With the popularity of our Holiday LinkedIn Ads analysis, we wanted to follow it up with the performance of additional holidays.

I especially wanted to see what would happen on Civil Rights Day because it’s a government holiday in the U.S., so many professionals get to take the day off of work. But it’s also not celebrated in the same, big manner as the winter holidays. (I don’t mean to minimize the importance of the civil rights topic–just to say it’s not as commercialized).

My hypotheses going in:

‍
1. Many take it off of work, so I expect to see fewer overall impressions, especially from Desktop. Probably not as few as the winter holidays but nowhere near a normal workday.

‍

2. I expect costs to rise anytime there are fewer impressions, so I expect to see CPCs and CPMs rise by some amount, but probably not a shocking amount.

The questions we set out to ask:
1. How much less usage of the LinkedIn platform would we see on a non-major holiday?

‍

2. What happens to costs before, during, and after each holiday and what insights can we extrapolate from them?

METHODOLOGY

We collected 8 accounts where no changes were made to ad creative, bidding, or budgets on or around the holiday.

Civil Rights Day was on Monday, Jan 16th, 2023.

We took the average performance of each ad type running for each day of the week (excluding the day before, the day of, and the day after the holiday). Then we took the performance of the holiday, as well as the days around it, to compare to an average day.

For instance, the day before the holiday was Sunday, Jan 15th, so we compared the ad performance on that day to the average Sunday as a baseline.

The results as you read them will be the performance on the holiday as a percentage of what a normal day would look like. So 110% in one category would mean 10% more than a normal, average day. And 50% would mean that performance on that day was half that of a normal day of the same day of the week. Makes sense?

THE RESULTS

Without further ado, here’s the exciting part.

The measures that we chose to track were:

Engagement (measured by clickthrough rate)

Click Cost (by average cost per click)

Impression Cost (by average cost per impression)

ENGAGEMENT

You’ll notice very little deviation in engagement across the holiday. Although some of the accounts showed a marked increase in engagement, many others showed a decrease that brought the results within a whisper of where they would normally be. The day before showed a 2% increase, followed by a 1% decrease on the holiday, and the day after was 1% higher than average. Nothing worthy of writing home about.

CLICK COSTS

Here’s where we saw some results that were actually worth mentioning. Costs were 100% the Sunday before the holiday, simply meaning that they were average. The day of the holiday, though, costs rose 10%, which isn’t surprising, given that we usually see costs rise on holidays. Click costs were still up 6% the day after.

IMPRESSION COSTS

Costs per impression closely mirrored that of their click costs.

TRAFFIC PATTERNS

Seeing the results of the holiday’s effect on performance is certainly interesting, but I was even more excited to see the traffic patterns.

How active were people on LinkedIn? And how do we know?

5 of the 8 accounts were running Dynamic Ads, which only serve on Desktop. So we had enough data to compare how many impressions occurred on holidays where most white-collar business professionals wouldn’t be at their computers.

We know that Dynamic Ads have relaxed frequency caps compared to Sponsored Content ads which allows them to be shown more often, which actually makes the data a bit more robust since it better represents more activity on the platform, since each prospect can generate 5+ impressions rather than 1 impression per day like Sponsored Content shows.

Here’s the data:

DYNAMIC ADS

The story that Dynamic Ads show is that prospects likely took the whole weekend off of work to enjoy the holiday weekend, and then were back to work to make up for the day off with a bit of extra productivity.

The Dynamic Ads data show before the holiday, only 54% of the impressions were generated compared to an average Sunday. On the holiday itself, 42% of average impressions happened. Then the day after, 5% more activity was registered.

SPONSORED CONTENT

Sponsored Content told a bit different of a story. We see less impressions served on the day before (91% of average), but the day of the holiday, we actually saw a 5% increase over an average work day. These are mobile users since we don’t see this level of activity reflected on Dynamic Ads.

The big shocker is that Sponsored Content impressions jumped to 71% above average on the day after the holiday. Desktop impressions certainly increased on the following day, but mobile usage almost doubled! I’m not sure what to attribute that activity to other than professionals jumping into work and making up for the lost productivity of having the previous day off.

Did you see the same effect in your account? Did the results surprise you?

What do you think the results will be like on Valentine’s day? It’s a minor holiday that is highly celebrated, but most are still at work in the US. What are your hypotheses to what’s going to happen to your ad performance?

‍

Written by AJ Wilcox

‍