How much should you spend on Google Ads?

The only truly bad budget is the one that can’t teach you anything. I’ve lost count of how many accounts I’ve opened where someone said, “We tried Google Ads. Didn’t work.” Then I look and see £300 dripped across six campaigns, ten keywords apiece, one broad match free-for-all, and a homepage as the landing page. That’s not a verdict on Ads; that’s a verdict on starving a test.

I budget from the unit economics up. Not dashboards. Not hunches. Just facts. If you're here looking for how much you should spend on Google Ads today, then the guide below should help.

Start with economics, not ad settings

Before I touch Google Ads I write five numbers on a pad:

• Average order value or revenue per customer

• Gross margin (or the % of revenue you can afford to spend on acquisition)

• Lead-to-sale close rate

• Website conversion rate (visitor → lead)

• Rough CPC range for the intent you want (you can pull this from past data, Keyword Planner, or common sense and a small test)

From those, I back into a target cost per lead (CPL) and, if you sell online, a target CPA/ROAS.

Example:

• You sell a service for £1,000 with 50% gross margin → £500 contribution

• Sales closes 25% of qualified leads

• You can afford to spend £200 per sale on ads (leaving room for overhead and profit)

• That means you can afford £50 per lead (because 1 in 4 leads becomes a sale)

Now we have a real target. The budget conversation gets easy.

Turn targets into a budget you can actually run

Let’s say your target CPL is £50. Your landing page converts at 5%. That means you need 20 clicks to generate one lead. If the average CPC is £2.50, then one lead costs £50. Good.

How many leads do you want in a month? If the answer is 20 leads, your monthly ad budget is 20 × £50 = £1,000. Daily that’s roughly £33.

You can do this with any numbers:

• If conversion rate is 3% and CPC is £3, your effective CPL is £100 (33 clicks × £3)

• If conversion rate is 8% and CPC is £2, your CPL is £25 (12–13 clicks × £2)

The dials you actually control are conversion rate and click quality. Budget follows from them, not the other way around.

Minimum budgets that don’t waste your time

Google’s algorithms need signal. Humans do too. If you run so thin that you only get a trickle of clicks, you’ll never know what works.

My rule of thumb:

• Aim for at least 10–20 clicks per day on your main ad group so you learn something every week

• If your CPC is £2–£3, that’s £20–£60/day

• If your CPC is £5–£8 (legal, finance, B2B niches), £50–£160/day is more realistic for non-brand

That’s not me trying to spend your money. It’s me trying to get you out of limbo where you burn £300 a month and learn nothing.

What if I only have £500 this month?

Still workable, just narrow the scope:

• One service

• One location

• One intent (high intent, e.g., “emergency plumber cambridge” not “plumbing tips”)

• Exact/phrase match only, with a tight negative list

• One purpose-built landing page with a phone number, simple form, and proof

You won’t conquer the market, but you can buy clean signal. If it bites, you scale next month.

What to set aside for testing

I split budgets 70/20/10:

• 70% proven core (your highest-intent campaign)

• 20% controlled tests (new keywords, new ad angles, new landing page)

• 10% exploration (discovery, broad match with tight negatives, or a single Performance Max test if you have clean conversion tracking)

This stops “test” spend from eating the farm, but keeps real progress coming.

When to raise or cut spend

Raise when:

• You’re hitting or beating target CPL/CPA and Search Impression Share lost to budget is >20% on a converting campaign

• You’re capped by budget most days and the landing page is holding 5–10% conversion consistently

• You’ve got at least 30 conversions per month per campaign and Smart Bidding is stable

Cut when:

• CPL has drifted 25–30% above target for two weeks in a row

• You’re buying brand clicks you’d get anyway (protect brand only if competitors are bidding on it or you need the sitelinks)

• Tracking is broken—pause until it’s fixed; otherwise you’re flying blind

Don’t spend a pound until these are in place

• Conversion tracking that maps to real outcomes (form submits, calls with duration, e-comm revenue). Import offline conversions if sales happen by phone or later in the CRM.

• A landing page that matches intent. If the query is “pricing”, show pricing signals. If it’s “emergency”, show phone first.

• Speed on mobile. If it takes 5–6 seconds to load you are paying a tax on every click.

• Negative keywords from day one. Add the obvious junk terms before the first impression.

How I’d budget the first 90 days

Weeks 1–2 — Set up and sanity

• Tracking in GA4 + Google Ads, call tracking turned on, conversion values set

• Build one high-intent search campaign with 5–15 tightly themed keywords

• Write three ad variants and ship one clean landing page

• Start at the bottom end of your calculated budget; prove the plumbing works

Weeks 3–4 — Prove the unit economics

• Watch search terms daily; add negatives; tighten matches

• Fix obvious landing-page leaks (headline, form length, proof)

• If you’re within 10–20% of target CPL, hold; if you’re miles off, fix page/message before adding budget

Weeks 5–8 — Add controlled tests

• Duplicate the landing page with a different hero and offer; 50/50 the traffic

• Expand keywords cautiously; try phrase → exact, not the other way round

• Consider Smart Bidding only once you have 15–30 conversions in the last 30 days in that campaign

Weeks 9–12 — Scale what’s working

• Increase budget on the winners by 10–20% at a time; avoid big jumps

• If impression share is capped by budget and CPL is on target, keep nudging up

• Add one remarketing list with simple creative that pushes to the same winning page

At the end of each month, I write a one-pager: what moved, what stalled, what we’re changing. If your budget can’t be tied to a change on that page, we’re spending for the sake of it.

Common budget killers I see

• Too many campaigns. You split a small budget across five ideas and starve them all. One solid campaign beats five anaemic ones.

• Bidding on everything. Broad match across vague terms with no negatives is a money shredder.

• Sending to the homepage. Every extra click to find the form is a leak you paid for.

• Chasing position instead of profit. Top of page looks great in a report; second position with half the CPC often wins the month.

So… how much should you spend on Google ads?

Use the math, not a moral stance about “big” or “small” budgets.

1. Set a target CPL/CPA from your real economics.

2. Estimate CPC and conversion rate for the intent you want.

3. Back into a monthly budget that can buy enough clicks to learn and enough conversions to judge.

4. Start at the low end, fix the funnel, then raise with proof.

If you want a quick sense check, send me your target, your best guess at CPC, and your conversion rate. I’ll tell you what budget I’d set for month one and where I’d spend the first pound. Then you can run with it, or ask me to do it. Either way, the goal is the same: buy clean signal, learn fast, and put money only where the math works.