If there is one word that we’ve been talking about constantly since 2023, it has to be Artificial Intelligence, and how it is disrupting our world. Of course, AI is or will have an impact on a lot of industries, but what will be the impact exactly on marketing? How will it change the way we do advertising? New AI-powered tools keep on arriving on the market so let’s analyse how this will impact us!

First of all, if you are not familiar with artificial intelligence tools, here is a brief explanation. AI marketing tools are programs that use AI to help marketers in all sorts of steps of marketing campaigns such as the tracking, giving insights in order to optimise, or even creating campaigns. The content of the campaigns can also be personalised thanks to those programs, in function of the target and this is all thanks to machine learning and advanced algorithms.

How is AI changing advertising?

The primary objective of artificial intelligence tools in advertising is to improve the results of advertisers, and ultimately have a better conversion rate. So how can AI tools help in achieving these improvements? First, they can easily analyse a very large volume of data and investigate all your important KPIs such as a click-through rate, a conversion rate, cost per conversion, etc. Ultimately, this will help you understand your target better, define the best performing ads, how each audience resonates with your different ads and thus better optimise your campaigns. It also helps to have more personalised ads for your target group, as AI powered tools can analyse each past consumer behaviour and use those insights to show the right ads, to the right audience at the right time.

On top of content personalisation, AI can also play a very big role in content creation. AI tools can generate copy, create images and videos based on the message of the advertisers, and very specific to a brand identity. As content creation can often be a blocking point to advertise, due to budget restrictions, inhouse availability or simply time, this can make a really big difference for a lot of advertisers in terms of cost reduction and efficiency.

What are the benefits of incorporating those tools in our day-to-day work?

Personalised ads: as explained above, advertisers can have more personalised ads for their audience thanks to AI technology, which is increasingly important for the users who want to have tailor-made ads for their likes and needs.

• Better ROI: with the improved analysis that AI brings to the table, campaigns are better optimised and thus results as well, increasing the return for advertisers.
• Gaining in efficiency: Artificial intelligence tools are allowing for lots of automated processes, whether it’s in the analysis of the data or in the creations of the ads and the content. This is a very big gain of time and efficiency for advertisers.
• Better audience targeting: with the analysis that AI tools are doing for each user interaction on websites or social platforms, they can easily identify new prospects. Finding a suitable audience for your product is much more accurate than before.
• Being ahead of the market: by adopting AI tools in the planning, strategy and optimisation of our campaigns, advertisers gain a competitive advantage with their competitors that don’t.

Besides the benefits, AI tools are bringing challenges to our industry

• Potential mistakes: like every tool, Artificial Intelligence tools can also lead to mistakes. Whether it’s due to incomplete data or errors in algorithms that the tools are using, mistakes can happen and they could easily be overlooked as we are fully trusting the tools
• Using AI for unethical purposes: like for many tools, there is a potential risk that those are used for other purposes than what they are intended for. It could be for collecting data without consent, or using personal data without consent so it’s important to keep this in mind as well
• One of the biggest challenges that AI tools might bring is the fact that these tools could be seen as a replacement for our work. With AI tools, it’s undeniable that there will be a lack of human input. But the big issue will be if we start relying too much on those tools, and the innovation in the advertising industry starts decreasing.

In conclusion, AI marketing tools are definitely changing the way we advertise and this in different ways. It can improve the performances of our campaigns, improve our way of working by making it more efficient but also more accurate. By leveraging the power of artificial intelligence, we can gain very valuable insight and improve our advertising by making it more personalised and powerful. Nonetheless, we need to keep being aware of the challenges that AI brings to the table, in order to use it in the right way and to stay top innovators in the advertising industry.

{snippet astridmerenne-en}

In a previous article, we highlighted some reasons explaining why your data might look different between UA & GA4. One of the reasons was linked to the Reporting Identity and the 3 ways of identifying users within GA4:

• Blended: method taking into account both observed data (User ID, Google Signals, Device ID) and modeled data. User activity will be modeled when observed data isn’t fully available and Google Consent Mode is implemented.
• Observed: method considering only observed data (User ID, Google Signals, Device ID). It is similar to the Blended identity approach but does not involve modeling.
• Device-based: method relying solely on Device ID as an identity space.

You can find Reporting Identity in Property under the Admin section of GA4 (click on ‘Show all’ at the bottom right to see the 3 of them):

Identifying if modeling is active

When using GA4, it’s sometimes hard to know whether you have modeled data and you can check that in Reporting Identity.

When there is no modeling:

When modeling is active:

Determining if data is modeled

You can change the Reporting Identity at any time: it will alter the way data appears in the reports, both currently and retroactively, without affecting data collection. This means that at any moment you can switch between the three methods to compare them, which can be very handy. As mentioned, altering the Reporting Identity does not affect data collection but, if you change it, be aware that this change impacts all users with access to the property.

To illustrate the impact of Reporting Identity with a practical use case: for one of our clients we noticed a data discrepancy between UA & GA4. In GA4, the website generated 621K sessions (with Reporting Identity set to ‘Blended’), while UA reported 520K sessions for the same date range. Switching to ‘Device-based’ revealed 523K sessions in GA4, closely aligning with UA data. The data discrepancy could thus be attributed to modeling.

Choosing the right Reporting Identity

The choice of Reporting Identity depends on your specific needs:

• For comparing UA and GA4 Data, opt for the ‘Device-based’ method, as it closely resembles UA data.
• For data representation closer to reality, consider the ‘Blended’ method, as it models data that can’t be observed directly.

By understanding and effectively managing Reporting Identity in GA4, you can ensure more accurate and actionable insights for your analytics efforts.

Need any support on deep diving into your GA4 data? Reach out to This email address is being protected from spambots. You need JavaScript enabled to view it. or visit untangled.semetis.com and fill out the contact form.

{snippet justineheeren-en}

In the ever-changing world of digital marketing, staying at the forefront of innovation is essential. Semetis has demonstrated its commitment to pushing boundaries by being among the first agencies to launch an advertising campaign on Netflix in France with “Atol Mon Opticien”.

Atol Mon Opticien: A Visionary Client

Atol Mon Opticien is a well-known optical retailer in France. They have a rich history of providing high-quality eyewear and eye care services to customers. They recognize the importance of embracing digital platforms to expand their reach. Being a visionary company, they were immediately drawn to Netflix's innovative approach.

Netflix: A New Frontier for Advertising

While Netflix has primarily been known as a streaming giant that offers uninterrupted entertainment, it has recently opened its doors to advertising opportunities. This shift enables brands to tap into Netflix's extensive subscriber base and engage audiences more deeply through video ads.
They offer the opportunity to create impactful 15-second or 30-second video ads that blend into the viewer's streaming experience. These ads can appear as pre-roll or mid-roll placements, though advertisers do not yet have the liberty to choose when their ads will appear. It is good to know that the ads are non-skippable, ensuring that the message is delivered in its entirety to the audience.
In terms of targeting, advertisers can fine-tune their campaigns using a variety of options, including geo-targeting to reach specific regions, selecting genres of movies, leveraging the popularity of Netflix's top 10 movies or TV shows, and even blocking sensitive content to maintain brand integrity. Additionally, age and gender targeting provide further precision in reaching the most relevant audience segments.

Semetis: the digital agency behind the campaign

Semetis and Atol Mon Opticien capitalized on this innovative advertising platform by including it in their media mix for this last quarter of 2023. Given Atol's campaign focus on promoting accessible French eyewear with incentives of up to 100 € on French glasses fitted with lenses, the synergy with Netflix proved ideal. It allows creating massive awareness to a broad audience while ensuring attention of users whom are actively consuming content. This, mix perfectly with other digital and non-digital efforts made to promote the offer to the French audience. In terms of results, we are generating 97.5% completion rate with CPM at 32€ which falls within the average cost for connected TV.

Conclusion

Semetis' innovative approach to advertising on Netflix with Atol Mon Opticien has broken new ground in the digital marketing landscape. This collaboration not only enabled Atol Mon Opticien to extend its reach to a wider audience but also showcased the potential of advertising on Netflix. On top of that, it is a good complement to TV and Connected as well as online video ads as it is mixing the both of them. With a strong 97.5% completion rate and a cost-effective CPM, this endeavour highlights the promising opportunities available to marketers in the ever-evolving digital landscape which is only the beginning as Amazon Prime and Disney + will both be showing video Ads in 2024.

{snippet emilieandries-en}

Server-side tagging in Google Tag Manager is still a rather new concept to most digital marketeers, and it’s normal that some common issues arise. Apart from it being a new concept, it is on the crossroads of marketing and IT, so a close collaboration with teams of different backgrounds is often needed. At Semetis, we’ve been experimenting with server-side tagging over the last years, and did some interesting studies. One of them was the impact server-side tagging had on page speed.

If you’re completely new to the concept of server-side tagging in GTM, there are a couple of interesting resources we can recommend. The most complete resource to this day is definitely the Server-side Tagging In Google Tag Manager course by Simo Ahava. Julius Fedorovicius from Analytics Mania has also written an impressive guide, complemented with some videos. On top, this is free, while the course by Simo Ahava isn’t.

Today, we will be looking at one of the most common issues we saw arise since the conception of server-side tagging in Google Tag Manager: why some GA4 events might not show up in the preview mode of your server container. Note that in this article, we’re using a set-up in which we are sending GA4 HTTP requests to our server container.

There are a couple of primary reasons why your GA4 events might not be showing up in your server container while previewing. Here’s a checklist to go through:

1. Make sure you’re referring to the right GA4 Configuration Tag in your GA4 event tag.
2. Your Configuration Tag needs to be pointing to the right server URL.
3. Make sure that the tag that fires the event is firing on the correct pages or in the correct situations.
4. The data layer is used to pass data to the GA4 event tag. Make sure that the data layer is being pushed correctly and that it contains all of the necessary information for the event.

If you’ve checked all of the reasons above, there might still be another reason why your GA4 event is not showing up: the configuration tag is firing too late. The GA4 configuration tag is responsible for routing hits to the correct destination, whether it's the Google Analytics 4 property or the server container. Make sure that the GA4 configuration tag is firing before the GA4 event tag.

In the example I’ll be using below, we had recently implemented consent mode. This set-up pushes a new data layer event gtm_consent_update which we in turn use to trigger our GA4 configuration tag (and any other pageview tag). The downside of this is that the newly created event gtm_consent_update fires quite late, after all other events have been fired. If you’re pushing an event on a pageload, it can happen that gtm_consent_update is only being fired after that event has been passed in the data layer. Let’s take the example of the data layer event view_item.

As you can see from the screenshot above, the view_item event has been fired way before the gtm_consent_update event has been fired. The GA4 view_item event tag, that is supposed to be sent to the server, fires before the GA4 configuration tag, which fires on gtm_consent_update. Since the configuration tag is responsible for routing the event tag to the right destination, the event is not being sent to the server. The GA4 event in turn doesn’t show up in my server container.

One way of working around this is making sure your event only fires after your configuration tag, but in most cases this requires development support.

Another workaround is to add an extra parameter to your GA4 event tags. When adding server_container_url as an event parameter, the event tag in theory doesn’t need the configuration tag to know where it needs to send data. Even if the event fires before it can use information from the configuration tag, it sends an HTTP request to the server. The result is that my event now shows up in my server container, even before the page_view event sent by the configuration tag.

Need any support on your server-side implementation? Reach out to This email address is being protected from spambots. You need JavaScript enabled to view it. or visit untangled.semetis.com and fill out the contact form.

{snippet robbedesmyttere-en}