November is when many businesses lock in next year’s budget. And the Google Ads budget usually gets decided in one of two ways: last year’s figure plus a percentage, or whatever’s left once everything else has been allocated. Neither answers the question that matters: how much can your business spend on Google Ads and still make money, and how much does it need to spend for the campaigns to work at all?
This post is the method I use to set a Google Ads budget for a full year. It isn’t a pricing article. What a click or a lead costs by industry is covered in how much Google Ads costs. Here we start from your margins and your demand and work towards an annual figure, split by month, with a slice set aside for testing and clear rules for when to raise or cut it.
In 30 seconds:
- Your margins set the ceiling: work out your maximum CPA or break-even ROAS before you open any tool.
- Demand sets the size: Keyword Planner tells you how many searches there are and roughly what advertisers pay for them.
- The algorithm sets the floor. Google recommends evaluating Target CPA over at least 30 conversions in 30 days, and a Performance Max daily budget of at least 3 times your CPA.
- Since August 2026, a budget-limited campaign on Target CPA or Target ROAS bids towards that target. If you leave it loose and count on the campaign to beat it, it will spend to reach it: budget with the target you actually set.
- The monthly split comes from your own seasonality, not from dividing by twelve.
- Keep a separate testing pot, and allow 1 to 2 conversion cycles before you judge any budget change.
The 6-step method at a glance
| Step | Question | What you get |
|---|---|---|
| 1 | How much can I pay for a conversion? | Maximum CPA or break-even ROAS |
| 2 | How much demand is there, and at what price? | Possible clicks, cost and conversions per month |
| 3 | What’s the minimum the bidding needs to learn? | A budget floor per campaign |
| 4 | How does it spread across the year? | Monthly budget based on seasonality |
| 5 | How much do I keep for testing new things? | A testing pot separate from the core budget |
| 6 | When do I change it? | Rules for raising or cutting during 2027 |
The order matters. Start at step 2 and you end up with a budget that soaks up demand without knowing whether it’s profitable. Start at step 4 and you’re splitting a number nobody has justified.
Step 1: how much can you pay for each conversion?
Before you open Google Ads, open your margin sheet. A Google Ads budget has an economic ceiling, and you can work it out with two simple formulas:
- Maximum CPA (ecommerce): average order value × gross margin. That’s what each order leaves you before advertising.
- Break-even ROAS: 1 ÷ gross margin. With a 40% margin, break-even ROAS is 2.5 (250%). Below that, every sale costs you money.
- Maximum cost per lead (services and B2B): gross profit per customer × lead-to-customer close rate.
A hypothetical example with round numbers. A shop with a 90 € average order and a 45% gross margin makes 40.50 € per order before advertising. That’s its break-even CPA. If it wants every order to leave at least 13.50 € of profit, its target CPA is 27 €, which works out to a target ROAS of 3.33 (90 ÷ 27).
Services work the same way. A 3,000 € project at a 30% margin leaves 900 €. If you close one in five leads, you can pay up to 180 € per lead without losing money. See why a 120 € CPL can be excellent for one business and ruinous for another?
If what you sell repeats (subscriptions, consumables, customers who come back), you can use 12-month customer value instead of the first order. Only if you’ve got data to back it. I’d rather start with the first order and raise the target once repeat purchases are measured, not assumed. The ROAS & CPA calculator gives you these figures in a minute.
Step 2: how much demand is there, and what does it cost to capture?
With the ceiling clear, it’s time to size the market. The tool here is Google Ads Keyword Planner, and it’s worth knowing what each column shows, because it mixes two very different kinds of data (Google Ads Help):
| Metric | What it is | What I use it for |
|---|---|---|
| Avg. monthly searches | 12-month average by default, includes close variants | Market size and seasonality |
| Top of page bid (low range) | Roughly the 20th percentile of what advertisers have paid | Indicative CPC, cautious scenario |
| Top of page bid (high range) | Roughly the 80th percentile | Indicative CPC on contested keywords |
| Plan forecasts | Daily clicks, impressions and cost based on your bid and budget | Realistic spend scenario |
Google spells out the key difference on that same page: forecasts take bid, budget and seasonality into account, and historical metrics don’t. And forecasts are based on one week of data, averaged per day. They’re a snapshot, not a film. That’s why I use them for sizing and never as a promise.
My process for this step:
- List only keywords with buying or hiring intent. Informational terms inflate the volume and don’t sell.
- Build a plan with those keywords and the bid you’d actually use, then check the monthly click and cost forecast.
- Apply your real conversion rate, from your account or your site. If you don’t have one, use a cautious figure and label it as an assumption.
- Divide cost by conversions. If the resulting CPA is above your step 1 maximum, the problem isn’t the budget. It’s the offer, the site or the keywords you picked.
If the account already has history, Performance Planner is a better starting point for existing campaigns. Its forecasts refresh daily and are based on the last 7 to 10 days, adjusted for seasonality, and it only includes eligible campaigns, such as those that haven’t changed bid strategy in the last 10 days and have at least one conversion (Google Ads Help).
What about industry benchmarks?
They’re useful for checking that your assumptions aren’t detached from reality, not for planning. The most quoted report is LocaliQ’s, published with WordStream and updated on 1 June 2026: average CPC of $5.42, average conversion rate of 8.18% and average cost per lead of $66.69 on search, drawn from thousands of campaigns across more than 20 industries (LocaliQ, 2026 Search Advertising Benchmarks).
Two warnings before you use those figures. The report doesn’t say which countries the accounts are in, and LocaliQ is a US company, so don’t treat it as a benchmark for UK or European markets. And the average hides huge ranges: in the same report, lawyers pay $9.87 per click and travel pays $2.14. If your forecast CPC is in the same order of magnitude as your industry’s, fine. If it’s ten times lower, check the assumption.
Click prices vary a lot by sector and country. As a first-party reference, in the accounts I manage in Spain and Europe the average Search CPC between September 2025 and August 2026 was €0.64, with a per-account range of €0.43 to €0.87; the detail and its limits are in how much Google Ads costs.
Step 3: what’s the minimum budget for Smart Bidding to learn?
This is the step people skip most often, and the one that wrecks the most budgets. A budget can be profitable on paper and still too small for automated bidding to have any data to work with.
Google doesn’t set a universal minimum, but it does give concrete reference points by strategy and campaign type:
| Strategy or campaign | Google’s guidance | Source |
|---|---|---|
| Target CPA | Can start with no history; for evaluation, the last 30 days with at least 30 conversions | Google Ads Help |
| Target ROAS | Report values for 4 weeks or 3 conversion cycles (whichever is longer) before setting the target | Google Ads Help |
| Performance Max | Average daily budget of at least 3 times your CPA | Google Ads Help |
| Demand Gen with conversion bidding | Budget of at least 20 times the average target CPA | Google Ads Help |
In euros, using the 27 € target CPA from the example:
- To evaluate Target CPA on 30 conversions a month, the campaign needs to spend about 810 € a month (30 × 27 €).
- For Performance Max, 3 × 27 € is 81 € a day. At 30.4 days a month, that’s about 2,460 € a month for a single campaign.
What do I do with this? I count how many campaigns the budget can genuinely support. If step 2 says profitable demand is worth 1,600 € a month, I don’t build one Search campaign, one Performance Max and a Demand Gen on top. I build one well-fed campaign and let it learn. Spreading a small budget across many campaigns is the fastest way to make sure none of them ever gets enough data. I’ve covered when automated bidding pays off and when it doesn’t in the practical Smart Bidding guide, and how long each phase takes in how long Google Ads takes to work.
What changed in August 2026 for budget-limited campaigns?
On 17 August 2026, Google started rolling out a change to target-based bid strategies across all accounts: when a campaign is limited by budget, bids are now optimised more consistently to its target (Google Ads Help). It affects Target CPA and Target ROAS in Search, Shopping, Performance Max, Demand Gen and Travel campaigns. Google says the change doesn’t directly increase spend, and daily and monthly budget limits still apply.
Google’s own recommendation gives away what it means: it asks you to review above all the budget-limited campaigns that were beating their target, because they now bid towards that target. The options it gives are updating the target to recent performance, setting a different one, or switching to Maximize conversions or Maximize conversion value.
For 2027, the rule is simple: the target CPA you set is the CPA you should budget with, not an optimistic figure the campaign will beat on its own.
Step 4: how do you split the Google Ads budget by month?
Dividing the annual budget by twelve is convenient and almost always wrong. Demand isn’t flat, and your budget shouldn’t be either.
For the split I use, in this order:
- Your own account’s conversions by month over the last 12 to 24 months. It’s the best data you have.
- Seasonality from Keyword Planner and Google Trends if there’s no history. I walked through the full Trends workflow in how to calculate your budget with Google Trends.
- Your trading calendar: launches, sales, trade shows, end of season.
From that you give each month a weight. A normal month is 1; a quiet one, 0.7; November for an online shop might be 1.6. The month’s budget is the base budget times its weight.
Within each month, two technical details are worth keeping in mind. Daily budget is the monthly figure divided by 30.4, which is the monthly cap: a campaign can spend up to twice its average daily budget on a given day, but by the end of the month it won’t go over 30.4 times that budget (Google Ads Help). And for short peaks, like Black Friday or a three-day promotion, Google offers seasonality adjustments, designed for events of 1 to 7 days, which may not work as well if used for more than 14 days at a time (Google Ads Help).
There’s a useful change for 2027. Since January 2026, campaign total budgets have been in open beta for Search, Performance Max and Shopping: you set an amount for a period of a few days to a few weeks and Google paces spend to use it by the end date (Google, 15 January 2026). For campaigns with a fixed start and end date, it saves me adjusting the daily budget every other day. If you’re also splitting spend between Google and Meta in the last quarter, I covered that separately in how to split your Q4 budget between Google and Meta.
Step 5: how much should you keep for testing versus scaling?
The annual budget has two parts, and it pays to keep them apart from day one, both in the spreadsheet and in your head:
| Core budget | Testing pot | |
|---|---|---|
| What it’s for | Campaigns already delivering CPA or ROAS within target | New campaigns, campaign types, markets or keywords |
| How it’s judged | Against the step 1 target, every month | Against one specific question, at the end of the test |
| If it fails | It gets adjusted | It gets closed and the learning gets written down |
| Size | Most of the money | Whatever it takes to reach a conclusion |
I usually set aside 10 to 20% of the total for testing. That isn’t a Google figure: it’s my own rule so learning doesn’t eat into what already works. What matters more is the size of each test. A test that never reaches 30 conversions tells you nothing. With a 27 € CPA, each campaign-type test needs at least 810 € or so for the comparison to stand up. If the testing pot is 2,000 €, there’s room for two proper tests, not six half-finished ones.
For tests on existing campaigns, I use Google Ads experiments, which the Target CPA help page itself recommends for comparing new settings against the original campaign (Google Ads Help).
Step 6: when should you raise the budget, and when should you cut it?
Your 2027 budget isn’t a fixed number until December. It’s a starting point with rules for moving it. These are mine:
| Signal | What it usually means | What I do |
|---|---|---|
| CPA or ROAS on target and high impression share lost to budget | There’s profitable demand you’re not buying | Raise the budget in steps |
| As you raise it, marginal CPA gets worse month after month | You’re buying less profitable clicks | Go back to the previous step |
| CPA above break-even for 2 conversion cycles, with tracking validated | The campaign isn’t profitable at this level | Cut or pause, then review the offer and site |
| Sudden drop in conversions with no changes on your side | Could be broken tracking, not performance | Check tracking before touching the budget |
| Planned seasonal peak | Conversion rate will shift for a few days | Seasonality adjustment or campaign total budget |
The metric in the first row is Search lost impression share (budget): the percentage of time your ads weren’t shown on the Search Network because the budget ran short (Search Ads 360 Help). If that figure is high and CPA is fine, budget is the brake.
Google separates that loss from the share lost to Ad Rank (Google Ads Help), and the distinction changes the decision. If a profitable campaign loses a good share to budget, spending more should bring more volume at a similar cost. If it loses it to rank, raising the budget barely changes anything; you need to improve ads, landing page or bids first. And if you lose little to either, you’re already close to the ceiling: spending more on those same searches just makes each conversion more expensive.
The timing rule is the hardest one to stick to. Google recommends you allow 1 to 2 conversion cycles for the system to stabilise after major changes to bidding or budget before adjusting again, and evaluate over at least two full cycles or 50 conversions (Google Ads Help). If your customers take two weeks to buy, judging a budget increase after four days means judging noise. That’s why I raise in moderate steps and wait, rather than doubling overnight.
A full example: the 2027 budget on one sheet
Putting the six steps together with the hypothetical shop from earlier (90 € average order, 45% margin, 27 € target CPA):
- Ceiling: break-even CPA of 40.50 €, target of 27 € (target ROAS of 3.33).
- Demand: the Keyword Planner plan forecasts about 1,800 clicks a month at a 0.90 € CPC, roughly 1,620 €. At a 3% conversion rate, that’s 54 orders and a 30 € CPA. Slightly above target, so 2027 work starts with the site and the offer, not with spending more.
- Floor: 54 conversions a month clears the 30-conversion reference for a Target CPA campaign. Performance Max would need about 2,460 € a month, more than demand justifies, so the shop starts with a single Search campaign.
- Split: with weights of 2.6 in Q1, 3.0 in Q2, 2.6 in Q3 and 4.0 in Q4 (12.2 in total), the annual core budget is 19,764 € (1,620 € × 12.2). The weights add up to 12.2 rather than 12 because the Q4 peak brings demand that doesn’t exist the rest of the year; it isn’t just money moved from one month to another. And while CPA stays at 30 €, above the 27 € target, the core budget doesn’t grow beyond this split: it scales once CPA comes down to target.
| Quarter | Weight | Core budget |
|---|---|---|
| Q1 | 2.6 | 4,212 € |
| Q2 | 3.0 | 4,860 € |
| Q3 | 2.6 | 4,212 € |
| Q4 | 4.0 | 6,480 € |
| Total | 12.2 | 19,764 € |
- Testing: an extra 15%, about 2,965 €, for two or three tests of at least 810 € each.
- Approved total: about 22,700 € for 2027, with a quarterly review against the step 6 table.
The weights and conversion rate in this example are made up to show the maths. Yours need to come from your account. To run the CPA, target conversions and CPC part in one place, use the campaign budget calculator.
When does it make sense to get help setting the budget?
If you’ve got clean history, clear margins and one or two campaigns, you can apply this method in-house in an afternoon.
I’d bring in outside help in three cases. When tracking isn’t reliable, because the whole calculation rests on real CPA and a badly measured CPA gives you a badly calculated budget. When you have to choose between several campaign types on a budget that can’t fund them all. And when the budget needs sign-off from someone who wants the reasoning in writing.
In those cases I work as a Google Ads consultant: I check the tracking, run the six steps on your account data and hand over the sheet with the monthly budget and the adjustment rules.
Frequently asked questions
What’s the minimum Google Ads budget I need for 2027?
There’s no official minimum to advertise, but there is one for automated bidding to work with real data. Google recommends evaluating Target CPA over at least 30 conversions in 30 days, so the practical minimum for one campaign is roughly 30 times your CPA per month. For Performance Max, at least 3 times your CPA every day.
How do I calculate a Google Ads budget from ROAS?
Start with break-even ROAS: 1 divided by your gross margin. Set a target ROAS above that point, turn it into a CPA (average order value divided by target ROAS) and multiply that CPA by the conversions demand allows you to get each month.
Is a fixed or variable monthly budget better?
Variable, if your business is seasonal. Give each month a weight based on your historical conversions and split the annual budget using those weights. A flat budget overspends in the quiet months and runs short in the strong ones.
How long should I wait before judging a budget increase?
Google recommends allowing 1 to 2 conversion cycles after a major bidding or budget change before adjusting again. If your sales cycle is a week, wait at least two weeks before drawing conclusions.
Are CPC benchmarks useful for setting my budget?
Only as a sanity check. LocaliQ’s 2026 averages ($5.42 CPC, 8.18% conversion rate) cover more than 20 industries and don’t state a country. For planning, use Keyword Planner forecasts with your own keywords and your real conversion rate.
A budget you can defend
A good Google Ads budget for 2027 isn’t the highest or the lowest. It’s the one you can explain line by line: how much you can pay per customer, how much demand there is, what the algorithm needs, how it spreads across the year and what you’ll do if the numbers move. With that in hand, the conversation with leadership stops being “how much do we put in?” and becomes “which scenario do we pick?”.
If you’re finalising your 2027 budget and want to go over the numbers with someone who does this every year, book a 45-minute call and we’ll work through it with your account data.
Sources
- Google Ads Help. About Keyword Planner forecasts. support.google.com/google-ads/answer/3022575
- Google Ads Help. About Performance Planner. support.google.com/google-ads/answer/9230124
- LocaliQ. 2026 Search Advertising Benchmarks. Updated 1 June 2026. localiq.com/blog/search-advertising-benchmarks
- Google Ads Help. About Target CPA bidding. support.google.com/google-ads/answer/6268632
- Google Ads Help. About Target ROAS bidding. support.google.com/google-ads/answer/6268633
- Google Ads Help. Step 5: Set your budget (Performance Max). support.google.com/google-ads/answer/15863846
- Google Ads Help. Demand Gen campaign performance fluctuations: Budget settings. support.google.com/google-ads/answer/14715995
- Google Ads Help. About overdelivery and your average daily budget. support.google.com/google-ads/answer/1704443
- Google Ads Help. About seasonality adjustments. support.google.com/google-ads/answer/10369906
- Google. Campaign total budgets are now available in Search, Performance Max and Shopping campaigns. 15 January 2026. blog.google/products/ads-commerce/campaign-total-budgets
- Search Ads 360 Help. Get impression share data. support.google.com/sa360/answer/9633436
- Google Ads Help. About impression share metrics. support.google.com/google-ads/answer/7103314
- Google Ads Help. Frequently asked questions about changes to Target-based bid strategies. support.google.com/google-ads/answer/17125145
- Lionel Fenestraz. How much Google Ads costs (first-party CPC data, September 2025 to August 2026). lionelz.com/en/blog/how-much-does-google-ads-cost
Lionel Fenestraz is a freelance Google Ads and Meta Ads consultant based in Spain, working with ecommerce, SMB and B2B advertisers. Google Partner and CXL certified.


