The minimum Google Ads budget that actually works in 2026

Google will happily take $10 a day. That is the source of most of the disappointment in small Google Ads accounts: the platform imposes almost no minimum, so people assume there isn't one, then spend six months at a budget that cannot produce the conversion volume the bidding system needs to do its job. The real floor is not set by Google's billing rules. It is set by arithmetic, and you can work yours out in about ten minutes.

Why "minimum budget" is the wrong question

Google's own minimums are almost irrelevant. Search campaigns have no account-level minimum spend at all. The only hard floor Google has introduced recently is narrow: from 1 April 2026, Demand Gen campaigns created through the Google Ads API must carry a daily budget of at least $5, returning a BUDGET_BELOW_DAILY_MINIMUM error below that. Google's stated reason is worth reading closely — the minimum exists to help campaigns get through the cold-start phase with enough spend for the models to learn. That is the platform admitting, in one sentence, what the real constraint is.

The constraint is conversion volume. Every automated bidding strategy is a prediction engine that needs examples. Starve it of examples and it does not fail loudly; it just makes worse predictions, spends unevenly, and produces the pattern every small advertiser recognises: three good weeks, one terrible one, no idea which lever caused either.

So the useful question is not "what is the minimum I can spend" but "what is the minimum that buys enough conversions for the system to learn". That number is specific to you, and it is a multiplication.

The formula

Google Ads documents a minimum of 15 conversions in the past 30 days for Target CPA, and 15 over 30 days for Target ROAS, while recommending you evaluate performance over periods containing at least 30 conversions — 50 for Target ROAS. Practitioner consensus through 2026 lands in the same place: 15 is where the strategy will run, 30 is where it starts behaving predictably, and tROAS with variable order values wants 50 or more before you trust it.

Put that together with your own economics:

That is the whole model. It looks trivially simple because it is, and it is still the calculation almost nobody does before launching. Note the second line: your cost per conversion is not a number Google hands you, it is your click price divided by how well your page converts. Two businesses paying the same CPC can have budget floors that differ by a factor of four.

The floor table

Find your cost per conversion in the left column and read across. The middle columns are monthly budget floors at the three volume thresholds that matter.

The last column contains the shortcut worth memorising. Thirty conversions in thirty days is one conversion per day, so a daily budget roughly equal to your target CPA is the point where Target CPA bidding starts behaving. If your target CPA is $150 and your daily budget is $50, you are asking the system to learn from one conversion every three days. It will not learn much.

Google's own budget guidance for Demand Gen sits higher still — a daily budget of at least ten times your target CPA — which is a different and more demanding bar aimed at campaigns that need reach before they need efficiency. Treat the 1× figure as the floor below which automated bidding degrades, and 10× as what Google would like you to spend on its upper-funnel inventory.

If you have never run ads, estimate the CPA first

The formula needs a cost per conversion, and a business that has not run Search before does not have one. You can estimate it credibly, and the estimate is more useful than any generic "spend $1,000 to start" advice.

Click prices vary enormously by vertical. WordStream's 2026 benchmarks, drawn from more than 13,000 search campaigns measured between April 2025 and March 2026, put the all-industry average CPC at $5.42, with a roughly sixfold spread underneath it: arts and entertainment at $1.63 and restaurants and food at $2.05 at the bottom, attorneys and legal services at $9.87 at the top. The same dataset gives all-industry averages of 6.64% CTR, 8.18% conversion rate and a $66.69 cost per lead.

Multiply your vertical's click price by your page's likely conversion rate and you have your CPA estimate:

Read that table sideways rather than down and you find the thing most budget conversations miss. A legal firm with a 2% landing page pays $500 a lead and needs $15,000 a month to reach stable volume. The same firm, same keywords, same click price, with a 12% landing page pays $83 a lead and needs $2,490. The landing page moves the budget floor by a factor of six. Media budget is the expensive way to buy conversion volume; conversion rate is the cheap way, and it is the lever available to advertisers who cannot fund the floor.

To get your own CPC estimate rather than a category average, use Keyword Planner on the twenty keywords you would actually bid on, in the locations you would actually target, and take the top-of-page bid range rather than the headline average. It runs high, which is the right direction to be wrong in when you are setting a floor.

What changed in 2026, and why old advice is now wrong

Two 2026 changes make underfunded campaigns behave worse than they used to.

The first is the bidding change that landed on 17 August 2026. Google altered how budget-limited campaigns are handled so that budget governs spend and the efficiency target governs efficiency, as two separate levers, instead of the system holding spend down to protect the target. The campaigns most affected were the ones that had been comfortably beating their targets on small budgets — exactly the profile of a well-run small account. If your CPA drifted upward in late August without anything else changing, that is the likely cause; we covered the mechanics and the three responses in what the August 17 bidding change does to your CPA.

The second is structural. As Google consolidates campaign types into AI-driven formats — AI Max absorbing Dynamic Search Ads, automatically created assets and campaign-level broad match from September 2026, Display folding into Demand Gen — the share of your account that depends on model learning rises and the share you can steer manually falls. Automated formats amplify good conversion data and amplify thin conversion data just as faithfully. The budget floor matters more each year, not less.

Below the floor? Six things that work, in order

Most businesses reading this will find their available budget sits below the floor their CPA implies. That is a solvable problem, and the solution is almost never "spend more anyway".

1. Narrow until you clear the floor. Cut the keyword list to the ten terms closest to the transaction, tighten the geography to where you actually convert, drop the hours nobody answers the phone. Spending $1,500 concentrated on one intent beats $1,500 spread across four. Concentration is the only free way to raise conversions per campaign.
2. Fix the landing page before you raise the budget. Per the table above, doubling page conversion rate halves the budget floor. It is the highest-leverage move available and the only one that improves every other channel at the same time. Start with landing pages built for paid traffic rather than sending ads to your homepage.
3. Consolidate campaigns. Conversion data pools at the campaign level for Target CPA. Four campaigns with eight conversions each learn nothing; one campaign with thirty-two learns. Splitting by match type, device or ad group theme is a habit from an era when it bought control. Now it mostly buys fragmentation.
4. Use Maximize clicks or manual CPC below the threshold. If you genuinely cannot reach 15 conversions a month, do not run Target CPA. Buy clicks deliberately, watch search terms weekly, build the negative list, and switch to Smart Bidding when the volume arrives. This is unfashionable advice and it is correct for small accounts.
5. Move the conversion action upstream, carefully. Counting a higher-volume action — a call started, a quote form opened, a booking page reached — gives the model more examples. The risk is that you optimise toward cheap actions that never become customers, which is how cost per qualified lead quietly doubles while cost per lead falls. If you do this, hold the two metrics side by side; our cost per qualified lead model sets out the break-even.
6. Spend the money somewhere else. If your floor is $6,000 a month and your budget is $1,200, Google Ads is not a cheap channel you are under-using. It is an expensive channel you cannot afford yet. Local SEO, a Google Business Profile worth clicking, email, and organic content have lower minimum viable budgets, and there is no prize for losing money on the prestigious channel. A good media plan starts by naming the channel you can afford to run properly, not the one with the best reputation.

Budget floor is monthly. Commitment is quarterly

One more number belongs in the plan. The floor tells you what a month costs; it does not tell you how many months you need before the answer is trustworthy. Smart Bidding re-enters a learning period after significant changes, typically one to two weeks, and a single month of data at the 15-conversion threshold is far too noisy to judge. A reasonable minimum commitment is three times your monthly floor, run without structural changes, before you decide whether the channel works for you.

Advertisers who cannot commit that tend to change bid strategy every three weeks, which resets learning each time and guarantees the disappointing result they were trying to avoid. If three months at the floor is not fundable, that is genuinely useful information: it tells you to go and fix the conversion rate, or pick a different channel, rather than to start and stall.

And once you are above the floor and stable, the constraint changes shape entirely. The question stops being "is there enough data" and becomes "at what point does buying more volume start costing more per customer than it returns" — a different problem with a different answer, which we work through in scaling ad spend without watching CAC explode.

Frequently asked questions

What is the minimum budget for Google Ads in 2026?

Google sets no general minimum spend for Search campaigns; the only recent hard floor is $5 a day for Demand Gen campaigns created through the Google Ads API, in force since 1 April 2026. The practical minimum is set by Smart Bidding's need for conversion volume: Google documents a minimum of 15 conversions in 30 days for Target CPA, and 30 or more for stable performance. Multiply the conversions you need by your cost per conversion to get your own floor. At the 2026 all-industry average cost per lead of $66.69, 30 conversions a month works out at roughly $2,000.

Is $500 a month enough for Google Ads?

It is enough only if your cost per conversion is under about $33, which in practice means a low-competition local vertical with a landing page that converts well. At $500 a month with a $100 cost per conversion you buy five conversions, a third of what Target CPA needs to run at all. If you are in that position, run Maximize clicks or manual CPC on a tightly narrowed campaign rather than a Smart Bidding strategy that cannot learn, and put the effort into landing page conversion rate.

How much does Google Ads cost per click in 2026?

WordStream's 2026 benchmarks, based on more than 13,000 search campaigns measured from April 2025 to March 2026, put the all-industry average cost per click at $5.42. The spread by industry is close to sixfold: arts and entertainment averaged $1.63 and restaurants and food $2.05, while attorneys and legal services averaged $9.87. The same dataset reports all-industry averages of 6.64% click-through rate, 8.18% conversion rate and $66.69 cost per lead.

How many conversions does Smart Bidding need per month?

Google's documented guidance is at least 15 conversions in the past 30 days for Target CPA and 15 over 30 days for Target ROAS, while recommending that performance be evaluated over periods containing at least 30 conversions, or 50 for Target ROAS. In practice, 15 is the point where the strategy will run, 30 is where it becomes reasonably predictable, and Target ROAS with variable order values needs 50 or more before the targets mean much.

Does a bigger budget lower your cost per click on Google Ads?

No. Budget controls how much you spend, not what you pay per click; auction price is driven by competition, Ad Rank and quality. What a bigger budget does buy is conversion volume, which improves the bidding model's predictions and can lower cost per conversion over time. This distinction matters after Google's 17 August 2026 bidding change, which separated the two levers more explicitly: budget governs spend, the efficiency target governs efficiency.

How should a small business split a limited budget across campaigns?

Concentrate it. Conversion data pools at the campaign level for Target CPA, so four campaigns with eight conversions each learn nothing while one campaign with thirty-two learns properly. Run one tightly themed campaign on the keywords closest to the transaction, in the locations that actually convert, and only split out a second campaign once the first is clearing 30 conversions a month on its own.

What should I do if my budget is below the floor?

Work through it in order: narrow the campaign until the concentrated budget clears the floor, fix the landing page conversion rate (doubling it halves the floor), consolidate fragmented campaigns, drop to Maximize clicks or manual CPC while volume is low, consider counting a higher-volume upstream conversion action while watching lead quality, and if none of that closes the gap, put the money into channels with lower minimum viable budgets until your economics change.

The takeaway

Google's lack of a minimum spend is not evidence that small budgets work; it just moves the failure somewhere harder to see. Your floor is conversions needed times cost per conversion, which at 30 conversions a month usually means a daily budget close to your target CPA. Work out that number before you launch, then commit to three months of it or spend the money elsewhere. And notice which side of the equation is actually cheap to move: media budget buys conversion volume at full price, while a landing page that converts twice as well halves the floor permanently and improves every other channel while it is at it.