Pakistan
USA
Canada
UK
Dubai
Australia
Germany
- Target CPA aims for an average cost per conversion; Target ROAS aims for a return on spend.
- Lead generation usually starts with CPA; ecommerce with ROAS, if values are tracked.
- Set the first target from your own recent results, not from what you wish they were.
- Change targets in small steps and give bidding time to adjust.
How does each strategy work?
Both are Smart Bidding strategies. Google sets a bid for each auction based on how likely that search is to convert, using signals such as device, location, time, query and audience. The difference is the goal you give it. Neither strategy controls your budget; the daily budget still caps spend, and the target only shapes how Google bids within it. Both also depend on the same thing: conversion data that reflects what your business actually values.
With Target CPA, you tell Google the average cost per conversion you want. Some conversions will cost more and some less; the average is what it aims for. With Target ROAS, you tell Google the return you want, as conversion value divided by spend. It will bid more for searches likely to bring high-value orders and less for low-value ones. In Search campaigns, both now sit inside Maximise Conversions and Maximise Conversion Value as an optional target.
Which suits lead generation and which suits ecommerce?
Start from how your conversions are valued.
Situation | Usually better fit | Why |
|---|---|---|
Lead forms or calls of similar worth | Target CPA | Each lead is treated as equal, so cost per lead is the right goal. |
Leads with very different values | Target ROAS with lead values | Lets bidding favour, for example, large projects over small enquiries. |
Online store with varied basket sizes | Target ROAS | Bidding can chase higher order values, not just more orders. |
New account with few conversions | Maximise Conversions with no target | Lets the system learn before you set a limit. |
Ecommerce with no values passed | Fix tracking first | Target ROAS needs accurate values to work at all. |
What data do I need first?
Both strategies are only as good as the conversions they learn from. Check that each primary conversion is a real business outcome, counted once, and that secondary actions such as page views do not steer bidding. If conversions are unreliable, our guide to why Google Ads are not converting covers the common causes.
For Target ROAS you also need values. Ecommerce stores should pass the actual order value, in the account currency. Lead businesses can assign values per lead type, or better, import the value of closed deals from their CRM using offline conversion imports. Google also expects a reasonable conversion history before targets work well; check the current Help page for its guidance.
How do I set a realistic target?
Start with what the campaign has actually achieved over the last few weeks. If your recent cost per conversion is a given figure, set the first Target CPA close to it. Setting it far lower usually makes Google bid so cautiously that volume drops, and the campaign may stop spending.
For Target ROAS, do the same with your recent return. Then work out the target your business needs from margins, not from industry averages. A store with thin margins needs a higher ROAS than one with high margins. A lead business needs to know what share of leads become customers and what a customer is worth. Move towards that number gradually.
What are the common mistakes?
Most bidding problems are data or patience problems.
Targets set from hope
An ambitious target that the campaign has never reached usually cuts volume rather than cost.
Frequent changes
Changing the target every few days resets learning and makes results harder to judge.
Wrong conversions as primary
Counting newsletter sign-ups or clicks as primary makes cheap, low-value actions look like success.
Budget too tight
A budget that cannot afford many conversions at your target leaves little room to learn.
Ignoring delayed conversions
Sales that report days after the click make recent days look worse than they are.
How do I switch strategies safely?
Treat any switch as a test, not a flip.
- Check conversion historyMake sure the campaign has steady conversions and, for ROAS, values that look right.
- Use a campaign experimentRun the new strategy against the old one with a split of traffic, so you have a fair comparison.
- Set the first target from recent resultsUse the actual average from recent weeks, not a goal.
- Wait out the learning periodExpect results to move around for the first period; do not judge it in a few days.
- Adjust in small stepsTighten or loosen the target gradually, then wait again before the next change.
What changes for Pakistani, Gulf and UK advertisers?
Values must be in the account currency. A UAE store with an account in dirhams should pass order values in dirhams, and a Pakistani account set up in rupees should not receive values in dollars. Mixed currencies make Target ROAS meaningless. Google can convert values if you pass a currency code, so set it on the tag.
In Pakistan and the Gulf, many sales close on WhatsApp or by phone rather than on the website. If you only count website orders, bidding sees a small part of the picture. Track WhatsApp and call leads, then import confirmed sales where you can. Seasonal events such as Ramadan, Eid, Black Friday and the UK Christmas period change conversion rates, so Google's seasonality adjustments can help for short, predictable spikes.
How we handle this for clients
We start with tracking: which conversions are primary, whether values are accurate, and whether offline sales can be fed back. Then we choose the strategy that matches how the business makes money, set the first target from recent performance, and run larger changes as experiments.
Targets are reviewed against margins and lead quality, not only against what Google reports. This sits within our Google Ads audit and optimisation service, and builds on our guide to choosing the right bidding strategy. We document each change, so the client can see why a target moved and what happened after.
Junaid has more than 18 years in SEO, paid media, web development and custom software. He founded SEOnumber1, builds WordPress plugins, CRMs and web applications, has served 500+ clients in 48 countries, and holds Google Ads, Analytics and Google Blue and Green Belt certifications. He is a National Level Award Winner in Digital Marketing. Read his story or ask a question about this article.