You send Google a value that reflects what a customer is expected to be worth, not just what their first order was worth. In practice that means a conservative predicted LTV, applied to new customers only, with a payback period short enough that you can finance the gap. Bidding on full projected LTV is how businesses run out of cash while hitting target.
If you only ever bid what a first order is worth, you cap acquisition at what you can afford on day one. Any competitor willing to look past the first order can outbid you for the same customer, permanently.
That is the real argument for LTV bidding. It is not optimism, it is the recognition that a customer with a 40% repeat rate is worth materially more than one order, and pricing them as one order hands them to someone else.
The opposite mistake is worse. Bidding a three-year projected LTV means spending cash today against revenue that may arrive in 2029, from customers who may churn, on a forecast built when conditions were different.
Accounts do not fail because LTV was wrong in the long run. They fail because the payback period was longer than the bank account.
The discipline is a payback window: how long until this customer has repaid what they cost. Ninety days is comfortable for most businesses. Twelve months needs funding and nerve.
That gives you most of the competitive advantage with a fraction of the risk.
Three practical routes, in rising order of sophistication:
Start at the top of that list. A flat, conservative uplift applied correctly beats a sophisticated model applied to unreliable data.
Reported ROAS will rise, because you are sending bigger values. That is not performance improving, it is the unit of measurement changing, and it is the single easiest way to fool yourself.
Judge it on new customer volume, nCAC, and whether the repeat rate you assumed is actually materialising. If the cohorts are not repeating at the modelled rate, the bids are wrong and the dashboard will be the last thing to tell you.
Whatever your cash position can carry, not what your competitors claim. Ninety days is a common and comfortable window for eCommerce. Anything beyond six months needs genuine confidence in the repeat data and the funding to sit through the gap.
Not really. If people buy once and rarely return, first-order contribution margin is the right ceiling and LTV bidding is just a way of overpaying. It is worth the work where repeat purchase or subscription is genuine and measurable.
Yes, and that is the trap. Sending larger values inflates the reported ratio without anything improving. The only honest comparisons are new customer volume, nCAC, and whether cohorts repeat as modelled.
Not sure where your budget is going?A free Google Ads audit, written down and walked through on a screenshare. You keep the findings either way.
Request an auditThirty minutes, no obligation and no sales script. If I am not the right fit I will say so on the call and point you somewhere better.
I will reply within one working day. If you would rather not wait, pick a time that suits you now.
Book a time now30 minutes, no obligation.