How Does Traffic Seasonality Influence Your CPM?

Traffic fluctuations caused by seasonality is present in most industries and is linked to an actual season itself or a given day of the week. However, seasonality can be impacted by both unpredictables, such as erupting volcanoes or global pandemics, and predictables, such as economic upturns or downturns. The Programmatic industry has always to remain flexible in order to meet such challenges.

What is seasonality?

In the Digital Publishing industry, Ad Revenue Seasonality refers to the correlation between the volume of traffic and advertising spend generated by Advertisers.

While media houses and brands plan their expenses in a way that is usually easy to predict (annual budgets are divided into shorter periods – quarterly and monthly goals, so that the beginnings of each of these periods are usually “worse” for Publishers – because not all budgets are yet active), traffic (the presence of users on the website and sessions generated by them) is associated both with a fixed calendar of annual events as well as with unexpected random situations and sudden events that affect the behaviour of people using the Internet.

Whatever the reasons and no matter the industry, the impact of seasonality on Ad Revenue is undisputed and every Digital Publisher should be aware of what events and how affect his inventory.

Types of Seasonality

Annual Traffic Distribution

To present a graph that is considered to be the most standard annual distribution of website traffic, we look at one of the few industries that were somewhat immune to the changes and drops resulting from the pandemic – Hobbies, Games and Leisure. Collectively these are very good examples of how traffic can fluctuate usually during the year.

In the Digital Advertising calendar, Q1 is usually flat and the expenses of Advertisers start to grow steadily only as of March. There are small ups during Easter and Children’s Day. What we can see is that after the summer holidays this number starts to increase steadily as consumers tend to search more for activities picked up during the holiday season.

We could expect that this rise in interest for this industry would last at least until mid-December, but that is not the case. In fact, it drops quite significantly in the first week of December and declines further up to the New Year. The reason for that is Black Friday that generates a big jump in searches not only for bigger promotions but also in searches for Christmas presents.

Monthly Traffic Distribution

Over the years, we at Yieldbird have had the opportunity to support hundreds of Publishers in web monetization and – while annual trends, which are anchored in large-scale global events, are already a recognizable and accepted justification for fluctuations in earnings – declines at the turn of the month often raise additional questions about the reason. Why is the Ad Revenue low at the beginning of the month?

Well, it’s a matter of formalities – Advertisers who divide their annual budgets into smaller “chunks” (quarterly and monthly goals) need these 2-3 days (and sometimes more) in order to confirm and launch new campaigns. This is also the reason for the earnings growth at the end of the month – the projected budget is worth using as planned.

Weekdays Traffic Distribution

Internet users have different habits during different days of the week, and this is particularly in evidence during the start of the weekend – and obviously varies at the industry level.

Below are the examples, but we strongly recommend that you look at your traffic stats in Google Analytics (number of users and sessions will be a good starting point) and look at them week by week over a longer period of time in order to gauge the regularity of trends.

What may cause traffic changes on a specific day?

Monday – Tuesday: the beginning of the week may mean a slight decline in traffic for the lifestyle, cooking, leisure sites. On the other hand, for business-related and news websites, increased interest of users can be expected.

Wednesday – Thursday: it is the middle of the week and the users, who were already able to catch up with responsibilities at work, still have plenty of time to finish them by the end of the working week. Usually, all the categories may expect a slight increase in traffic.

Friday: end of the working week meets us with a drop in traffic, as most of the users prefer to spend the afternoon on activities involving other people.

Saturday – Sunday: out-of-home activities reduce the scale of traffic to almost all categories of websites, especially from the desktop. The closer to the end of the weekend, the more traffic begins to regain scale.

Is there one proven pattern for traffic fluctuations?

Absolutely not. The category of the website and the subject of its content is indisputably related to how its users will behave. A great example of how trends move through the week in an entirely different way than described above is a site with cooking and baking recipes.

As we can clearly see, it is very low at the beginning of the week and gradually increases as the weekend approaches. This is the time when people look for recipes because they will have enough time to plan shopping and cooking their weekend meals. During the week there is definitely less time for this kind of activity.

Traffic fluctuations: expect the unexpected

The patterns are great, but in the real world, we have plenty of sudden events that are unpredictable and can cause short or long term changes in the online advertising market.

With Yield Optimization it is impossible to be ready for everything. No one, for example, could have predicted the scale of the impact of Covid-19 on Programmatic Advertising and how long it would last. Fortunately, there are still some things that we can do with the unpredictable cases – a good tool to counter the effects of unexpected events would be Price Genius. It adjusts the prices of your ad inventory automatically and seeks to give higher prices; and not during the sudden drops.

We also strongly recommend using your knowledge of your inventory’s traffic trends for planning tests and implementations. Things like verification of the UX ad grid, new kinds of ads can be safely introduced in the low-traffic period, to minimize the risk of any abnormal and unwanted behaviour.