How Much Does AppLovin Cost to Advertise? Minimum Budget, CPCs and Real Test Budgets (2026)

How much does AppLovin cost? For an advertiser, AppLovin Ads can launch with as little as $10/day, and what you pay is your media budget, funded on prepay for new accounts. A test that actually tells you whether the channel works costs more than that floor. The real number depends on your cost per purchase, the ROAS or cost-per-purchase goal you set, and the creative you have to produce.

This page is about what a brand pays to run ads on AppLovin. If you searched for the APP share price, that is a different question and this isn’t the page for it.

I’m Tommy Patterson. WKND runs AppLovin Ads for ecommerce brands, so below I’ll separate what AppLovin charges, what the auction decides, what an agency charges, and what nobody puts on the invoice.

The short version

Cost line	What it is	Who sets it	Where the figure comes from
Platform entry	You can launch at $10/day	AppLovin	AppLovin’s consumer-brands FAQ
Media	Your daily budget, spent against a ROAS, cost-per-purchase (CPP) or cost-per-lead (CPL) goal	You set budget and goal; Axon prices delivery	AppLovin
Billing	New accounts use prepay	AppLovin	WKND’s setup requirements
CPC / CPM	Not a list price. An outcome of the auction. WKND’s anonymized cases saw $1.11 and $1.75 CPC	The auction	WKND case results (below)
Management	Separate from media if you use an agency	The agency	WKND’s fee statement
Creative	At least one 9:16 video and one Interactive Page per Creative Set, refreshed as it fatigues	You or your agency	WKND / AppLovin specs
Measurement	Test time and, ideally, a holdout	You	WKND’s test method

The $10/day floor tells you how cheaply you can switch the channel on. It does not tell you what a useful test costs. Most of this article is about the gap between those two numbers.

AppLovin pricing: you pay for media, not a subscription

AppLovin Ads is a performance channel. In its June 22, 2026 announcement opening the platform to every advertiser, AppLovin’s CEO wrote that the company doesn’t do brand marketing. You buy toward a measurable result: a ROAS target, a cost per purchaser, or leads (AppLovin).

What you are billed for is that media. AppLovin’s SEC filing describes AppLovin Ads revenue as the amount charged to advertisers, dynamically priced based on campaign goals, less what is paid to publishers (Form 10-Q revenue note). AppLovin’s Billing help page describes charges based on ad spend only: your card is charged (or your balance deducted) daily, and it lists no separate subscription, seat licence or platform fee.

Three practical points:

• The floor is $10/day. AppLovin’s consumer-brands FAQ says you can launch “with a budget as low as $10/day” (AppLovin).
• New accounts pay up front. WKND’s setup requirements: “New accounts use prepay. Add a backup card and 2FA.” AppLovin’s billing page confirms it: all new accounts are on prepay, and select accounts may be eligible for postpay through an AppLovin representative. The card is charged at 0:00 UTC for the next day’s budget, and any unused funds return to your balance around 03:00 UTC. Plan cash flow so a campaign doesn’t stall on a failed charge; AppLovin recommends one or two backup cards.
• There is no rate card. You don’t buy a fixed CPM or CPC. The auction and your goal set the price.

If you want the other side of the ledger, including how AppLovin pays publishers and why there is no fixed “pay per click” rate, that is covered in How Does AppLovin Make Money?. This page stays on the advertiser’s bill.

AppLovin minimum budget: $10/day is the floor, not the plan

Being allowed to launch and having enough budget to learn are different questions.

AppLovin’s own web setup guidance, as we summarized in our guide to how AppLovin Ads works, step by step, says a budget capable of driving around 10 purchases per day on your existing social channels is a reasonable starting point for giving the model enough signal (AppLovin setup guide). That is guidance, not a minimum. But it gives you a far better planning number than $10:

Starting daily budget ≈ 10 × your current cost per purchase on social.

For example (illustrative), a brand with a $40 CPA on Meta is looking at roughly $400/day by that yardstick, not $10.

Two cautions:

• AppLovin’s app-campaign guidance (roughly 15 to 20 conversions per day) is for app advertisers (AppLovin apps setup). Don’t copy it into a web campaign as a rule.
• Floors vary by platform. TikTok requires campaign budgets above $50 and ad group budgets above $20 (TikTok Ads Manager help). A lower floor on AppLovin doesn’t make a test cheaper. The conversions you need to learn from set the cost.

How ROAS, CPP and CPL goals shape what you pay

You don’t bid a CPM or a CPC on AppLovin Ads. You pick a goal and a target, and Axon decides how much to pay for each impression it thinks can hit that target. The target works like a price ceiling.

Goal	What you’re telling the engine	What happens to spend
ROAS target	“Only buy impressions you expect to return at least X in revenue per dollar.”	A tight target caps delivery. Loosening it in steps opens up volume.
Cost per purchaser (CPP)	“Pay up to about $Y for a purchaser.”	Spend follows how many purchasers the engine can find at or under $Y.
Leads (CPL / value per lead)	“Buy leads worth at least this much.”	Same logic, for lead-gen businesses.

This is why raising the budget alone often does nothing. WKND’s published diagnosis for accounts stuck on a plateau: “The ROAS goal is doing its job: it caps delivery at the price the engine can hit. Raising the budget does nothing; loosening the target in steps does.”

So the most important cost decision isn’t the budget. It’s the target, and it should come from your economics. WKND’s rule: “Set the target from contribution margin and CAC.” The simplest anchor is breakeven:

Breakeven ROAS = 1 ÷ gross margin.

At a 50% gross margin, breakeven ROAS is 2.0x. At 55%, it’s about 1.82x. Set a first-order target far above breakeven and you’ll buy very little. Set it at or below breakeven and every sale you buy loses money unless repeat purchases pay it back.

Two more settings move cost:

• The day target (the conversion window the campaign optimizes to) changes the objective itself, not just reporting. A higher-AOV product forced onto a same-day target can look worse than it is. See the day-target section of our setup walkthrough.
• Budget changes. AppLovin recommends incremental daily increases while goals are being met, and making larger changes at 12:00 AM UTC, when campaigns reset (AppLovin).

AppLovin CPM and CPC: what our accounts actually saw

AppLovin doesn’t publish a CPM or CPC rate card in the materials we reviewed, and it treats both as secondary metrics. Its campaign-analysis guidance says to judge campaigns first on the ROAS, CPP or CPL goal, and warns against using CPM or CTR in isolation. A high CPM can still show up on a profitable day (AppLovin).

Here is what WKND’s anonymized case results show. These are the only click-cost numbers we publish:

Brand type	Window	Spend	CPC	CTR	NC D28 ROAS
Jewelry brand	March 4 to 21, 2026 (18 days)	$127,000 total	$1.11	not published	2.94x
Food & beverage CPG brand	January 24 to February 10, 2026 (18 days)	$68,437.26	not published	4.77%	2.52x
Footwear brand	August 14 to 31, 2026 (18-day validation)	$4,062.79	$1.75	4.78%	3.52x

“Reporting shown is anonymized. ROAS figures are AppLovin Ads new-customer cohort metrics from the stated reporting windows. Platform attribution can differ from blended or incrementality measurement, and results vary by account.” NC D28 ROAS is AppLovin Ads’ new-customer cohort ROAS over 28 days.

What to take from this:

• CPM: no number here. WKND doesn’t publish a CPM figure from these accounts, and I’m not going to make one up. CPM varies with your goal, target, creative and competition for the same inventory. If you find a CPM benchmark elsewhere, check its date, whether it’s from app or web campaigns, and how many accounts it covers.
• CPC doesn’t predict profit. The footwear account paid more per click than the jewelry account ($1.75 vs $1.11) and still posted the higher NC D28 ROAS in its window. Budget from cost per purchase and margin. Treat CPC and CTR as diagnostics, mainly for creative.
• Small samples mislead. The footwear result is, in WKND’s own words, “an efficient validation result, not a scale case.” Your CPC during a small validation and your CPC at scale won’t be the same number.

Realistic AppLovin ads budget for a first test

There’s no official “test budget.” Here are the reference points that exist, and who they come from:

Reference point	Figure	Source	What it is
Platform floor	$10/day	AppLovin	The minimum to launch
Learning signal	Budget for ~10 purchases/day at your social CPA	AppLovin web setup guide	Guidance, not a minimum
WKND qualification threshold	At least $500 per day	WKND	Who WKND takes on. Not an AppLovin platform minimum and not a performance guarantee
DDU Media recommendation	$10–15k over 4–6 weeks for a test; $30k+/mo total paid spend	DDU Media	A third-party agency’s recommendation
Avenue Z	Start “with as little as $500 per day”	Avenue Z	A third-party agency’s claim: “Start with as little as $500 per day to test campaign viability”

WKND’s services page puts its threshold this way: “AppLovin Ads allows advertisers to get started with as little as $10 per day. WKND is built for ecommerce brands prepared to invest at least $500 per day … The $500 figure is WKND’s qualification threshold, not an AppLovin platform minimum or a performance guarantee.”

The part most budgets miss is measurement. A test isn’t finished when platform ROAS looks good. It’s finished when you know whether the sales were incremental. WKND’s rule of thumb: “a channel under roughly 10% of total spend will not move blended MER detectably in a month.” That changes how you design the test:

• If AppLovin will be 10% or more of total paid spend, a geo holdout is possible.
• If it will be under 10%, WKND uses an on/off test of about 2 weeks instead.
• Either way: take a 4-week baseline, run the test for 2–4 weeks, and hold everything else still.

That timeline is part of the cost. So is the budget you hold back from other channels while you test. The full method is in AppLovin attribution and incrementality.

How much does AppLovin cost for a real test? A worked example (ILLUSTRATIVE)

Every number in this example is invented to show the arithmetic. It is not a WKND client, a benchmark or a forecast. Swap in your own figures.

A brand sells a product with a $90 average order value and a 50% gross margin. It spends $150,000 a month across paid channels, and its Meta CPA is $45.

1. Breakeven ROAS. 1 ÷ 0.50 = 2.0x. Below that, first orders lose money.
2. Breakeven CPA. $90 × 50% = $45 of gross profit per first order. That’s the most one new customer can cost before the first order stops paying for itself.
3. Starting daily budget. Using AppLovin’s ~10 purchases/day guidance at the current $45 social CPA: 10 × $45 = $450/day.
4. Test length and media cost. A 21-day test (inside WKND’s 2–4 week window): 21 × $450 = $9,450 in media.
5. Share of spend. $450/day is about $13,500 a month, or 9% of $150,000. That’s under the ~10% rule of thumb, so blended MER probably won’t show the effect. Plan an on/off test rather than trusting a before-and-after MER read.
6. Decision line. WKND’s decision table compares incremental ROAS to breakeven. At 30% or more above breakeven (here, 2.6x or better), scale in steps and retest at 2x spend. Within 30% of breakeven, hold, fix creative or feed, and retest. Below breakeven, cut or restructure.
7. Add what isn’t media. Creative production for the Creative Sets, any tracking or catalog work, and management fees if you use an agency. None of those are in the $9,450.

The point of the exercise: this brand’s real test costs roughly $9,450 in media plus creative and setup. That’s what to budget. $10 a day isn’t.

The hidden costs: management fees, creative and setup

Agency management fees

If you hire an agency, its fee sits on top of media. WKND’s published wording: “Your AppLovin Ads media budget funds campaign delivery. WKND’s management fee is separate.” Pricing “depends on monthly spend, campaign complexity, creative volume, tracking or catalog work, and measurement needs,” and “every proposal separates media, management, and optional creative or technical work.”

WKND doesn’t publish a fee number on this page. Whoever you talk to, ask for the same three-way split: media you pay AppLovin, the management fee, and creative or technical work. If a proposal blends them, you can’t tell what the channel itself costs.

Creative production

This is the cost most first tests underestimate. WKND’s requirement: “Each Creative Set needs at least one 9:16 video and one Interactive Page.” AppLovin’s specs call for full-screen vertical video up to 60 seconds. It encourages a range of lengths, ideally including longer videos of 30 seconds or more, varied styles, and captions for sound-off viewing (AppLovin ad specs).

Your Meta ads are a starting point, not a finished library. WKND’s diagnosis when Meta creative underperforms on AppLovin: it’s “a format problem before it is a messaging problem,” and hook, pacing and end card usually get rebuilt. Creative also fatigues, so a test that scales needs fresh Creative Sets. Budget editing and design time as its own line.

Tracking and catalog setup

Before any spend is worth reading, the Shopify app or AppLovin Pixel has to be installed and validated: required ecommerce events, purchase value, item data, transaction IDs, click parameters and server-event deduplication. Your catalog also has to be matched on item and variant IDs, prices, availability, images and URLs. It’s one-time work, but it’s real work, and a broken pixel wastes every dollar spent before it’s fixed.

Ad credits

Some agencies and creators advertise a spend-match offer, roughly $5K in AppLovin ad credits after $5K of spend, through referral links. Its terms were not found on an AppLovin page when we checked. AppLovin’s partner page does say partners earn 3% of ad spend for the first 12 months after a brand signs up (AppLovin partner program). Don’t build a test budget that depends on credits until AppLovin confirms the terms in your account. More in AppLovin referral codes and ad credits.

How WKND handles this

We price the test before we price anything else. A first call with WKND covers the requirements that decide cost:

• Billing: prepay set up, a backup card and 2FA added, WKND invited at the right access level.
• Tracking: Shopify app or Pixel via GTM, validated before spend.
• Purchase history: Shopify connected or order history uploaded in Events Manager, so Prospecting and Discovery audiences work.
• Catalog: Shopify, a Pixel-built catalog or a daily CSV feed, matched to the site.
• Creative: at least one 9:16 video and one Interactive Page per Creative Set.
• Goal: ROAS, CPP or CPL, set from contribution margin and CAC, with breakeven written down before launch.

What that has looked like in practice, across $5M+ in Axon spend on 50+ accounts and 5,000+ Creative Sets deployed:

• A jewelry brand went from $268.12/day on March 4 to $24,784.28/day on March 21, 2026 (“$268/DAY → $24.8K/DAY IN 18 DAYS”).
• A food & beverage CPG brand went from $111.12/day to $12,954.25/day between January 24 and February 10, 2026.
• A footwear brand ran an 18-day validation (August 14 to 31, 2026) on $4,062.79 before any scale decision.

Same caveat as above: anonymized, platform-reported new-customer cohort metrics. Platform attribution can differ from blended or incrementality measurement, and results vary by account. The full breakdowns are in our AppLovin case studies.

On setup speed, Anand Dixit of Groomie: “Dude we were live in 5 days. That has to be a record lol.” On the creative line item, Jenny Trinh of Daniel Patrick: “Those new end cards are sick!! CTR literally doubled.”

We’re not the right fit for every budget. WKND is built for brands with proven product-market fit, “prior paid scale of $10k+/day with positive unit economics,” and the creative volume to feed the channel. If you’re still validating product-market fit, or you need a small budget to pay back immediately, we’ll tell you so. If that’s you, it’s a better use of money to wait. If it isn’t, here’s how our AppLovin Ads management works: setup, ROAS-goal bidding, creative for app inventory and measurement that holds up against blended MER.

For our own management fee, see AppLovin agency pricing: what WKND charges.

FAQ

How much does AppLovin cost per month?

None of the AppLovin materials reviewed for this article describe a monthly subscription. You pay for the media you run. The platform floor is $10/day, so about $300 a month if you ran the minimum every day. A useful ecommerce test usually costs much more, because it needs enough purchases for the engine to learn from. Size it from your cost per purchase, not the floor.

What is the minimum budget for AppLovin ads?

AppLovin says you can launch with a budget as low as $10/day. That is a platform minimum, not a recommended test budget. AppLovin’s web setup guidance points to a budget that could drive around 10 purchases a day on your existing social channels as a reasonable starting point.

What is the CPM on AppLovin?

AppLovin doesn’t publish a fixed CPM, and WKND doesn’t publish a CPM figure from its accounts here. CPM varies with your goal, target, creative and competition. AppLovin treats CPM as a secondary metric and recommends judging campaigns on ROAS, CPP or CPL.

How much does AppLovin pay per click?

That question usually comes from publishers or from people researching AppLovin’s business model. There’s no universal AppLovin pay-per-click rate. We explain the publisher and business-model side in How Does AppLovin Make Money?. On the advertiser side, WKND’s anonymized cases saw $1.11 and $1.75 CPCs, which are outcomes of the auction, not prices.

Is AppLovin cheaper than Meta?

Some agencies advertise CPAs 20–40% lower than Meta. That claim is unsourced, and you shouldn’t budget on it. Cost depends on your creative, your target and whether the customers are incremental. Run a controlled test and compare incremental cost per customer. Our AppLovin vs Meta Ads comparison covers how the two channels differ.

Does AppLovin charge a platform fee?

None of the AppLovin materials reviewed here describe a separate platform or subscription fee. AppLovin Ads is billed on media, dynamically priced against your campaign goal, and new accounts pay by prepay. If you use an agency, its management fee is separate from media.

How long should an AppLovin test run?

WKND’s method is a 4-week baseline, then a 2–4 week test with everything else held still, then a comparison of incremental ROAS to breakeven. DDU Media recommends $10–15k over 4–6 weeks. Either way, plan in weeks, not days, and decide how you’ll measure incrementality before you start.

Get the test sized before you spend

The cheapest AppLovin test is one set up right the first time: the goal set from your margin, a validated pixel, Creative Sets built for full-screen app inventory, and a measurement plan agreed before day one.

Request an AppLovin readiness review → We’ll check billing, tracking, purchase history, catalog, creative and goal settings, and tell you what a meaningful first test would cost for your account.

Already spending on AppLovin? GET AN AXON ACCOUNT REVIEW: goal history, feed and event health, creative fatigue, incrementality readiness. Written findings inside five business days. Start at /applovin-axon-services/.

Sources

1. AppLovin Ads for consumer brands (FAQ: $10/day), AppLovin, read 2026-09-27.
2. AppLovin Ads is now open to all advertisers, AppLovin, June 22, 2026.
3. Setting up a campaign (Web), AppLovin.
4. Setting up a campaign (Apps), AppLovin.
5. Analyzing campaign performance, AppLovin.
6. Optimizing your campaign, AppLovin.
7. Ad specs and requirements (web), AppLovin Support.
8. AppLovin Form 10-Q revenue note, filed August 5, 2026.
9. AppLovin partner program, AppLovin.
10. AppLovin Axon Ads Manager Agency, WKND, read 2026-09-27.
11. AppLovin Ads Agency for E-Commerce, DDU Media, read 2026-09-27.
12. AppLovin Ads Agency for eCommerce Brand Growth, Avenue Z, read 2026-09-27.