When should an app start Meta ads, and what does it cost?

By gr0w Research · Updated · From 185 founder interviews · 6 min read

Short answer

In an analysis of 185 founder interviews, 44 used Meta ads, and most started after content, creators or influencers had shown what worked. Only 3 of the 38 who said how they started began with paid ads. Founders began at about $20–$100 a day, reported $0.39–$2.50 per install (Prayer Screen, Pep AI) and about $10 per trial (Prayer Lock), and raised budgets in steps while ROAS stayed above 1.

44 of 185
case studies where Meta ads drove growth at some stage (median monthly revenue $55K among the 34 that gave one)
3 of 38
founders who said how they started and began with paid ads; 15 began by posting content themselves
24 of 44
Meta-ads case studies that also used organic TikTok (UGC creators: 18 of 44)
$0.39–$2.50
cost per install founders reported on Meta (Prayer Screen's best US day vs. Pep AI's average)

When should an app start Meta ads?

Most founders started Meta ads after they had a piece of organic or creator content that already worked. Of the 38 Meta-ads founders who said how they got their first users, 15 posted content themselves, 5 did influencer deals, 4 had an existing audience, and only 3 began with paid ads.

The pattern is the same across several of them:

  • Prayer Screen ran UGC on 10–12 TikTok accounts, then repurposed the one winning video, which drove about 70% of installs, into Meta ads [1].
  • Pep AI signed nearly every peptide creator first, then put the winners on Meta [2].
  • Canary posted a lot, found an organic winner, made variations and boosted the winner with Meta spend [3].
  • Puff Count founder Steven said you should find what goes viral organically so you can spend on ads with more confidence [4].

The exceptions had a reason. Boost App Social started at $100 a day because the team already knew its Instagram-services audience [5]. Cal Buddy ran ads in Israel, where he said CPMs are far lower than in the US [6].

How much do founders spend on Meta ads?

Founders usually tested at $20–$100 a day, and budgets after that ranged widely. The numbers are self-reported and not like-for-like.

  • Test budgets: Prayer Lock's founder started Meta at $20 a day [7]. SideShift's Nick Lawton suggested about $100 a day across roughly three videos for 14 days [8]. Gravl started under $50 a day in South America [9].
  • Mid-stage: Pep AI was spending about $400 a day in total, plus about $150 on Apple Search Ads [2]. Cal Buddy went from $1K in month one to about $20K a month in four months [6].
  • At scale: Boost App Social went from $100 to $500 to $1K to $10K a day, and spent $1.5–1.8M on Meta over about 2.5 years [5]. Letterly said it spends about $200K a month on advertising against $250K a month in revenue [10]. Gravl said ad spend is about a third of revenue [9].

Sizzle AI's growth lead suggested a $5K starting budget spread across many creatives instead of a few [11].

What cost per install or trial do founders get on Meta?

Reported Meta costs ranged from $0.30 to $2.50 per install and about $10 per trial, and the low figures came from creatives already proven organically. The library has few cost-per-trial figures, so treat these as examples, not benchmarks.

  • Prayer Screen: $0.39 per install on the best US day and $0.70–$0.80 on average. The founder said most people are happy at about $1.50 [1]. A related portfolio founder cited $0.30–$0.50 on a Pray Screen launch [12].
  • Pep AI: about $2.50 per download, with a rule of thumb that under $5 is usually fine [2].
  • Prayer Lock: about $10 per trial started on Meta at $20 a day, against $4 on TikTok, with a $50 yearly plan [7].
  • AbMax: 2.2–2.5x ROAS on Meta and 4–5x on TikTok with the same creatives [13].

App acquirer Steve Young said he would not consider Meta below roughly $3 revenue per install or a $10–15 cost per trial. One of his apps earned about $1 per install [14]. Puff Count's founder found TikTok traffic cheaper but lower quality, and Meta pricier but better [4].

Which creatives work best on Meta: UGC, Spark ads or statics?

UGC-style video that first proved itself organically was the most common winner, but founders also warned that most creatives fail. Spark ads are a TikTok format. For Meta, the founders' equivalent was reposting the creator's video as a partnership ad.

  • Organic winners as ads: Snag treated any UGC video over 50,000 views as a Meta candidate [15].
  • Spark and partnership ads: SideShift's Drew Levin said they generally do better than whitelisted ads on a branded account [16]. Lawton said Spark ads give cheap CPMs but no download button, while Meta gives attributed installs [8].
  • Volume: Boost App Social said 1 of 85 videos got the spend, and about $800K of its spend was wasted finding them [5]. Gravl paid creators as little as $50 a piece and used AI video tools [9].
  • Statics: Pep AI said AI-generated app pictures performed well [2]. Marcus Burke suggested starting with video but not skipping statics, because Facebook Feed suits static content [17].
  • Failures: Pep AI's scripted copies of other apps' ads lost money [2].

What are the rules for scaling Meta ad spend?

Founders scaled in steps and stopped when returns turned negative. They also kept feeding the account new creatives.

  • Step up gradually: Snag tested at about $50 a day, then went 100, 200 and 300 until ROAS turned negative [15]. Frederick James, who ran TikTok ads, said never raise a budget more than 20% every few days [18].
  • Tier your tests: Reflectly's Ryan Thorp tested each video at $10–20, moved winners to $100–500 ad sets, and gave the top one or two evergreens 90% of spend [19].
  • Don't scale too fast: Wrestle AI's George said spending faster raised CAC, and the real limit was producing enough creatives [20].
  • Fight fatigue: Snag said Meta has diminishing returns and creatives fatigue, so keep making more [15].
  • Structure: Burke split ad sets by media type, format and message, and gave outlier ads their own broad-targeted ad set [17].
  • Event data: Puff Count optimized for installs, then trials, then subscriptions as data built up [4].
  • Seasonality: Thorp said he spent about 75% of the annual budget in January, when acquisition cost is about half [19].

What can go wrong when you start Meta ads too early?

The main risk is spending on creatives that never had a winning signal, or running ads on an app whose revenue per install is too low. The library has several examples.

  • Wasted spend: Boost App Social lost about $800K of its $1.5–1.8M finding working creatives. A duplicated ad also burned $8K in a blip [5].
  • Platform risk: It was banned 6–8 times and had no organic channel, so rising CPMs could have ended profitability [5].
  • Weak economics: Young's Meta test on a roughly $3K-a-month organic app showed about $1 revenue per install, which he said was too low [14].
  • Product first: Journable's founder, speaking about Google Ads, said to raise budgets only as subscription conversion improves [21].

Some founders did well without Meta. Gravl said it found iOS ads unaffordable and ran Android Google campaigns instead [21]. AbMax found TikTok ads returned more on the same creatives [13].

What eight founders did before and on Meta ads, with the numbers they reported
Founder / appWhen ads startedBudgetResult they reported
Prayer ScreenAfter one organic TikTok winner [1]Not said$0.39–$0.80 per US install [1]
Pep AIAfter signing creators and finding winners [2]About $400/day total [2]About $2.50 per download [2]
SnagOn videos over 50K views [15]$50/day test, then 100, 200, 300 [15]Keep if ROAS above 1 [15]
Boost App SocialFrom the start, using UGC creatives [5]$100/day up to $10K/day [5]$100K MRR; 1 in 85 videos worked [5]
AbMaxAfter cutting UGC and influencer deals [13]Not said2.2–2.5x ROAS on Meta [13]
Prayer LockMeta first, then TikTok [7]$20/dayAbout $10 per trial on Meta [7]
GravlAfter a Reddit launch and a paywall [9]Under $50/day in South America [9]Subscription within 10 minutes [9]
Cal BuddyAds from month one in Israel [6]$1K in month one to about $20K/month [6]$20K/month after four months [6]

Frequently asked questions

How many of the founders in the library used Meta ads?

44 of the 185 interviews list Meta ads as a growth driver at some stage. Among the 34 that gave monthly revenue, the median was $55K, ranging from $10K to $1.5M.

Do you need organic content before running Meta ads?

Not strictly, but most founders did it that way. Only 3 of the 38 who said how they started began with paid ads, and several said organic winners made their ad spend cheaper and less risky.

What is a good cost per install on Meta?

Pep AI's founder said under $5 per download is usually doing well, and his was about $2.50. Prayer Screen's founder said $0.70–$0.80 per US install is far better than the roughly $1.50 most people accept.

How much should I budget to test Meta ads?

Founders tested at $20–$100 a day. Sizzle AI's growth lead suggested a $5K starting budget spread across many creatives, and SideShift's Lawton suggested a $2K–$4K first month on creator content to find winning formats.

How fast can I raise a Meta ad budget?

Founders raised budgets in steps. Snag went $50, $100, $200, $300 a day until ROAS turned negative. One TikTok advertiser said never raise more than 20% every few days. Wrestle AI said scaling too fast raised its CAC.

Sources

Founder interviews on YouTube; numbers are as the founders said them. How this research is done.

  1. [1]He Built a Simple $120k/Yr App, Then Sold It For Multiple 6 Figures! · Prayer Screen, Bible Buddy, Magic Music, Toxic Traits (Superwall)
  2. [2]I speedran an ai app to $100k in 70 days. · Pep AI (Superwall)
  3. [3]I turned a viral idea into a $25K/month app · Canary (Starter Story)
  4. [4]The $1M Mobile App Playbook · Puff Count, Posted (Starter Story)
  5. [5]Copy This Facebook Ads Strategy, It'll Blow Up Your App · Boost App Social (Superwall)
  6. [6]I Copied A Huge App. Now I Make $80K/Month · Cal Buddy (Starter Story)
  7. [7]I Built a $20K/Month Mobile App: Here’s My Entire Playbook · Prayer Lock (Starter Story)
  8. [8]#1 UGC EXPERT Reveals How To Market Consumer Apps · SideShift (Superwall)
  9. [9]How I Built It: $400K/Month Mobile App (Gravl) · Gravl (Starter Story)
  10. [10]How he makes $250K per month from a simple app (Letterly Breakdown) · Letterly (Starter Story)
  11. [11]How I Grow Viral Apps to 1M+ Downloads (just copy me) · Sizzle AI (Superwall)
  12. [12]I flipped 4 apps and made $500,000 · Bible Buddy, Magic Music, Toxic Traits, Pray Screen (Starter Story)
  13. [13]This crazy simple app makes $50K/month · AbMax (Starter Story)
  14. [14]Beginners Guide to Buying & Selling Apps 2026 · App Masters (app acquisitions portfolio) (Superwall)
  15. [15]I Built A $30K/Month App: Here's My Exact Process · Snag (Starter Story)
  16. [16]I Studied 1,000,000 UGC Ads, This Is What Will Grow Your App · SideShift (Superwall)
  17. [17]How to Build a $2M/Month App (Copy Me) (Superwall)
  18. [18]I Built a $10K/Month App in 88 Days (Starter Story)
  19. [19]Meet The Guy Dominating The App Store · Reflectly, Done (Do Habit), The Mindfulness App, Flo (Superwall)
  20. [20]Meet the ex-wrestler making $41k/mo with AI apps · Wrestle AI, Brawler, AbMax (Superwall)
  21. [21]I Built a $100K/Month Android App · Journable (Starter Story)