At a certain stage of development, companies begin to launch media advertising in addition to performance tools. It warms up the market, makes the brand recognizable, and the audience loyal. For medium and large businesses, media can account for more than 80% of the budget. But few people correctly evaluate its effectiveness and understand how display advertising affects traffic sources. In this article, newage. will explain how display advertising affects other channels. We will also show an assessment of this impact using the example of the Prom client.
Place of display advertising in the digital mix
Display advertising is an important component of digital campaigns. According to VRK, banners and video ads accounted for 40% of all ad traffic in 2018-19. According to our observations, up to 82% of the budget of advertisers who use display advertising falls on this channel.
If a business is willing to invest so much in display advertising, it is important to understand what these investments will bring. This is where the question of performance analysis comes in.
How to evaluate the impact of display advertising
In most cases, media advertising is evaluated like this:
The user saw an advertisement for a smartphone, he immediately went to the site and made a conversion. Obviously, there are few such users, because when you see an ad on Youtube, you are more likely to skip it than follow the CTA.
And if the ad is shown on SmartTV, then the chance of a click and conversion is not higher than zero.
Our first task as media professionals is to evaluate the delayed actions of the viewer after viewing the ad. Comprehensive Analysis and data on post-view and cross-device conversions help with this.
- Post-view is the actions of users who viewed ads and made a conversion from the same device or browser.
- Cross-device is the actions of users who move from one device to another. It often happens that a user sees an ad on a smartphone on Youtube, but moves from the desktop because it is more familiar.
But besides the direct impact of media advertising on the target audience, its contribution to the work of performance tools is also interesting. We asked ourselves this question when we were working on Prom campaigns in 2020. And below we will describe step by step how we built the analytics system and what results we got.
BrandLift is a great way to measure the effectiveness of your ads. Read our article “What BrandLift is and How it Works”.
Prom case — how to evaluate the impact of media campaigns on performance
Prom is a large marketplace. In 2020, he actively used display advertising. There were 5 big flight videos and banners with quite big budgets. The main task was to increase brand awareness, brand loyalty.
Please note that the case is a process case, that is, we applied certain coefficients to further figures to show the structure of the work, but not the data under the NDA.
Analysis tools
You need to understand what you want to get as an output; what is your goal and what data do you want to receive. Depending on this, different systems will be required. When it comes to site conversions, you need a web analytics system with customized goals. If the goal is a sale, CRM and access to it are necessary, as was the case in our case.
In addition, to account for post-view and cross-device conversions, you need tracking with Google Campaign Manager and setting up the transfer of parameters that will determine the user who made the conversion. It can be User ID, Client ID, cookies – your choice.
Also, depending on the scale of the campaign, you may need a big data application. We used BigQuery, but for some, simple Excel or Google spreadsheets may be enough.
How we measure the impact of display advertising
For this experiment, we also used Google Campaign Manager: before launching the campaign, we installed its tags on the client’s website. Due to this, at the time of the conversion, the Client ID of the person who made the conversion was transmitted to the dashboard.
And at the same time, Google Campaign Manager acted as an auditor and standardly marked everyone who saw the ad.
Thus, we got two cohorts of users:
- those who saw the advertisement;
- those who have not seen the advertisement.
By separating users, we were able to compare the behavior of both cohorts.
On average, those who saw the ad were 12% more likely to convert compared to those who didn’t interact with the media. If you look at this figure, we can say that it is small. But it is important to take into account that we conducted analytics on Prom data, and this is the e-commerce giant in Ukraine, which occupies almost 50% of the uanet coverage, and this is a lot.
Single campaigns
In terms of different campaigns, the increase was 9-16% between those who saw and did not see ads. The dynamics are uneven, but noticeable, given how many people the client has on the site.
Let’s take a look at the main conclusions on the example of one of the campaigns – “Back to school”. This campaign performed the best for three reasons.
- We focused on a relatively narrow audience. It had the most successful targetings.
- We focused on those audiences and segments that drove the most conversions. That is, they worked not with a cold, but with a neutral and warm audience.
- The campaign took place at a good time. In the conditions of quarantine, when it was not clear whether the children would go to school or not. Then, when it finally became clear that the children would go to school, everyone immediately began to prepare, and we were just having a campaign.
We compared the data of 2019 and 2020 and overlaid here the data on the media effect for the campaigns that we ran. Compared to 2019 and 2020, the overall website traffic has grown:
- web by 15%;
- application by 40%.
Prom has really grown the application a lot, it is influenced by many factors. Obviously, 2020 was fundamentally different from 2019 and users used online more. But the presence of Prom in the media field also influenced here, compared to 2019.
If we superimpose data on the media effect, then such insights are visible.
- In paid search, both during and after display campaigns, there is a drop in the cost per click and a slight increase in the percentage of conversions.
- If you look at direct and email, then these are the two channels that we most influenced by advertising campaigns. People who were in contact with advertising through these channels had more active contact than those who came through these channels and did not see the ad.
If you are estimating growth even without these meta-parameters, you can still calculate ROI by imagining how media influences other channels.
Conclusions
- Display advertising is an important component of digital promotion. It accounts for about half of advertising traffic, and for medium and large businesses, it can take up to 80% of the marketing budget.
- It is incorrect to evaluate the impact of display advertising by post-click conversions, because this is only a small part of its effect. Furthermore, it is necessary to analyze post-view and cross-device conversions in order to track user actions delayed in time.
- When exploring the impact of media campaigns on other channels for the Prom client, we used the capabilities of Google Campaign Manager. This made it possible to compare the behavior of those who saw and did not see the advertisement.
- The ad audience was 12% more likely to convert. Growth was unevenly distributed across advertising channels and campaigns, but display advertising had a positive effect on customer behavior.
- In the Prom case, direct and email channels grew the most. There was also a drop in the cost per click in PPC and an increase in the percentage of conversions. Social is seeing an impact on increasing traffic, but its conversion rate is lagging behind the previous year. And in Display, we see an increase in traffic and an increase in conversions when conducting media campaigns.
