What used to be a straightforward task — tracking and converting customers — has become quite a challenge, especially for small advertisers. Only 60% of brands feel prepared for the deprecation of third-party cookies. Moreover, small companies often lack access to first-party data or data management platforms, and their experimentation budgets are limited. Add here the lack of internal tools for collecting and processing data, and you get the vivid yet troubling picture. However, there’s a solution — contextual targeting.
Why control is crucial for small advertisers
The primary problem of small advertisers is a lack of data. For example, one e-commerce company noticed a drop in conversions. The reason was issues with identifying website visitors and retargeting them. Without cookies, the brand couldn’t track potential customers who browsed products but didn’t make a purchase — and reach out to them.
We’ve been hearing a lot of stories like this from our clients over the last few months. And statistics back up our observations: according to Adobe research, over one-third of marketers have already noticed that the deprecation of third-party cookies has affected their ability to track and target customers.
So, if you can’t rely on third-party data and don’t have your own first-party data, like most small advertisers, then what’s left?
You can target them based on the content they consume. It’s the essence of contextual targeting — it analyzes the page’s content and displays relevant ads. It doesn’t rely on cookies, yet it can deliver great results. Some brands achieve 10-fold or even bigger growth in ROAS with contextual targeting compared to the old cookie-dependent marketing strategies.
Platforms that support contextual targeting allow advertisers to choose the content in which their ads appear. By defining keywords, topics, and even tone of voice, small companies can find the best “frame” for their ads.
How to make the context work for you
The logic of contextual targeting differs from what many advertisers are used to. It doesn’t try to identify a visitor’s ID — instead, the context works as an intent signal. After all, if a person is browsing specific content, there’s usually an intent behind it.
DPSs that support contextual targeting can match the context to your campaign’s settings, and if you choose the right placement, you can stabilize your click-through rates and CPA. Moreover, such DSPs allow you to experiment with various content topics and categories and choose the best-performing contexts. So, even without access to user data, you still gain control.
To make the most of contextual targeting, you can use the STS model.
S – Setup
This is all about setting up contextual targeting in your DSP:
- Сreate a word taxonomy. A list of 25-40 key terms related to the desired context. These words should be connected to your product or service category and describe your audience’s interests and intents. Think about what your customers are looking for and what they’re reading or watching.
- Avoid broad terms. For example, for a natural cosmetics brand, “skincare” or “self-care” would be too generic, but “chemical-free moisturizer” or “eco-friendly packaging” could hit the right spot.
- Set up a brand-safe supply allowlist and a blacklist. Keywords aren’t enough to ensure high-quality placements; you should aim for traffic sources that align with your brand values. Some DSPs automatically optimize traffic sources, so you don’t have to whitelist or blacklist them. However, when brand safety requirements are strict, advertisers should manage these lists themselves.
- Make sure your creatives and landing page content align. After clicking on an ad, customers expect to find the same information. If they land on a generic page instead, their trust and your performance will suffer. So, to avoid a drop in conversions, check your creatives and content for consistency.
These settings help you build trust, increase click-through rates, and keep up conversions. Over time, you can create a context-aware environment that no longer relies on third-party cookies, enabling you to implement your programmatic advertising strategy efficiently.
T — Test
Contextual setup is just the first move; optimizing and scaling this effort is equally essential. According to our experience, here are the most effective steps small advertisers can take:
- Test various contexts for targeting. For instance, define several content clusters that match your customers’ interests, and test ad placements across relevant websites or apps. Then, compare the performance (click-through rates, bounce rates, etc.).
- Define the best-performing contexts (keywords, topics, websites) by calculating CPA, CTR, and ROAS. Next time, allocate your budget with this analysis in mind — and keep refining the contexts.
If your DSP allows, you can create rules to increase ad placement bids for contexts that drive more clicks and conversions, so optimization happens automatically.
S — Scale
The final step is scaling, which should be performed mindfully. For small advertisers, it’s crucial to move slowly but steadily. First, ensure that at least 2-3 topics are delivering decent results. Then, allocate not more than 20% of your budget to experimenting with new content clusters. Finally, test 1-2 new topics or keywords for a week and analyze the results.
Following this model can become the path to stable performance for small advertisers in the cookieless world. For example, I’ve seen how a small DTC skincare brand almost doubled its CTR (from 0.6% to 1.1%) after moving from broad placements on beauty sites to contextual clusters around topics like “sensitive skin routine” and “chemical-free moisturizer.”
How to measure the performance of your contextual stack
Utilizing the full power of contextual targeting requires a shift in your analytics approach. The key question is no longer “who is the customer?” but rather “where and why did they interact with your ad and convert?”
Hence, the most critical metrics focus on the content quality and post-click actions. The two main groups are:
- 1. Pre-click metrics. The first of them is viewability — after all, if a customer can’t see your ad, it doesn’t matter if the content on the page is relevant to your brand.
The second metric is click-through rate, which shows if your ad is well-matched to the content around it. Also, metrics like hover time (time spent with a mouse cursor over an ad without clicking) can help you estimate a customer’s engagement with your ad.
- 2. Post-click metrics. We’ve already mentioned bounce rate — if it’s too high (over 70-80%), it makes sense to check whether your ad and the landing page content align. Other essential metrics are scroll depth and session duration — they show how effectively you selected the ad context.
Scroll depth over 60-75% is considered good, while 40% or less indicates problems. Session duration varies widely by industry. According to Databox, the average session duration is 77.61 seconds for B2B companies and 92.33 seconds for B2C companies.
However, it makes sense to research the average session duration in your specific industry and use it as a benchmark. If your bounce rate, scroll depth, or session duration consistently fall outside these benchmarks, it’s worth checking whether your contextual signals are truly aligned with user intent and whether your landing page content aligns with your ads.
Finally, actions such as adding products to the shopping cart and filling out forms can help you better understand a customer.
Since you’ll be experimenting with different contexts, comparing their performance is a must. Moreover, tracking week-to-week dynamics will help you understand how effectively your programmatic advertising strategy works in the new reality.
Key takeaways
As third-party cookies continue to decline, small advertisers are becoming increasingly vulnerable to losing conversions. On top of that, they usually don’t have access to first-party data or large budgets.
At the same time, contextual targeting can provide much-needed control and transparency, allowing advertisers to experiment with different contexts.
To get the most out of contextual targeting, you may take these steps:
- Create a whitelist of publishers — and a blacklist, too.
- Blacklist keywords or topics — and update this list regularly.
- Experiment with different contexts weekly, measure performance, and choose the best.
Contextual targeting has enormous potential to keep your conversion rates stable and help you scale your efforts. So, give it a try!