Catch 22: Resolving LinkedIn’s Frequency Capping Paradox
Coined by Joseph Heller in his 1961 satire of the same name, Catch-22 refers to a paradoxical absurdity in which there is no escape because of mutually conflicting conditions. In the novel, Doc Daneeka, a WWII army psychiatrist explains that a pilot requesting mental evaluation — hopeful of being diagnosed with insanity to avoid flying dangerous missions — is, ironically, demonstrating their sanity by making the request in the first place.
Albeit a tad less dramatic, B2B marketers face a similar dilemma when it comes to their LinkedIn ads. By design, LinkedIn offers advertisers limited control over the frequency distribution of their ads. That is to say, advertisers don’t really have a choice in how many times each target user/account is served an ad. This results in the majority (≈80%) of ad impressions being served to just about 10% of audiences.
This lopsided ad distribution occurs primarily because bigger companies with larger social presence tend to inadvertently eat up the majority of impressions (and consequently, budgets). In turn, advertisers are impacted by a combination of overexposure or ad fatigue amongst a few accounts, underexposure with the majority of accounts, and worst of all: woeful, wasted marketing dollars.
🧢 Read about the impact of frequency capping (or the lack thereof) on your LinkedIn Ads. 🧢
To solve for this, B2B advertisers typically segment their campaigns based on firmographic properties: big accounts in one bucket, medium sized accounts in another bucket, and small accounts in a third bucket. Given that accounts within each bucket are of similar sizes, ads tend to be distributed more evenly across the board. While this is an effective stop gap, it’s far from perfect:
- For one, it’s a tedious, time-consuming chore to configure multiple campaigns for each segment, especially since LinkedIn also requires individual campaigns per ad format, bid strategy, etc. Before you know it, the number of campaigns you’re managing start to spiral out of control.
- More importantly, however, the ultimate objective of reducing wasted ad spend and improving reach is failed to achieve because campaign cost per mille (or CPM) rises significantly as the campaign’s audience size shrinks. This defeats the purpose of exposure control in the first place.
In other words, larger audiences result in lower CPMs but lopsided ad frequency distribution. Smaller, segmented audiences result in better frequency distribution but increased CPMs. And the end result in either case? Wasted ad spends. This edition of Factors Labs crunches the numbers to qualify this Catch-22 and highlights how Factors solves for it.
Crunching the numbers: Audience size vs. CPM
If you’ve run LinkedIn ad campaigns, you’re likely already familiar with the negative correlation between audience sizes and cost per mille (CPMs). As a quick refresher, here’s what our data look like:
While this slope is relatively shallow, the result is clear: the bigger the audience, the cheaper the ads. Especially at scale, the difference between a CPM of $42 and $25 can make considerable difference. However, as we qualified in the pie chart, consolidating all your target accounts into a single audience inevitably leads to asymmetrical, unbalanced ad distribution. Here’s a detailed breakdown of the same:
| Ad Account | No. of target accounts | Total impressions | Impr delivered to top 10% of target accounts | % of Impr delivered to top 10% of target accounts |
|---|---|---|---|---|
| A | 187 | 4,280 | 2,346 | 54.8% |
| B | 10,000 | 1,555,369 | 935,463 | 60.1% |
| C | 5,801 | 668,462 | 478,176 | 71.5% |
| D | 4,640 | 4,064,176 | 3,474,681 | 85.4% |
| E | 6,113 | 171,126 | 97,009 | 56.6% |
| F | 2,043 | 104,526 | 80,300 | 76.8% |
| G | 2,628 | 557,381 | 516,090 | 92.5% |
| H | 10,000 | 431,642 | 265,132 | 61.4% |
| I | 4,687 | 3,516,461 | 2,754,925 | 78.3% |
| J | 6,912 | 499,642 | 357,044 | 71.4% |
| K | 1,532 | 299,426 | 235,817 | 78.7% |
| L | 815 | 13,904 | 9,138 | 65.7% |
| M | 9,081 | |||
