get every new user to a finished, watermark-free export before the paywall, with a priority queue for the first render, which raised first-session exports from 31% to 54%.

Written by Emma Park

Prefer a conversation? Book a call

The company and program

ClientA Series A AI video generation app
StageSeries A
IndustryAI creative tools
PillarsCreator marketing + Paid ads + Product growth
Timeline8 months, Feb 1 – Sep 30, 2026: two 90-day sprints and a 60-day third sprint
ChannelsTikTok and YouTube creators, TikTok Spark Ads, Meta partnership ads, onboarding, paywall and lifecycle email
TeamA dedicated Kiko growth lead plus agents for creator research, creative tagging, ad reporting and cohort pulls. The client's engineers built the onboarding and paywall changes Kiko specced
Headline resultMRR from $180K to $640K in 8 months
Measured inProduct analytics, RevenueCat and Stripe, TikTok and Meta ads managers, the app's attribution provider and tracked creator codes

The approach

  • Hypothesis: in AI video the output is the ad. If creators publish clips they actually made, ads reuse those clips, and onboarding gets a new user to their own finished export before the paywall, then creator, paid and organic users all subscribe at higher rates.
  • Comparison: January 2026 was the baseline: TikTok and Meta ads built from brand footage at $77 per subscriber from paid, 1,000 new subscribers, a blended CAC of $46 and 31% of sign-ups exporting a video in their first session. Each later month was read against it, by source.
  • Decision rule: a creator post got 7 days of organic data and went into paid only if its sign-ups exported in the first session at 45% or better. Ad concepts got $500–$1,000 or 7 days and were killed above a $70 cost per subscriber. Onboarding and paywall changes shipped only after a 50/50 test lifted subscriptions per sign-up by 10% or more.

How the three pillars connected

Creator marketing

We booked 12 creators in the first 30 days and 26 over the engagement: 9 YouTube channels and 17 TikTok creators. On YouTube we used AI video testers with 100K–300K subscribers and AI filmmaking teachers. On TikTok we used short-form creators who post before-and-after edits. Every brief asked for one finished clip made in the app, with YouTube's altered-content setting turned on for realistic output. Creators brought 8,200 subscribers at $26 each.

The baseline ads used brand footage and cost $77 per subscriber from paid. We rebuilt the account around creator clips run as TikTok Spark Ads and Meta partnership ads, with 90 days of paid usage negotiated into each creator's base fee. Monthly media rose by about a third from February to September, and paid brought 9,500 subscribers.

Product growth

New users had to choose a model, write a prompt and wait in a shared render queue. We specced a template-first start that matches the ad the user came from ("photo to 5-second clip"), a priority queue for each user's first render, a watermark-free first export, and a paywall shown after that export with the user's remaining credits. First export in the first session rose from 31% to 54%, and trial-to-paid from 18% to 27%.

How they fed each other

The creator clips that converted best were photo-to-video edits of pets and old family photos, so those became the first templates in onboarding and the hooks in paid. At paid volume, analytics showed 36% of new users waiting more than 3 minutes for a first render and leaving; the priority queue fixed that for every channel. Higher conversion then cut paid CAC from $77 to $49 as media scaled. Exported clips shared with the app's end card grew organic subscribers from 400 to 2,930 a month, about half of all new subscribers by September.

What we changed

  1. Weeks 1–2: event audit, a first-value definition (first exported video), the January baseline by source, and a 12-creator shortlist scored on Kiko's six-point rubric, with AI-label rules in every brief.
  2. Weeks 3–6: first creator posts live in week 3. Sixteen ad concepts tested at $500–$1,000 each, and the priority first render into a 50/50 test.
  3. Weeks 7–13: template-first onboarding and the priority queue shipped to all users. Six creator clips passed the rule and ran as Spark Ads and partnership ads.
  4. Weeks 14–26 (sprint 2): the paywall moved after the first export and an annual plan was added. Fourteen more creators booked, and media scaled on the winning clips.
  5. Weeks 27–35 (sprint 3): media reached its highest month, creative was refreshed every 10–14 days from the creator library, and a win-back email for lapsed subscribers launched.
Kiko program report showing the before and after results across the full program. Recreated from the case-study results; account details omitted.
Kiko program report showing the before and after results across the full program. Recreated from the case-study results; account details omitted. View full report

The results

PillarMetricBefore (Jan 2026)After (Sep 2026)Change
Creator marketingNew subscribers a month from creator codes and links01,300+1,300
Paid adsCost per subscriber from paid$77$49−36%
Product growthSign-ups exporting a video in the first session31%54%+23 pts
CombinedMRR$180,000$640,0003.6x

Blended CAC fell from $46 to $18 as organic grew to about half of new subscribers, and monthly subscriber churn fell from 11.2% to 7.8%.

Cohort comparison in PostHog’s dashboard layout, using the same before and after periods as the results table. Recreated from the case-study results; account details omitted.
Cohort comparison in PostHog’s dashboard layout, using the same before and after periods as the results table. Recreated from the case-study results; account details omitted. View full report

What didn't work

  • Desktop-workflow reviews. Two long YouTube reviews brought 310 subscribers at about $50 each, nearly twice the creator average. Both creators worked in a desktop editing workflow, and viewers who installed the mobile app could not repeat what they had seen.
  • A weekly plan. A low-priced weekly plan lifted first purchases but doubled churn among those buyers. We removed it after 5 weeks.
  • Reels-only creators. Creators who posted only on Instagram Reels brought subscribers at about twice the TikTok creator cost, so we moved that budget to TikTok.

What we learned

  • Turn the best-converting creator clip into the first onboarding template, so the first thing a new user makes matches what they just watched.
  • Give every new user a fast first render; queue time was the biggest leak in the funnel.
  • Negotiate 90 days of paid usage into every creator fee up front, so winning clips can run as ads the week they prove out.

Frequently asked questions

What does it cost to grow an AI video app with creators, paid ads and onboarding work together?

Each new subscriber cost $21 in creator fees and media over the 8 months: $56 from paid ads and $26 from creators, with organic bringing about half. Series A programs like this one run $80,000–$100,000 a month in creator fees and media.

How long did it take to see MRR move?

Creator posts went live in week 3. The priority first render shipped to everyone in week 7, with the first creator-clip Spark Ads the same sprint. MRR went from $180,000 to $640,000 over 8 months as media scaled.

Would this work for a seed-stage AI image app with no paid budget yet?

The onboarding half would: a template-first start and a fast first render need engineers, not media. The paid half would not; this program ran paid media at a Series A budget from month 1 and scaled on creator clips with paid usage rights.

Same industry, other angles:

Playbooks and lists:

Plan your next growth sprint

Prefer a conversation? Book a call