Smart Bidding in Google Ads: When to Use It and When Not To

Google Ads lets you decide how much to pay for clicks. You can set a fixed amount and hope for the best, or you can let Google's AI take the wheel and adjust your bids automatically. That second option is called smart bidding, and it's where most businesses end up—but many get it wrong.

The problem isn't that smart bidding doesn't work. It does. The problem is timing. Jump in too early, and you'll waste money while Google figures out what it's doing. Wait too long, and you're leaving money on the table by manually adjusting bids yourself.

This guide will help you understand the three main smart bidding strategies, work out if you're ready for them, and know what to expect when you switch over.

What Is Smart Bidding?

Smart bidding means letting Google's algorithms automatically adjust your bids based on the likelihood that a click will turn into what you actually want—a sale, a phone call, a form submission, or whatever your goal is.

Instead of you saying "I'll pay 50p per click for everyone," you're saying to Google: "I want a customer acquisition to cost me £40, and I'll pay whatever makes sense to hit that target." Google then adjusts your bids in real time, paying more for clicks that look valuable and less for clicks that don't.

It sounds brilliant. And when it works, it is. But there's a catch: Google needs data to learn from, and most small businesses don't realise how much data that actually means.

The Three Types of Smart Bidding

Target CPA (Cost Per Acquisition)

Target CPA means you set a maximum cost you're willing to pay per customer or qualified lead.

Let's say you're a plumber in Manchester. You get enough profit from a job to spend £60 acquiring the customer. You tell Google: "My target CPA is £60." Google then bids higher on clicks from people who look like they'll actually call you, and bids lower on clicks from people just browsing.

When it works: You get customers at roughly your target cost, and Google optimises across your campaigns automatically.

The catch: Google needs to see about 50 conversions before it really starts working properly. For many small businesses, that takes weeks or months.

Target ROAS (Return on Ad Spend)

Target ROAS is about profit margin, not just cost.

If you sell something for £100 and it costs you £40 to make, your profit is £60. You might be willing to spend £30 in ads to get that sale (50% return). You tell Google: "I want a 3:1 ROAS" (meaning £3 revenue for every £1 spent). Google then optimises to hit that number across your campaigns.

When it works: You maintain healthy profit margins and scale without losing money.

The catch: This needs even more data than Target CPA—ideally 100+ conversions—because Google needs to understand the actual value of each conversion, not just whether it happened.

Maximise Conversions

Maximise Conversions is simpler: "Get me as many customers as possible within my daily budget."

You set a budget (say, £50 a day) and Google spends all of it trying to get the most conversions it can.

When it works: You're new to ads and just want volume while Google learns what converts.

The catch: It ignores profitability entirely. You might get 10 customers for your £50, or you might get 3. You won't know if they're actually good customers until later.

How Much Data Do You Actually Need?

This is the honest bit: Google says you need 30 conversions to switch smart bidding on. They're being optimistic.

Here's what actually happens:

• 0–30 conversions: Google's making educated guesses. Your costs will be all over the place. One week you pay £25 per customer, the next week £80. It's learning, but poorly.
• 30–100 conversions: Things are stabilising. Google's starting to recognise patterns. Costs become more predictable, but there's still noise.
• 100+ conversions: This is where smart bidding actually starts earning its keep. Google has real data and can make intelligent decisions.

For a local services business (plumber, electrician, roofer), 30 conversions might take 2–3 months. For an e-commerce business with a lower price point, it might be 2 weeks.

The learning period is real, and it's frustrating. During this time, your cost per acquisition will probably be higher than it would be with manual bidding. That's normal. You're paying Google to teach its algorithm about your customers.

When You Absolutely Shouldn't Use Smart Bidding Yet

If any of these apply to you, stick with manual bidding for now:

• You have fewer than 20 conversions per month. You're not generating enough data. Google will take forever to learn.
• Your conversion data is messy or unreliable. If you're not tracking properly (see below), smart bidding will optimise towards rubbish. Rubbish in, rubbish out.
• You can't afford the learning period. If every pound matters and you can't absorb higher costs for 8–12 weeks, wait.
• Your business is super seasonal. Smart bidding learns from your data, but if you only get busy at Christmas, Google won't have learned anything by July.
• You've just launched and have no conversion history. You need a baseline first.

Getting Your Tracking Right (This Is Non-Negotiable)

Before you even think about smart bidding, sort out your conversion tracking.

A conversion is whatever matters to your business. For a plumber, it's a phone call or a booking. For an online shop, it's a purchase. For a consultancy, it's a completed contact form.

Google needs to see this happen reliably, every time. If your tracking is dodgy—missing half your conversions, or counting the same person twice—smart bidding will optimise based on incorrect data.

Check your tracking by:

• Installing Google Ads conversion tags properly (or hire someone to do it if you're not confident)
• Checking your conversion numbers daily for a week—do they match what you actually experienced?
• Making sure you're not double-counting (like counting both a form submission and a phone call as separate conversions if they're the same person)

If you're unsure, talk to someone who knows. BrightClick can audit your tracking and sort it out—it's usually just an hour's work and it'll save you hundreds in wasted ad spend.

The Switch: When You're Ready

Once you've got 50+ reliable conversions under your belt, you can think about switching.

Do it gradually:

1. Keep one campaign on manual bidding as a control 2. Switch a test campaign to smart bidding 3. Run them together for 2–4 weeks 4. Compare costs and quality of leads 5. If smart bidding is better, switch the rest over

Expect the first 2 weeks to be messy. Costs will spike. You'll second-guess yourself. That's Google's learning period. Give it at least 3 weeks before deciding it's not working.

What You Should Do Today

If you're running Google Ads right now:

1. Check how many conversions you've had in the last 30 days. If it's fewer than 30, don't switch to smart bidding yet. 2. Verify your conversion tracking is accurate. Pick a day, count your actual conversions manually, and check it matches Google's number. 3. If you have 50+ conversions: Try switching one small campaign to Target CPA as a test. Set your target CPA to what you're currently paying (roughly), and let it run for three weeks. 4. If you're unsure about any of this: Document your current costs, get them right, then decide.

Smart bidding isn't magic. It's a tool that works brilliantly when you have data and breaks down when you don't. Respecting that boundary is what separates businesses that grow on ads from those that bleed money into them.