If you are an e-commerce business then you have the question of how to handle repeats.

Conventional wisdom is that you should be happy to pay more for a new customer than a repeat customer.

Google has some tools to make it easier to do this:

New Customer Acquisition Goal

Google allows you to add extra conversion value to a ‘new’ conversion (a new customer rather than a repeat/existing customer).

If you are using Smart Bidding (which you should be) then this ‘extra conversion value’ allows Google to bid more aggressively on ‘new’ customers – if you set the new customer acquisition goal to increase conversion value by 50%, then, because of this add-on conv. value, the associated spend can also go up by 50% and Google can achieve the same Target ROAS.

You can also use New Customer Acquisition Goal to only bid on new customers for that campaign.

We don’t like New Customer Acquisition Goal because:

Google can not perfectly identify all your existing customers, therefore it is likely to bid higher on some existing customers with the expectation that if they convert they will attract the additional new customer goal uplift in conversion value (which means bidding more for your existing customers).

The net impact of you and your competitors using the New Customer Acquisition Goal is higher CPCs and collectively higher Google marketing spend for your industry (unless everyone is very good at increasing ROAS targets to offset the new customer acquisition goal).

We think the smarter approach is to adjust the bidding for existing customers.

Split your biggest campaigns into two: one targeting repeat customers and one targeting non-customers and then run experiments on both the repeat and non-customer campaigns to work out the optimal Target ROAS for each.

You can assign an audience segment for existing customers on the existing customers’ campaign (though Google won’t be able to identify all existing customers, this audience shouldn’t include non-customers).

Performance Max does not allow for target audiences but you do have the option to target new customers only. You can then set an identical campaign that targets all audiences but has a higher Target ROAS and will hopefully, therefore, become a repeats campaign primarily.



Why should you expect a higher ROAS for repeat customers?

You run a shoe store. An existing customer and a non-customer with the same profile (bar one being a customer and one not) both search for a pair of tennis shoes on Google and both convert on your site.

Should you value the two conversions differently?

The value of a new customer should be based on the number of orders in that customer’s lifetime rather than the value of that customer’s first order only.

How do we value a repeat customer then? One option is to value the order based on the expected number of future orders after the point of that particular repeat order – for a lot of companies this will be more than the lifetime value of an initial order.

What’s very difficult with repeat orders is to definitively say that 1) we wouldn’t have got that order if the customer hadn’t clicked on a specific ad (perhaps they would have come via a different route) and 2) that all future orders are dependent on that specific order (i.e. failing to get that order means the customer is gone for good).

We think a reasonable compromise is to think of repeat orders being worth the value (CM2) in that order only.

New customers are worth the new order value (CM2) plus any expected future free orders CM2 in the lifetime (normally set at 2-3 years).


Should we bid more, less or the same for the existing customer versus the new customer?

Let’s do a thought experiment:

A user does a Google search.  An ad for your company plus ads for three competitors appear on the results page. If the user has not purchased from any of the companies before then that user is most likely to click on the highest-ranked ad.

We think that if the same user has previously only bought from your website of the websites that have a paid ad showing (and the user had a good experience purchasing from you), then that user is more likely to click on your ad than any of the other ads even if your ad is not at the absolute top of the page. (She may even click on your organic listing if an ad for your site is shown).

Because existing customers are aware of your brand they are more likely to click on your ad regardless of its ad position (as long as it’s above the organic listings) and more likely to search for your website in the organic listings if your ad doesn’t appear.

So logically we can bid less for an existing customer than a non-customer because we believe existing customers’ clicks will be less correlated to ad position.

Google however will naturally bid more for an existing customer because an existing customer’s conversion rate on your website is usually higher ⇒ a higher conversion rate means higher conversion value per click which permits a higher cost per click while maintaining the same Target ROAS.

So we want to offset the higher conversion rate expectation of the existing customer (and a little extra) by setting a higher ROAS target for repeat customers.

Lastly, an existing customer is more likely to have interacted with your website via other non-paid channels such as email or direct.

The customer journey of a new customer might be three paid visits whereas the customer journey of an existing customer might be organic > paid > email. Google Ads will attribute a conversion to Google Ads if any visit in the customer journey is from Google Ads.

There will likely be a bigger discrepancy for repeat customers between the Google Ads conv. value and your own attribution estimate, than for new customers.

So make allowance for Google Ads’ generous attribution model for repeats by having a higher Target ROAS on repeats.


Some repeat customers are looking for support – not to buy

If your brand terms have quite a high CPC then make sure you set up a separate paid search campaign to account for people looking for faqs, help etc. – these people are not looking to buy or be tempted by a competitor, so put a low bid and let them come on an organic link if necessary (as long as your organic listing ranks high).


How does Google know if a user is an existing customer?

Google can identify an existing customer by:

  • A cookie on the user’s browser from a previous conversion (which lasts up to 540 days). If the user uses a different device/browser (or has a new phone/computer) or has deleted their cookies then the customer won’t be identified.
  • Customer match of customer email to Google account (or you’ve supplied a customer’s email to Google via enhanced conversions). The user must be signed in to the Google account for Google to match the email address.

You need to add an audience segment that targets repeat customers to the repeat campaign and exclude that audience from the non-customer campaign.

Read about audience segments.


What if I have too many campaigns and am loth to create more campaigns by splitting out repeats and/or running experiments?

We advise running separate repeat and experiments on your largest campaigns.

For PMax and smaller campaigns then you can use the ‘Campaign Target ROAS simulator’ instead of experiments to try and understand the optimal Target ROAS.

You can also put smaller campaigns into a portfolio with a single Target ROAS target and view the ‘Campaign Target ROAS simulator’ at the portfolio level – as long as these campaigns have very similar natural ROAS and product margins.

You can also combine together campaigns that have similar product margin structure and max. ROAS. Read more about optimal campaign structure.


Conversion value rules

If you have added observable audience segments then you will be able to track the performance of existing customers versus non-customers.

You can then use conversion value rules to adjust the conversion value of an existing customer.

Conversion value rules let you change the conversion value that is parsed into Google based on the user’s location, device or (in this case) audience.

The downside of conversion value rules is that it is parsing into Google Ads a different conversion value than has actually been achieved, which can be very misleading when looking at the data (we find it helpful to think conv. value equals revenue attributed to Google Ads so we can compare campaigns and ROASs fairly). Plus it’s very hard to know you have it optimised.

On the whole, we would advise not using conversion value rules. If you think one group is more valuable than another than target that group in a separate campaign and optimise the ROAS on it.


What about Performance Max campaigns where I can’t run experiments?

Performance Max does not permit experiments.

It also doesn’t allow for audience segments either so you can’t segment performance between existing and non-customers.(note: you can create a PMax campaign that only targets new customers, plus you can help google learn what audiences you think should be interested in your products using ‘Audience Signals’).

You can use conversion value rules with PMax but we would recommend not doing so, as per above.

To assess whether the ROAS should be higher or lower use the Campaign Target ROAS simulator. Google gives you an indication of how many more/fewer orders you would receive at different spend levels and you can use mapflo to calculate the different profits (or lifetime profits).

With Performance Max you should have separate campaigns for product groups that have substantially different customer lifetime values or gross margins – read more in the checklist.


What about differentiating prior visitors from new visitors?

The other core audience split you can consider is differentiating prior visitors from non-visitors to your site.

The audience of prior visitors (but non-buyers) is probably more likely to click through to your site than non-visitors as they have more awareness of your brand (but not as much as existing customers).

Make sure you set up prior visitors as an observable audience segment on your search campaigns so you can see how differently they behave from non-visitors. You can then decide whether to split them out of large campaigns too and optimise as per existing customers.


What about other audience segments?

Existing customers and prior visitors are a special case as these people have existing brand awareness.

Nonetheless it is worth adding a few demographic audience segments as observable audiences to your campaigns as you may find that a particular audience performs much better than others and it might be worth splitting that into its own campaign if it has sufficient volume. You can use some of these audience insights to improve audience targeting on Display campaigns.


>>> Read the next article: How to track performance and take action

>>> Back to main menu


The interactive video below highlights some of the analyses we cover:


Please be really careful making changes to your Google Ads account

  • Google doesn’t always respond how you (or we) think it will. The way we think about Google Ads may not be the best set-up for your account.
  • Only change one thing at a time.
  • If possible, always use an experiment to test a change – particularly for significant changes such as moving bidding strategy to Maximize conversion value (Target ROAS).
  • Protect your financial downside by testing with limited spend in the experiment/change. Note that moving to a smart bidding strategy requires a learning phase where Google may not be efficient.
  • Be careful if adding/removing primary conversion actions – changing what Google is converting to can radically change what and who Google targets and how much it’s willing to spend.
  • Remember, all changes to your account are at your own risk. Mapflo shall not be liable for any damages; losses; lost revenue or lost profit.


Glossary of Terms

AOV = Average Order Value

CM1 = Contribution Margin 1 = revenue minus COGS (cost of goods sold) in an order.

CM2 = Contribution Margin 2 = margin on an order after all costs directly attributable to that order such as COGS, shipping, payment fees, customer service etc. (except for marketing).

CM3 = Contribution Margin 3 = CM2 less marketing spend. An ‘Estimated CM3’ value uses an assumed CM2 %.

CPA = Cost Per Action. In this report taken to mean cost per conversion or cost per order.

Keywords = words or phrases (assigned to an ad group) that match a user’s search term and trigger Google to bid to show an ad.

Lifetime CM3 = CM3 from all orders (or subscription payments) for a customer.

Profit = CM3 less all fixed overheads (such as salaries and office rent). Hence Optimising CM3 also optimises profit at the same cost base

ROAS = ‘Return On Ad Spend’ = conversion value divided by cost. A ROAS of 400% means you get four pounds of revenue back for every pound of ad spend.

Search term = the word or phrase that a user searches for on Google.