Which bidding strategy fits your account

The bidding strategy is the setting with the biggest effect on your results and the least clear name. One wrong choice can double your cost per enquiry without any warning.

What the strategies do

Maximise clicks. Google gets as many clicks as it can out of your budget. It does not look at what happens afterwards. Useful for a new campaign with no conversion data, and dangerous if you leave it there, because cheap clicks are usually cheap for a reason.

Maximise conversions. Google goes for as many conversions as possible within your budget, with no ceiling on what a conversion may cost. Works if your tracking is right and you have enough conversions to learn from. Without a ceiling, the price per enquiry can climb.

Target CPA. The same, but you state roughly what a conversion may cost. Set that target too low and Google barely dares to bid, so your impression share collapses. Set it too high and you overpay.

Maximise conversion value and target ROAS. For webshops, where one order is worth more than another. This only works if you pass conversion value and that value is correct. Without value, ROAS is an empty number, and then we do not show it.

Impression share as a target. Steers on visibility, not results. Sensible for brand campaigns on your own name, rarely beyond that.

The question that decides

Not "which is best", but "what does my account already know". Smart strategies learn from conversion data. If you do not have it, they have nothing to steer on.

A practical order that works in most accounts:

1. If your conversion tracking is wrong, no smart strategy makes sense. Fix that first.
2. If you barely have conversions yet, start with maximise clicks or maximise conversions without a target, and accept that the first weeks are expensive.
3. Once you get conversions steadily for a few weeks, you can move to target CPA. Set the target at what you currently pay on average, not at what you would like to pay.
4. To bring your CPA down, do it in steps of about ten percent and give each step a few weeks. Big jumps stall the campaign.

How many conversions you need before a smart strategy becomes worthwhile depends on your sector and the spread in your data. Google publishes guide figures per campaign type; we deliberately name no number, because it varies too much between accounts to turn into a rule.

What almost nobody tells you

Every change costs a learning period. Change the strategy or the target and the campaign starts learning again. During that period your figures are not comparable with the ones before it. Anyone drawing conclusions in that week draws the wrong ones.

The strategy can ignore your margin. Target CPA does not know that product A earns three times what product B does. If that difference is large, it belongs in your conversion value, or you are steering on the wrong thing.

A good CPA does not exist apart from your margin. What an enquiry may cost follows from what a customer is worth to you and how many enquiries become customers. That is what the article on CPA and ROAS covers.

What AdWarp does and does not do here

Warp shows what your current strategy is doing: what you pay per conversion, how that compares to the previous period, and whether your impression share is slipping because of a target that is too tight. It also flags when a strategy was changed, so you know why the figures moved.

What Warp does not do is quietly switch your bidding strategy. That touches your budget and your reach, so it always waits for your approval, with the expected effect stated. And you will not get black-and-white advice: a bidding strategy depends on your margin, your stock and your capacity, and we do not know those three better than you do.