GTM Sauce

What apps pay for ads and views

Costs guests reported for paid ads and for creator content priced per view. Platform, country and creative change these a lot; each row says which applied.

38 figures · updated 2026-09-17

CPM (per 1,000 views or impressions)

Median $1.5 across 22 figures, from $0.1 to $10.

WhoMetricValueContextSaid in
SunflowercpmUSD 0.1
approximate
host's read of Sunflower's influencer-driven CPM; guest responds about sponsorship arbitrage without confirming the figureThe Superwall Podcast
2026-04-03
31:06
aggregateClover campaigns on cloned content (base plus performance deal)cpmUSD 0.3–0.5
approximate
Performance CPM paired with $3/post baseThe Superwall Podcast
2026-09-17
28:07
DecartcpmUSD 0.3
approximate
Upper bound: 'sub-30-cent' all-in UGC CPMThe Superwall Podcast
2026-03-16
4:42
aggregateAll-in CPM (retainers + bonuses) of SideShift programs spending >$100K/monthcpmUSD 0.8–0.9
approximate
The Superwall Podcast
2026-06-04
47:53
Cal AIcpmUSD 1–2
approximate
Influencer deal CPMs that were achievable earlier and became much harder to find in the last 6 monthsThe Superwall Podcast
2026-08-26
48:44
benchmarkcreator view bonus on top of retainercpmUSD 1–2
approximate
view bonus paid to creators per 1,000 views, capped or uncappedThe Superwall Podcast
2026-08-06
8:31
LockedcpmUSD 1–2
approximate
typical effective CPM paid to influencersThe Superwall Podcast
2026-06-11
9:38
aggregateSpark ads on high-engagement UGC videos across SideShift programscpmUSD 1
approximate
The Superwall Podcast
2026-06-04
30:18
Halo AIcpmUSD 1
approximate
expected creator-pay CPM under base+milestone; actual is lower (unpaid views between milestones, only one of 4 posts paid)The Superwall Podcast
2026-04-04
36:19
CluelycpmUSD 1
approximate
Upper bound: 'sub-$1' CPM for the early UGC programThe Superwall Podcast
2026-03-16
6:12
NomadtablecpmUSD 1–2per 1000 views
approximate
Performance-only CPM paid to UGC creators, no retainer; he calls $1-2 standard and decentThe Superwall Podcast
2026-03-15
40:44
CardstockcpmUSD 1
approximate
Effective CPM on Nois slideshow campaign, host's reading of the dashboardThe Superwall Podcast
2026-01-06
44:31
ArisecpmUSD 1–2per 1000 views
approximate
What influencers and meme pages were paid per 1,000 viewsThe Superwall Podcast
2025-12-02
18:50
ErlycpmUSD 2–3
approximate
Effective CPM paid to influencers under view-guarantee dealsStarter Story
2026-07-05
11:14
benchmarkAll-in UGC program CPM by month 3cpmUSD 2–3
approximate
The Superwall Podcast
2026-06-04
46:32
aggregatePopular UGC creator deal ~4 months earlier: view bonus ratecpmUSD 2
approximate
creator view bonus paid per 1,000 viewsThe Superwall Podcast
2026-06-04
43:33
Jenni AIcpmUSD 2
approximate
Blended cost per 1,000 views across the UGC program (upper bound)The Superwall Podcast
2025-12-30
4:57
JunglecpmUSD 2per 1000 views
approximate
UGC program cost per 1,000 views; varies a bit; much better than influencer marketingThe Superwall Podcast
2025-12-09
23:03
aggregateHighest CPM other apps have offered Jay's UGC accounts in DMscpmUSD 3per 1000 viewsHighest offer seen; $1-2 is the usual rangeThe Superwall Podcast
2026-03-15
41:57
benchmarkUGC program CPM at which founders can step back from day-to-daycpmUSD 4–5
approximate
consistently sub $4-5The Superwall Podcast
2026-06-04
12:22
benchmarkAll-in UGC program CPM, month 1 (retainers + bonuses)cpmUSD 5–8
approximate
The Superwall Podcast
2026-06-04
46:26
Cal AIcpmUSD 10
approximate
Some mom influencers were profitable at a $10 CPMThe Superwall Podcast
2026-08-26
18:14

Cost per install

Median $0.73 across 10 figures, from $0.08 to $100.

WhoMetricValueContextSaid in
FittedcpiUSD 0.01–0.15
approximate
Cost per download when putting ad money behind viral organic videos. Host quotes it from something the guest said before; the guest explains the method without repeating the numbers.The Superwall Podcast
2025-12-23
1:20
NoisecpiUSD 0.08
approximate
For about three days while the 'don't get a second job' dishwashing video ranThe Superwall Podcast
2026-02-24
14:40
NoisecpiUSD 0.3
approximate
Average cost per install for Noise promoting itself on its own platformThe Superwall Podcast
2026-02-24
10:48
PraycreencpiUSD 0.39/dayBest day, US, Meta ads on InstagramThe Superwall Podcast
2025-11-03
6:01
PraycreencpiUSD 0.7–0.8
approximate
Average, US, Meta ads on InstagramThe Superwall Podcast
2025-11-03
6:01
Shot PatterncpiUSD 0.7
approximate
From a founder-made screen-recording video ad with CapCut captionsSub Club by RevenueCat
2025-07-09
32:15
Pep AIcpiUSD 2
approximate
best-performing static image ad (feature-list graphic)The Superwall Podcast
2026-07-01
47:13
Pep AIcpiUSD 2.5
approximate
Meta ads using repurposed organic creator winnersThe Superwall Podcast
2026-07-01
6:54
benchmarkCost per download considered good for a subscription app (heard secondhand)cpiUSD 5
approximate
under $5 is 'usually doing pretty good'The Superwall Podcast
2026-07-01
6:54
Large app account with no spend guardrailscpiUSD 100
approximate
CPI of the runaway creative ($100+)Sub Club by RevenueCat
2025-12-24

Cost per trial, purchase or customer

Median $46.75 across 6 figures, from $0 to $500.

WhoMetricValueContextSaid in
Architectural DigestcacUSD 0Incremental CAC of the post-purchase upsellSub Club by RevenueCat
2025-10-29
benchmarkMaximum cost per trial to consider Meta ads when optimizing for trialscpaUSD 10–15
estimate
cost per trialThe Superwall Podcast
2026-07-22
33:50
anonymized client ad account (charts shown)cpaUSD 32–35
approximate
lowest cost per result among ~30 ads; these ads had significant Facebook feed spendThe Superwall Podcast
2026-06-24
50:06
anonymized client ad account (charts shown)cpaUSD 60
estimate
cost per target event (likely trial) goal for this advertiserThe Superwall Podcast
2026-06-24
47:16
Sizzle AIcpaUSD 300per purchase
approximate
Cost per purchase within days of 3-4-way paywall and in-app experiments on winning-ad trafficThe Superwall Podcast
2025-08-04
18:19
Sizzle AIcpaUSD 500per purchase
approximate
Cost per purchase before rapid in-app testing (varies by targeting)The Superwall Podcast
2025-08-04
18:19

Figures are as stated in each interview and aren't independently verified. Goals, projections, illustrative numbers and before/after changes are left out. Ranges count as their midpoint in the median.

Tactics on this