Retargeting Calculator

How the retargeting math works

The calculator follows one chain from visitors to profit, and each step uses one input. The people you can retarget are your monthly visitors times the match rate. Impressions are those people times the frequency, and spend is impressions ÷ 1,000 × CPM. Clicks are impressions × click-through rate, conversions are clicks × conversion rate, and revenue is conversions × average order value. ROAS is revenue ÷ spend, and CPA is spend ÷ conversions.

Gross profit = revenue × gross margin − spend

Break-even ROAS is 1 ÷ gross margin, and break-even CPA is average order value × gross margin. In budget mode the chain starts from the money instead: impressions = budget ÷ CPM × 1,000, and frequency = impressions ÷ people. When a step would divide by zero, for example a CPA with no conversions, the result shows a dash, not a made-up number. Nothing is rounded until it is displayed, which is why 9.6 conversions stay 9.6.

A worked example you can redo by hand

These are the calculator's starting values. They are examples, not benchmarks.

1. People you can retarget: 10,000 × 50% = 5,000
2. Impressions: 5,000 × 8 = 40,000
3. Spend: 40,000 ÷ 1,000 × $10 = $400
4. Clicks: 40,000 × 0.8% = 320
5. Conversions: 320 × 3% = 9.6
6. Revenue: 9.6 × $80 = $768
7. ROAS: $768 ÷ $400 = 1.92×
8. Gross profit: $768 × 40% − $400 = $307.20 − $400 = −$92.80
9. Break-even ROAS: 1 ÷ 0.40 = 2.5×
10. CPA vs break-even CPA: $400 ÷ 9.6 = $41.67, against $80 × 40% = $32

A 1.92× ROAS looks like a win next to 1×, because sales are bigger than spend. That comparison uses revenue, and it is the wrong one. Of every $80 order, $48 goes on the goods themselves, so only $32 is left to pay for the ad that brought the sale in, and each sale here costs $41.67 in ads. Break-even is measured on gross margin, not revenue, so this campaign needs 2.5× just to stand still. The ROAS calculator and the CPA calculator run the same checks on a campaign you have already run, and break-even ROAS vs ROAS explains why the two numbers part ways.

What the model leaves out

It holds click-through and conversion rates flat however often people see the ad. It credits the ad with every conversion, including from people who were coming back anyway. If the conversion rate you enter comes from a report that includes view-through conversions, it will be higher than clicks alone produce. And it leaves out tool and agency fees, ours included.

How to set a retargeting budget

A retargeting budget has a ceiling that a prospecting budget doesn't: your own traffic. You can only show these ads to people who already visited, so the budget comes from how many of them there are and how often you want each one to see an ad.

Monthly budget = visitors × match rate × frequency × CPM ÷ 1,000

With the example numbers, 10,000 × 50% × 8 × $10 ÷ 1,000 = $400 a month. Spending more than that doesn't reach more people. It shows the same people more ads. Budget mode in the calculator answers the reverse question: enter what you can spend, and it works out the frequency that buys, with a warning above 20 a month.

Before you spend it, check the budget can pay for itself. Break-even CPM is the most you can pay per 1,000 impressions at your click-through rate, conversion rate, order value and margin:

Break-even CPM = 1,000 × CTR × conversion rate × order value × gross margin

In the example that is 1,000 × 0.008 × 0.03 × $80 × 0.40 = $7.68. The example pays $10, so every extra dollar of budget adds to the loss. A bigger budget cannot rescue a CPM above break-even. For prospecting, where the audience isn't capped by your traffic, use the ad budget calculator. The CPM calculator turns any budget and CPM into impressions, and the guide to budgets, pacing and pricing covers daily and lifetime budgets.

What changes the answer most

Move one input at a time away from the example and a pattern shows. Match rate and frequency change how much you spend and how many dollars you make or lose, but not the ROAS. Twice the impressions buy twice the clicks at the same rates. Whether retargeting pays at all comes down to CPM, click-through rate, conversion rate, order value and margin.

That flat ROAS is a property of the model, which holds click and conversion rates steady. In a real account they can fall as the same people see the ad again and again, so read the high-frequency rows as a best case.

Match rate

Match rate is the share of your visitors who end up in an audience you can actually reach, such as a Meta website custom audience or a Google remarketing list. A visitor whose consent choice or ad blocker stops your tag from running can't be added. Safari limits cookies too: WebKit announced in February 2019 that cookies created through document.cookie are “capped to a seven day expiry.” To estimate yours, compare the size your ad platform shows for a 30-day website audience with your analytics' users for the same 30 days. The ratio is your match rate; treat it as rough.

Frequency

To see the frequency you run at now, divide a past campaign's impressions by its reach for the same dates (the frequency vs reach explainer covers the difference). Google lets Display and Video campaigns cap impressions per person “per day, week, or month”. Meta documents its frequency cap for Awareness and Engagement campaigns using the Reach and ThruPlay goals; we could not find it documented for sales campaigns. Without a cap, budget and audience size set the frequency, which is what budget mode models. The warning above 20 a month is our threshold, not a platform rule.

CPM

Use a CPM from your own account, ideally from a past retargeting campaign. In Meta Ads Manager, open the Columns dropdown and choose Customize columns, then add CPM, which Meta defines as the amount spent “divided by impressions, multiplied by 1,000.” In Google Ads, take a Display campaign's Cost and Impressions for the same dates and work it out: cost ÷ impressions × 1,000. If CPM comes out above your break-even CPM, no amount of budget fixes it. Raising the click-through rate or the conversion rate raises the break-even CPM instead.

When retargeting is the wrong spend

Retargeting can look like the easiest win in paid ads. Sometimes it isn't, and a tool or agency that runs ads for you, ours included, usually gets paid whether or not it works. Four cases where the honest answer is to spend less, or nothing:

• Your CPM is above your break-even CPM.

  Scaling makes the loss bigger in proportion. Fix the offer, the order value or the landing page first; more retargeting won't.

• The audience is too small to learn from.

  Meta says ad sets usually leave the learning phase “after about 50 results in the week after the ad set's last significant edit”, and its Learning Limited page names “small audience size” as one cause of getting stuck. Here is what that means for a 1,000-person audience optimizing for purchases. At the example's 0.8% click-through and 3% conversion rates, each impression produces 0.008 × 0.03 = 0.00024 conversions. So 50 a week takes 50 ÷ 0.00024 = 208,333 impressions, which is 208 ads per person, every week. Meta also suggests waiting for “several hundred people” in a website audience before using it, and Google won't serve to a data segment with fewer than 100 active users in the last 30 days. The Meta learning phase calculator checks this for your own numbers.

• The buyers were coming back anyway.

  The calculator credits the ad with every retargeted sale. Only a holdout test, where part of the audience is deliberately not shown the ads, measures how many of those sales the ads caused. A healthy-looking ROAS can hide a much smaller real effect.

• Fees eat the margin, including ours.

  The calculator leaves tool and agency fees out. If retargeting were the only thing you used AdFlint for, the $40-a-month Launch plan would turn the example's $92.80 monthly loss into $132.80.

If you're choosing between finding new people and re-reaching old ones, prospecting vs retargeting sets out what each is for, and remarketing vs retargeting untangles the two names.

Sources

Every quote below was last verified on 2026-10-12. The formulas and worked examples on this page are arithmetic and need no source.

• Meta Business Help Center

  About the Learning Phase

  “ad sets exit the learning phase as soon as they can deliver stably. This usually occurs after about 50 results in the week after the ad set's last significant edit.”

  No last-updated date is shown on the page. Last verified 2026-10-12.

• Meta Business Help Center

  About Learning Limited

  “an ad set becomes learning limited when the ad set is limited by small audience size, low budget, low bid or cost control, high auction overlap, an infrequent optimization event, or other issues”

  No last-updated date is shown on the page. Last verified 2026-10-12.

• Meta Business Help Center

  Troubleshoot website custom audience size

  “Waiting until you have several hundred people in your website custom audience before using it.”

  Listed as one of Meta's suggestions. No last-updated date is shown on the page. Last verified 2026-10-12.

• Google Ads Help

  How your data segments work

  “Google Display Network must have a minimum of 100 active visitors or users within the last 30 days.” The page gives the same 100-user minimum for the Search Network and YouTube.

  Section “Size of your data segment”. No last-updated date is shown on the page. Last verified 2026-10-12.

• Google Ads Help

  Frequency capping: Definition

  “you manage the number of impressions you will allow an individual user to have on the campaign per day, week, or month.”

  Covers Display and Video campaigns. No last-updated date is shown on the page. Last verified 2026-10-12.

• Meta Business Help Center

  Choosing your frequency controls for a reservation campaign

  “Target frequency and frequency cap are supported in the awareness and engagement campaign objective's maximize reach and ThruPlay performance goals.”

  No last-updated date is shown on the page. Last verified 2026-10-12.

• Meta Business Help Center

  CPM (Cost per 1,000 Impressions)

  “CPM measures the amount spent on an ad campaign, divided by impressions, multiplied by 1,000.”

  No last-updated date is shown on the page. Last verified 2026-10-12.

• Meta Business Help Center

  Customize Columns In Meta Ads Manager

  “Click the Columns dropdown menu and then select Customize columns. Select the columns you want to see.”

  No last-updated date is shown on the page. Last verified 2026-10-12.

• WebKit blog

  Intelligent Tracking Prevention 2.1 (February 21, 2019)

  “all persistent client-side cookies, i.e. persistent cookies created through document.cookie, are capped to a seven day expiry.”

  Apple's WebKit team on Safari. Later WebKit posts extend the seven-day limit to other script-writable storage. Last verified 2026-10-12.

What we could not confirm

• Any average retargeting CPM, click-through rate or conversion rate published by Meta or Google. We found none, so every default in the calculator is an example, not a benchmark.
• A “right” retargeting frequency from either platform. The warning above 20 impressions per person a month is our own threshold.
• A platform rule that an audience of about 1,000 people cannot leave Meta's learning phase. Meta publishes the roughly-50-results-a-week guidance and lists small audience size as a cause of Learning Limited; the per-person arithmetic on this page is ours.
• How much consent banners, ad blockers and Safari shrink a typical site's match rate. It varies by site, and we found no neutral, dated measurement.
• How many retargeted buyers would have bought anyway. Neither Meta nor Google publishes a general figure that we could find.
• Google's help page for the Avg. CPM report column. We couldn't load one, so the Google instructions below work CPM out from cost and impressions instead.
• The longest a Meta website custom audience keeps a visitor. Meta's developer documentation gives 180 days for one setting and 365 days in its FAQ, so we don't state a figure.