Google Ads Bidding Strategies Explained: Which One Should You Use?

Why Bidding Strategy Matters

Your bidding strategy tells Google how to spend your money. Get it right and the algorithm works for you, finding the right people at the right price. Get it wrong and you’re either overpaying for clicks, missing conversions, or handing Google a blank cheque to spend however it likes.

The problem is that most advertisers either pick a strategy based on a Google recommendation (which optimises for Google’s revenue, not yours) or pick the most advanced option before they have the data to support it. Smart bidding needs data. Without enough conversions, the algorithm is guessing — and it guesses expensively.

The right bidding strategy depends on one thing: how much conversion data you have. Everything else follows from that.

The Strategies Ranked by Data Requirements

Here’s every bidding strategy Google offers, ordered from least data required to most. Think of it as a progression — you start at the top and work your way down as your account matures.

Manual CPC

What it does: You set the maximum cost per click for each keyword yourself. Google won’t spend more than your bid. You have complete control over what you pay.

When to use it: When you’re just starting out, launching a new campaign in an unfamiliar market, or when you need absolute control over costs. Also useful for low-volume accounts where automated strategies don’t have enough data to learn.

Pros:

• Total control over every keyword’s bid
• No data requirements — works from day one
• Good for understanding what clicks actually cost in your market
• Forces you to learn the fundamentals

Cons:

• Time-intensive — you’re adjusting bids manually
• Can’t react to real-time signals like device, location, time of day, or audience
• You’ll almost certainly leave money on the table once you have enough data for automation

The honest take: Manual CPC is underrated. Google wants you on automated bidding because it typically increases spend. But manual CPC teaches you what things cost, and it’s the only strategy where you know exactly what you’re paying for each click. Use it to build your baseline, then move on when you have the data.

Enhanced CPC (ECPC)

What it does: You still set manual bids, but Google can adjust them up or down based on how likely a click is to convert. It uses your conversion data to make these adjustments.

When to use it: When you have some conversion data but not enough for fully automated bidding. Think of it as manual CPC with training wheels for automation.

Pros:

• More control than fully automated strategies
• Google can increase bids for high-converting opportunities
• Lower risk than jumping straight to smart bidding

Cons:

• Google removed the cap on how much it can increase bids, so costs can spike
• Still requires manual bid management as a foundation
• Neither fully manual nor fully automated — sometimes the worst of both worlds

The honest take: ECPC used to be a solid middle ground. Since Google removed the bid cap, it’s become less predictable. It can work well as a transitional strategy, but many advertisers now skip it entirely, going from manual CPC straight to maximise conversions once they have the data.

Maximise Clicks

What it does: Google automatically sets bids to get as many clicks as possible within your daily budget. You can set a maximum CPC cap to prevent overspending on individual clicks.

When to use it: When you’re building initial traffic and data, particularly for brand campaigns or when you need to collect conversion data quickly to move to a smarter strategy.

Pros:

• Simple to set up — just set a budget and go
• Good for generating the click volume needed to collect conversion data
• Works well for brand campaigns where most clicks convert anyway

Cons:

• Google optimises for volume, not quality — you’ll get cheap clicks, but they might not be the right clicks
• Can blow through budget on low-intent searches
• No consideration for conversion likelihood

The honest take: Maximise clicks is a data collection strategy, not a performance strategy. Use it with a CPC cap to prevent Google from overspending on individual clicks, collect your data, and then switch to something conversion-focused. Don’t leave campaigns on maximise clicks indefinitely — that’s how budgets get wasted.

Maximise Conversions

What it does: Google automatically sets bids to get as many conversions as possible within your daily budget. The algorithm uses machine learning to predict which clicks are most likely to convert and bids accordingly.

When to use it: When you have at least 15–30 conversions per month in the campaign (not the account — the specific campaign). Your conversion tracking must be set up correctly and tracking meaningful actions, not page views or button clicks.

Pros:

• Genuinely effective when it has enough data
• Reacts to real-time signals that you can’t manually adjust for
• Saves significant management time
• Often discovers conversion opportunities you would have missed

Cons:

• Will spend your entire daily budget — no cost efficiency consideration
• Can be aggressive in the learning period (first 1–2 weeks)
• If your conversion tracking is wrong, the algorithm optimises for the wrong things
• Doesn’t care about cost per conversion — just total conversions

The honest take: Maximise conversions is the workhorse strategy for most accounts. But ‘maximise conversions’ means exactly that — Google will spend whatever it takes to get more conversions, even if some cost three times what they should. If cost per conversion matters (and it always should), add a target CPA or move to that strategy instead.

Target CPA (Cost Per Acquisition)

What it does: You tell Google how much you want to pay per conversion, and the algorithm adjusts bids to hit that target on average. Some conversions will cost more, some less, but it aims for the average you set.

When to use it: When you know your target cost per conversion and have at least 30 conversions per month. You need enough historical data for Google to understand what a conversion looks like in your account.

Pros:

• Controls cost per conversion, not just volume
• Balances volume with efficiency
• The algorithm gets better over time as it collects more data
• Good for lead generation businesses with clear conversion values

Cons:

• Set the target too low and Google will stop spending entirely
• Set it too high and you overpay for conversions
• Needs stable conversion data — seasonal fluctuations can throw it off
• Doesn’t account for conversion value differences

The honest take: Target CPA is where most non-e-commerce accounts should eventually land. Start with maximise conversions to collect data and understand your actual CPA, then switch to target CPA with a realistic target based on historical performance. Don’t set an aspirational CPA — set one based on what you’ve actually achieved, then gradually tighten it.

Target ROAS (Return on Ad Spend)

What it does: You tell Google the return you want on your ad spend (as a percentage), and the algorithm adjusts bids to hit that target. A target ROAS of 400% means you want £4 in revenue for every £1 spent.

When to use it: E-commerce accounts with revenue tracking set up properly. You need at least 50 conversions per month with revenue values for this to work well. The algorithm needs to understand not just what converts, but what converts at high value.

Pros:

• Optimises for revenue, not just conversion volume
• Prioritises high-value conversions over low-value ones
• Ideal for e-commerce with varied product prices
• Can significantly improve profitability when it has enough data

Cons:

• Requires accurate revenue tracking — garbage data in, garbage bids out
• High data requirements — needs significant conversion volume to work
• Can dramatically reduce traffic if the target is set too aggressively
• Doesn’t account for margins — a £100 sale with 10% margin is treated the same as a £100 sale with 50% margin

The honest take: Target ROAS is the gold standard for e-commerce. But it only works if your revenue tracking is bulletproof. If your conversion values are wrong, the algorithm is optimising towards fiction. Get your tracking right first, run on maximise conversion value to collect data, then add a ROAS target. Most e-commerce accounts should aim for this as their end state.

Maximise Conversion Value

What it does: Similar to maximise conversions, but instead of maximising the number of conversions, it maximises the total conversion value (revenue) within your budget. Google prioritises high-value conversions over low-value ones.

When to use it: E-commerce accounts that are ready for value-based bidding but don’t yet have enough data for target ROAS, or accounts where you want Google to go after the highest-value opportunities without a specific return target.

Pros:

• Focuses on revenue, not just conversions
• Good stepping stone to target ROAS
• Lets Google find high-value customers you might miss

Cons:

• Will spend your entire budget — no efficiency constraint
• Still requires accurate revenue tracking
• Can overspend to chase high-value conversions

The honest take: Think of maximise conversion value as the data collection phase for target ROAS. Use it to build up conversion value data, understand your actual ROAS, then add a target. Don’t run it indefinitely without a return target, or Google will spend aggressively to chase revenue regardless of profitability.

The Progression

Here’s how to move through bidding strategies as your account matures. This isn’t theoretical — it’s what we implement for most new accounts.

Week 1–4: Manual CPC or Maximise Clicks (with CPC cap)

Goal: collect data. Understand what clicks cost, which keywords drive traffic, and start building conversion history. Set up conversion tracking properly before you start spending. Use negative keywords aggressively to keep traffic relevant.

Month 2–3: Maximise Conversions (or Maximise Conversion Value for e-commerce)

Once you have 15–30 conversions per month, switch to maximise conversions. Let Google’s algorithm take over bid management while you focus on ad copy, landing pages, and keyword expansion. Expect a learning period of 1–2 weeks where performance may fluctuate.

Month 3+: Target CPA or Target ROAS

Once you understand your actual cost per conversion or ROAS, add a target. Start with a target that’s achievable based on historical data — typically your average CPA or ROAS from the previous 30 days. Then gradually tighten it by 10–15% at a time, giving the algorithm 2 weeks to adjust between changes.

Ongoing: Portfolio Bid Strategies

Once you have multiple campaigns with the same goal, consider portfolio bid strategies. These let Google optimise across campaigns rather than within individual ones, sharing learnings and balancing spend between campaigns to hit your overall target. Particularly useful for accounts with many campaigns that individually have low conversion volume but collectively have enough data.

Common Mistakes

1. Using smart bidding without conversion tracking. This is the most common and most expensive mistake. If you switch to maximise conversions but you’re tracking page views instead of actual leads or sales, Google will optimise for page views. You’ll get lots of cheap traffic that does absolutely nothing. Set up proper conversion tracking first. Always.

2. Setting an unrealistic target CPA. If your average CPA is £50, don’t set a target of £20 and expect the algorithm to magically find a way. It won’t. It’ll just stop bidding on most auctions, your traffic will collapse, and you’ll think smart bidding doesn’t work. Start with your actual CPA and reduce by 10–15% at a time.

3. Switching strategies too frequently. Every time you change bidding strategy, the algorithm enters a learning period. During this time (typically 1–2 weeks), performance is unpredictable. If you switch strategies every week because results aren’t perfect, the algorithm never gets out of the learning phase. Pick a strategy, give it 2–4 weeks, then evaluate.

4. Ignoring the learning period. When you switch to a smart bidding strategy, Google tells you the campaign is in a ‘learning’ phase. During this time, don’t make major changes. Don’t panic about daily fluctuations. Don’t add lots of new keywords or change budgets dramatically. Let the algorithm learn. Judge performance after the learning period, not during it.

5. Not using portfolio bid strategies when they make sense. If you have five campaigns each getting 8 conversions a month, individually they don’t have enough data for target CPA. But collectively, that’s 40 conversions — plenty. A portfolio bid strategy pools the data across campaigns, giving the algorithm enough signal to optimise properly. Most advertisers don’t use this feature, and they should.