Free tool
Ad Budget Calculator
Split a monthly ad budget the way media buyers do: 70% prospecting, 20% retargeting, 10% testing. A $3,000 month becomes $2,100, $600 and $300, each with a cost per conversion it has to hit.
Last updated: 2026-08-06
How much should you spend on ads is the wrong first question. Two accounts running the same $3,000 a month can produce completely different results, because one put all of it behind a single audience it had already exhausted and the other split it between finding new people, closing the people it had already reached, and finding out what to try next month.
The split most media buyers converge on is 70/20/10. Seventy percent goes to prospecting, the cold audiences that decide how big the account can get. Twenty percent goes to retargeting the visitors, viewers and cart abandoners you already paid to acquire, which is the cheapest revenue in the account and also the fastest to run dry. Ten percent goes to testing: new creative, new audiences, new offers, budget you have already decided you are willing to lose.
Sizing each bucket takes two numbers. Your monthly budget, and what an average order is worth to you. From there the platform benchmarks fill in the rest: a $3,000 month on Meta at a $14 CPM buys roughly 214,286 impressions and 3,000 clicks, and at the blended 8.8% landing page conversion rate that is about 264 conversions at $11 each. Those figures tell you whether each bucket is big enough to produce enough conversions to learn anything, which is the real constraint on a small budget.
The rule of thumb worth keeping is that any bucket needs roughly 30 to 50 conversions a month before its numbers mean anything, and each ad set needs enough daily budget to clear the platform learning phase rather than sitting in it forever. If the split leaves a bucket under that line, the answer is not a smaller split. It is fewer ad sets, or a bigger budget, or accepting that this month is a test and not a scale-up.
impressions = budget / CPM x 1000, clicks = impressions x CTR, conversions = clicks x CVR
Enter a monthly budget and your average order value to see the projection.
Every benchmark this page uses is printed further down, so you can read the numbers without entering anything at all.
Advertising benchmarks by platform
The default assumptions behind every projection on this page, plus what each one produces on a $3,000 month. The last two columns do not depend on your order value, so they are the fairest way to compare the five platforms against each other.
| Platform | CPM | CTR | Landing page CVR | Clicks on $3,000 | Conversions | Cost per conversion |
|---|---|---|---|---|---|---|
| Meta (Facebook and Instagram) | $14 | 1.4% | 8.8% | 3,000 | 264 | $11 |
| Google Ads | $38 | 4% | 5% | 3,158 | 158 | $19 |
| TikTok Ads | $10 | 0.9% | 2.5% | 2,700 | 68 | $44 |
| LinkedIn Ads | $33 | 0.6% | 4% | 545 | 22 | $138 |
| Pinterest Ads | $6 | 0.6% | 2% | 3,000 | 60 | $50 |
Benchmarks are estimates blended from public advertising benchmark roundups and platform-published averages, last reviewed 2026-08-03. They are all-industry blends, so your vertical, offer and creative can move every number here. Use them as a planning starting point, not a forecast.
Read the cost per conversion column as the order value you need just to break even, because revenue matches spend exactly at that point. Pinterest is the cheapest impression in the set at $6 per thousand and LinkedIn the most expensive customer at $138, which is why one sells homeware and the other sells software. The Google row is a search and display blend rather than pure search, so its 4% click-through sits below what a search-only account would report and its $38 CPM above what a display-only one would.
How the math works
Five steps, each one computed from the unrounded result of the step above it. The example below spends $3,000 on Meta (Facebook and Instagram) at a $60 average order value, using the platform benchmarks as they stand.
- 1impressions = budget / CPM x 1,000
$3,000 / $14 x 1,000 = 214,286 impressions
CPM is priced per thousand impressions, which is where the 1,000 comes from.
- 2clicks = impressions x CTR
214,286 x 1.4% = 3,000 clicks
Click-through rate is the first place creative shows up in the chain, and doubling it doubles everything below.
- 3conversions = clicks x conversion rate
3,000 x 8.8% = 264 conversions
This is the landing page rate, not the ad. It is the assumption that varies most between accounts.
- 4revenue = conversions x average order value
264 x $60 = $15,840
Order value is the only input the platform has no say in, and the one you can raise without spending more.
- 5ROAS = revenue / spend
$15,840 / $3,000 = 5.28x
Cost per conversion falls out of the same numbers: $3,000 / 264 = $11.
Why the assumptions matter more than the math
Hold everything else and move the landing page conversion rate from the 8.8% all-industry blend to a more typical ecommerce 2%, and the same $3,000 goes from 264 conversions at 5.28x to 60 at 1.20x, with cost per conversion rising from $11 to $50.
Order value swings it just as hard. The same $3,000 on LinkedIn Ads projects 0.44x against a $60 basket and 6.55x against a $900 contract, on identical delivery. LinkedIn is not a worse platform at the smaller order value, it is the wrong platform for it.
A $3,000 month, split 70/20/10
The same split the calculator produces for any budget you type. Testing is taken as the remainder rather than a third percentage, so the three lines always add up to exactly what you are spending.
| Bucket | Share | Monthly budget | What it buys |
|---|---|---|---|
| Prospecting | 70% | $2,100 | Cold audiences, broad targeting, the ads that decide how big this account can get. Run few enough ad sets that each one clears the learning phase. |
| Retargeting | 20% | $600 | Site visitors, video viewers, cart abandoners. Cheapest revenue in the account and the fastest to exhaust, so watch frequency rather than spend. |
| Testing | 10% | $300 | New creative, new hooks, new audiences. Money you have decided in advance you can lose, because it is the only line that produces next quarter. |
| Total | 100% | $3,000 | On Meta (Facebook and Instagram) benchmarks that month projects 264 conversions at $11 each. |
Benchmarks are estimates blended from public advertising benchmark roundups and platform-published averages, last reviewed 2026-08-03. They are all-industry blends, so your vertical, offer and creative can move every number here. Use them as a planning starting point, not a forecast.
Where these numbers come from
Benchmarks are estimates blended from public advertising benchmark roundups and platform-published averages, last reviewed 2026-08-03. They are all-industry blends, so your vertical, offer and creative can move every number here. Use them as a planning starting point, not a forecast.
Nothing on this page is measured from your ad account or from any private dataset. The benchmarks are public-source estimates, the emailed plan is built from exactly the same figures and the same formula you can read here, and the projection is arithmetic rather than a prediction: it tells you what a month at these assumptions would produce, which is the number worth arguing with before you spend anything.
How to spend the budget once it is split
- Give every ad set enough daily budget to leave the learning phase. Most platforms want somewhere near 50 conversions per ad set per week before they optimise properly, so four ad sets on $600 a month is four ad sets that will each stay in learning forever. Run one.
- Scale in 20% steps, not doubles. A large budget jump resets the learning phase and buys you a week of expensive, unrepresentative delivery. Raise the winner by about a fifth, wait three or four days, and raise it again if the cost per conversion held.
- Cap retargeting by audience size, not by the percentage. Twenty percent is a starting point: if your site had 4,000 visitors last month, $600 against that audience will hit frequency 6 and start annoying people. Spend what the audience can absorb and push the rest into prospecting.
- Keep the testing 10% ring-fenced even in a bad month. It is the only budget that produces next quarter, and it is always the first thing cut when this quarter looks thin, which is exactly how an account runs out of working creative.
- Judge an ad on cost per conversion, not on click-through rate. A high CTR ad that converts at half the rate of a quieter one is more expensive per customer, and the platform will happily spend your whole prospecting budget on it because clicks are what it was asked to buy.
- Set the target cost per conversion from your margin before launch, then write it down. Gross margin times average order value is the most you can pay for a customer and still break even on the first purchase; anything above that has to be justified by repeat business you can actually measure.
Frequently asked questions
How much should I spend on ads per month?
Enough for each active ad set to produce roughly 30 to 50 conversions a month, otherwise nothing you learn is reliable. Work backwards from cost per conversion: at $11 per conversion on Meta benchmarks, 50 conversions costs about $550 a month for one ad set, so a $1,000 month can support one prospecting ad set and a small retargeting one, and not much else. If your product costs more to sell, the floor rises with it.
Why split a budget 70/20/10?
Because the three jobs have different ceilings. Prospecting is the only bucket that can grow, so it gets the majority. Retargeting converts several times better but is capped by how many people you reached last month, so a larger share just raises frequency against the same small audience. Testing is a fixed cost you pay to have something to scale next quarter. The exact numbers are a convention rather than a law: a new account with no audience to retarget often starts nearer 85/5/10, and a brand with heavy organic traffic can justify 60/30/10.
How do I set a target cost per acquisition?
Multiply your average order value by your gross margin. A $60 order at a 40% margin leaves $24, so $24 is the most you can pay for a first purchase and still break even on it. Set the working target below that, commonly at 60% to 70% of the breakeven figure, so the account has room to absorb a bad week. If you know your repeat purchase rate well enough to defend it, you can pay above the first-order breakeven, but only if you are measuring the repeats rather than hoping for them.
When should I increase my budget?
When an ad set has held its cost per conversion for at least a week at its current spend, and the audience is not already saturated. Check frequency before you raise anything: a prospecting audience over about 2.5 impressions per person per week is telling you it is running out of new people, and more money there buys repetition rather than reach. Raise by roughly 20%, hold for three or four days, and repeat while the cost holds.
Do I get the full budget plan?
Yes, by email. The projection and every benchmark on this page are free to read with no address at all. The bucket by bucket split with dollar amounts, the cost per conversion target and the written assumption list unlock when you ask us to send them, and the emailed version uses exactly the same public estimates and the same arithmetic shown here.
Is my data stored?
The calculation runs entirely in your browser and nothing leaves the page while you type. We receive something only if you ask for the plan by email, and then it is your address, the figures you entered and which page you were on, purely so we can deliver it. Marketing email is a separate checkbox that starts unticked.
Run the plan without the spreadsheet
Postify Ads builds and launches campaigns across Meta, Google, TikTok, LinkedIn and Pinterest from one place, and splits a single daily budget across them automatically. The projection above is the plan; this is where it gets spent.
Related free tools
Keep the rest of the checklist close: each tool works on its own, no account needed.
Plan the paid budget, then keep the organic feed running
Postify supports nine platforms in one workflow: plan, schedule, and publish without juggling five different tools.