Practitioner · Lesson 2 of 5

Frequency caps and suppression lists

About 9 minutes · Yogender Kumar, Managing Partner

By the end of this lesson you'll be able to

  • Explain what a frequency cap and a suppression list do
  • Spot wasted spend in a frequency report
  • Calculate the effect of a cap and a suppression list on cost per order

Why this matters

Retargeting is cheap per impression, so waste is easy to miss. The same few people can see your ad dozens of times, including people who bought last week.

The core idea

A frequency cap sets the most times one person can see your ads in a period, such as eight a week. After that, the platform stops bidding on them.

A suppression list removes people from an audience completely. The usual ones are:

  • people who bought in the last 30 days
  • existing customers, for offers meant for new ones
  • staff, and anyone who has asked not to be contacted

A cap saves money on people who've seen enough. A suppression list saves money on people who've already done what you wanted. Anyone who has objected to marketing must always be suppressed.

Worked example

Larch & Lane is the imaginary furniture shop from Lesson 1, in a different month. The figures are illustrative. Its retargeting spends £3,000 a month and the platform reports 75 orders: £40 an order.

Step 1: the buyers. The audience report shows £900 of the £3,000 went on people who'd already bought. Those impressions produced 5 orders, mostly small add-ons. Larch & Lane adds a suppression list of last month's buyers.

Spend falls to £2,100. Orders fall to 70. Cost per order is now £30.

Step 2: the frequency. The frequency report splits the remaining orders by how many ads each person saw that week:

  • 1 to 3 ads: cost per order £25
  • 4 to 8 ads: cost per order £40
  • 9 or more ads: £300 of spend for 3 orders, which is £100 each

Larch & Lane caps frequency at eight a week. To keep the sums simple, assume that removes the £300 and those 3 orders. In real life some of those people would still buy from their first eight ads, so treat the result as a best case.

Spend is now £1,800 for 67 orders, or £26.87 an order. That's a third less than where it started, for about 11% fewer orders.

The £1,200 a month it no longer spends can go on finding new visitors.

Cost per order after each change

Start: £3,000 for 75 orders

£40

Buyers suppressed: £2,100 for 70

£30

Capped at 8 a week: £1,800 for 67

£26.87
Suppressing buyers and capping frequency cut cost per order by a third, for about 11% fewer orders.

In practice

  • Upload last month's buyers as a suppression list, and refresh it at least weekly.
  • Start with a cap around eight a week, then adjust from the frequency report.
  • Look at cost per order by frequency band each month.
  • Suppress existing customers from offers meant for new ones.
  • Use the same suppression lists on every platform, and cover customer list uploads in your privacy notice.

Common mistakes

  1. No suppression list. You pay to remind people of something they bought yesterday. Upload buyers first.
  2. A cap set once and forgotten. The right cap changes with the season and the offer. Check the frequency report monthly.
  3. Judging only the average. A £40 average can hide a band costing £100 an order. Split it by frequency.

Try it yourself

Retargeting costs £4,000 a month for 100 orders. £1,000 goes on recent buyers and brings 10 orders. Another £500 goes on people who saw ten or more ads and brings 5 orders. What's the cost per order before, and after removing both?

Show answer

Before: £4,000 ÷ 100 = £40. After: spend is £4,000 − £1,000 − £500 = £2,500, and orders are 100 − 10 − 5 = 85. Cost per order: £2,500 ÷ 85 = £29.41.

Quick check

Not scored. Pick an answer to see the explanation.

  1. What does a suppression list do?

  2. £600 of spend went to people who saw nine or more ads, for 4 orders. What did those orders cost each?

Key takeaways

  • A frequency cap limits how often one person sees your ads.
  • A suppression list removes people, such as recent buyers, completely.
  • Split cost per order by frequency band to find the waste.
  • Refresh suppression lists often and use them on every platform.