Independent study of 4,000 accounts finds tROAS floors beat tCPA caps in lead gen
A large cross-account dataset suggests value-based bidding outperforms cost caps even for lead generation — when values are set honestly.
- 4,012 accounts analysed across 14 months, £180M combined spend
- tROAS with CRM-fed values produced 19% more qualified pipeline per pound
- Accounts using proxy values (form fill = £50 flat) saw no advantage
- Benefit concentrated in accounts with 30+ offline conversions per month
The headline will get quoted as 'tROAS wins' and that's exactly the wrong takeaway. Read the third point again: flat proxy values killed the advantage entirely. Value based bidding is only as good as the values you feed it, and most lead gen accounts are feeding it fiction. The real lesson is about measurement discipline, if you can't connect a click to a closed deal, no bidding strategy will save you. Below 30 offline conversions a month the signal is too thin and you're better off with tCPA and ruthless lead qualification upstream. This is a data maturity test dressed up as a bidding study.
Questions people ask
Is target ROAS better than target CPA for lead generation?
Target ROAS can outperform target CPA for lead generation when CRM values reflect genuine differences in lead quality, but this study does not establish a universal winner. Accounts feeding CRM values into tROAS produced 19% more qualified pipeline per pound, while flat proxy values showed no advantage. I would want to see that pipeline turn into contracts before calling it better business.
Can I use target ROAS for lead generation if every form submission has the same value?
You can use target ROAS with the same value for every form submission, but that gives Google no information about which leads are worth more. Flat proxy values showed no advantage in this study, which is why I would fix the CRM feedback before changing the bidding strategy. Calling every form fill £50 does not make it worth £50.
How many offline conversions do I need each month before testing target ROAS for lead generation?
This study found the benefit concentrated in accounts with at least 30 offline conversions per month, but that is an observed pattern rather than a guaranteed threshold for success. Below that level, my preference would be tCPA focused on qualified leads rather than asking Google to optimise around thin value signals. Getting above 30 would make me consider a test, not assume the account is ready.
What conversion values should I send from my CRM to Google Ads for lead generation?
Send values grounded in actual deal outcomes or defensible estimates of expected revenue, rather than assigning every lead an arbitrary amount. I want the Salesforce view to show whether higher valued leads actually close more often or produce more valuable contracts. If those differences do not hold up, I would fix the values before giving tROAS more budget.
Want this level of thinking on your account?
Get a free AI powered PPC audit of your website and goals, with quick wins and a 90 day plan.
Get my free audit