Should you use Target CPA bidding for lead generation campaigns?

Note: This article was written a few years ago. Google Ads changes frequently so it might not be the best advice today.

Google really wants you to use Target CPA bidding. The Google Ads reps have been pushing it hard and I see notifications to switch on almost every account I manage.

In this article:

1. Risks and trade offs of Target CPA bidding.
2. When not to use Target CPA bidding.
3. Reducing the risk in testing Target CPA bidding.

Target CPA bidding is one of several automated bidding strategies. CPA stands for cost per acquisition. If you use Google Ads to generate leads you can think of it as target cost per lead.

Here's the simplified version of how it works.

You give Google a target cost per lead to aim for and a daily budget. Google takes over the management of your bids and tries to get you as many leads as possible at the target cost. You won't have to adjust keyword or ad group level bids anymore.

The thought behind it is that Google knows much more about the person searching than you do. Google will use that knowledge to predict if the searcher will become a lead. Google will adjust your bids to get more of the people who they think will become a lead to click your ad. They'll do the opposite when they think the searcher is unlikely to convert.

Target CPA risks and trade offs.

You may get fewer leads with Target CPA bidding. Like every other experiment in Google Ads, there is a chance that it will make your campaign worse. If it works, you enjoy the upside, but, you carry the risk of the downside.

But, apart from the risk of a campaign tanking, there are also some trade offs that aren't always obvious.

1. You’re giving Google control over how the campaign is managed. Google might run your campaign very differently to the way you would. You might see low CTR or eye-watering CPCs. Don't switch to Target CPA bidding unless you're comfortable ignoring the intermediate metrics like impressions, clicks etc.

2. Any device bid adjustments not set at -100% will be removed. This means you can’t bid more (or less) for clicks from a searcher on a mobile phone than you bid for a searcher using a computer.

Instead Google adjusts your target CPA by the device bid adjustment. For example, if you increase mobile bids by +200% then Google will increase your target CPA by + 200%.

In some cases this works fine -

I have a client who repairs automated driveway gates. Mobile phone leads are worth a lot more than computer leads. Someone Googling for gate repair on their phone is usually stuck on their driveway. They're an easy sell. Someone searching on their computer is usually not urgent and my client closes fewer of these leads. She might get 30 or 40 jobs from 100 mobile phone leads. She'd expect 8 to 12 jobs from 100 desktop leads.

In other cases the link between the cost of a click and the value of the conversion isn't as clear.

3. You lose control over keyword level bidding. Google assumes that every lead is worth the same, regardless of which keyword generated it. Again, this is not always true.

I have a client who makes and installs windows. The campaign includes the keywords [aluminium window prices] and [aluminium window price list].

Over the years we’ve seen that people searching for a price list are in the very early planning stages. People searching for prices are closer to buying now. A lead in the planning stage is worth less to my client than a lead who's waving money. We'd account for this by adjusting bidding at keyword level. We couldn't do this using Target CPA bidding without some clunky workarounds.

Don’t switch to Target CPA bidding if:

You’ve recently made significant changes. It’s best to let the last changes settle before applying the next major change.

You’ve had few conversions over the last month or so. The more conversions you’ve had recently the better. I don’t know what the smallest number of conversions you need is but I've seen suggestions that range from 15 - 200 a month.

Your business will be in trouble if you have a low-lead month. Any major change in Google Ads causes some turbulence. Switching bidding strategy to Target CPA is no different. You may have a period with fewer leads while Google works out how best to manage your campaign.

The number of leads you get doesn’t match the number of conversions in Google Ads.

In Google Ads, the term conversion means an action your visitor takes. That’s the A in Target CPA.

The action (conversion) might result in a lead - say filling in a contact form. But, the action could also be an intermediate step you’re tracking like clicking a link to the contact page.

With Target CPA bidding you’re asking Google to get as many conversions as they can at your target cost. If you include both actions - clicking the link to the contact page and filling in the form - in your conversion count, Google thinks it's scored twice.

You’d end up spending twice your target CPA per lead - once for them clicking to the contact page and once for them filling in the form.

Google would also consider that they’d hit their target if someone clicked the link to the contact page but didn’t fill in the form and become a lead.

Google includes the intermediate conversion in the total conversion count unless you specifically change it.

You may also overmeasure conversions if you have more than one way for a visitor to become a lead.

For instance, you might have told Google your Target CPA is $20. A lead might fill in an enquiry form, then engage with your live chat system and click to call your office. If you’re counting each of these as a conversion then Google would be proud if they spent 3 x $20 to get you 3 actions. You might not be as happy to pay $60 for one lead.

You may never get 100% match between conversion tracking and lead count. That’s OK as long as you know by how much you’re over measuring conversions so you can drop your target CPA.

Reducing the risk of testing Target CPA bidding.

There are two opposing forces at play any time you test in Google Ads:

• You need to give the change enough time and budget to have a fair shot at winning.
• You need to limit the damage if the experiment is a failure.

It’s the same with trying out Target CPA. You need to let it run for long enough to get past the turbulence so you can make a rational decision on if it's right for you.

So, how do you limit the potential for damage?

• If you’ve got several campaigns you can test Target CPA with one. (If you use shared budgets you might want to consider a separate budget for the Target CPA test campaign.)
• If you’ve got one campaign you can use the Experiments feature to make a copy, change it to Target CPA bidding and split the traffic 50/50 with your regular campaign.

Need some help with this?

I offer 1-to-1 mentoring and consulting. You’ll get help, advice, support and answers without having to commit to a long-term contract. Details here.

Want to increase your conversion rate?

I’ve seen conversion rates double, triple and even 10X after a few quick and easy changes to the enquiry form. Subscribe now, and I'll send you a free copy of my in-depth ebook High-Converting Enquiry Forms.

You're already paying for clicks. Now turn them into conversions. Privacy policy. Unsubscribe at any time.