How to get the best results with your Recency Retargeting Strategy

Tim van Rijt Implementation Manager OperationsMore about Tim

Recency targeting is a popular and effective strategy for retargeting consumers who have already shown an interest in your product or site, by contacting them again to move them further towards conversion. It also offers an opportunity to distinguish different retargeting segments (e.g. landing page, product page, and sales funnel), as well as make a distinction between recent and non-recent visitors. In this article, we’ll examine the best ways to use recency targeting, goals you should aim for with this strategy, and we share the results of an A/B test to demonstrate the effectiveness of a recency retargeting strategy.

This highly effective tactic is triggered by the commonly used last cookie counts conversion attribution model. This model credits a conversion solely to the last ad a consumer clicked on, or (if the consumer did not click) the last ad served. By rewarding only the last ad in the conversion path, serving this last ad proves vital to deliver a good eCPA in reports.

Rather than generating new leads, the primary goal of a performance campaign is to serve ads right before a consumer is planning to generate a conversion Twitter icon To make sure a decent amount of last-cookies are generated, retargeting campaigns primarily target recent website visitors. You can compare this to a salesman handing out discount coupon in-store, near the cashier. People will use this discount coupon, giving the salesman a very positive report: a lot of sales were finalised using the discount coupon. But did they finalise the sale because of the coupon? Probably not, they were probably going to buy the product anyway.

How to Recency Retargeting

Most will agree that serving an ad to a consumer five minutes after they’ve left the sales funnel of the corresponding website will not always add value. Chances are high that these consumers plan to convert anyway but weren’t yet ready to finalise the sale. As long as an advertiser keeps using last cookie counts attribution, this setup will not change. For a retargeting campaign to focus on providing real uplifts some testing is needed.

Since a retargeting strategy only targets website visitors, it’s hard to pinpoint at what time and in what way the retargeting campaign provides value. Targeted consumers have already shown interest in the advertiser by visiting the website, making it more likely that you will serve ads to consumers who were already planning to convert. This could be considered a waste of money; you pay a higher eCPM to show an ad to convince somebody of something they were, most likely, already convinced of.

How to Recency Retargeting

Testing the actual uplift is not difficult, but can be costly. This should be seen as an investment to improve the basis of this highly important tactic. The test results should focus on answering the following questions:

  • At what time after the last website visit will retargeting produce an uplift?
  • How many ads should be shown to a consumer to realise an uplift?
  • What is the maximum number of ads shown to a consumer to still generate an uplift?
  • Visitors of which webpages will provide the biggest uplift with a retargeting strategy?
  • What percentage of conversions were from consumers who would convert without retargeting?

These questions can be answered with multiple A/B tests. The A group, existing of 80% of the consumers, will be shown ads as usual through the current retargeting strategy. The B group, existing of the remaining 20% of the consumers, will be shown blank ads through an identical setup.

The results will show the real value of showing the actual ads. To explain in more detail, take a look at the below report:

CampaignImpressions ConversionsCosteCPA
A – Real ads3,200,000216€6,400.00€29.63
B – Blank ads800,00032€1,600.00€50.00

This report shows that the retargeting tactic will provide an eCPA of €50.00 without showing ads. The actual uplift of showing ads is 40%, or an eCPA improvement of €20.00.

It also shows that every 800,000 impressions will provide around 32 conversions without showing ads. All conversions above this 32 can be considered an uplift. In the above example, 32 conversions per 800,000 impressions are organic, meaning 128 conversions per 3,200,000 imps are organic. The amount of conversions realised by showing ads is, therefore, 216–128=88.

By testing this against a more extensive retargeting setup, it is possible to pinpoint the actual uplift per recency. The following example will provide a more detailed fictitious report, with campaign A again being the campaign with real ads and campaign B being the campaign with blank ads.

Campaign A – Actual ads

Website visited less 1 hour ago€500.0041€12.20
Website visited 1 – 24 hours ago€1,000.0058€17.24
Website visited 1 – 7 days ago€1,500.0049€30.61
Website visited 7 – 14 days ago€1,800.0042€42.86
Website visited 14 – 30 days ago€1,600.0026€61.54

Campaign B – Blank ads

Website visited less 1 hour ago€125.009€13.89
Website visited 1 – 24 hours ago€250.0012€20.83
Website visited 1 – 7 days ago€375.006€62.50
Website visited 7 – 14 days ago€450.004€112.50
Website visited 14 – 30 days ago€400.001€400.00

The results of campaign B give a good estimation of the amount of organic conversions delivered by campaign A. We can now deduce the amount of organic conversions from the results of campaign A.

Since campaign A had four times more budget than campaign B, we have to multiply all results from campaign B by four:

Organic conversionsConversions
Website visited less 1 hour ago36
Website visited 1 – 24 hours ago48
Website visited 1 – 7 days ago24
Website visited 7 – 14 days ago16
Website visited 14 – 30 days ago4

These conversions could be seen as the amount of organic conversions in campaign A. To see the uplift the banners provided, we will subtract the organic conversions from the results of Campaign A:

Website visited less 1 hour ago€500.005€100.00
Website visited 1 – 24 hours ago€1,000.0010€100.00
Website visited 1 – 7 days ago€1,500.0025€60.00
Website visited 7 – 14 days ago€1,800.0026€69.23
Website visited 14 – 30 days ago€1,600.0022€72.73

These results show a different perspective on retargeting. The actual uplift on consumers who visited the website up to a day ago is much less than initially thought. A large part of these conversions turned out to be organic conversions.

The biggest uplift is achieved by targeting consumers who visited the website one to seven days ago, closely followed by website visitors seven to 30 days ago.

These results could have a big impact on the way retargeting is treated. Rather than focusing on consumers who recently visited, the focus should be on consumers who visited the website one to 30 days ago. Keep in mind that these results could differ per campaign/brand.

This testing can be followed up by a deeper analysis of the amount of ads shown to a consumer, the time between showing the ads, the websites on which the ads were shown and any other analysis giving more insight in the uplift in conversions in relation to the organic conversions. More in-depth tests will provide more detailed results, although it would require a higher investment to acquire more in-depth test results.

We recommend conducting a similar test to the one described here to get a good understanding of your retargeting strategy and where the actual uplift is produced. This will make sure your most important performance strategy is focused on the right consumers while limiting waste.

Related content: