Hello everyone, I previously worked for over five years as an engineer on the advertising algorithm team at Meta, and I later held a senior engineering position at Reddit focusing on ads as well. Drawing from my experience assisting more than 120 brands after my tenure at Meta, I’d like to share some often-overlooked strategies that could enhance revenue for those already involved in commerce or B2B.
Optimization
Through my experience with Meta’s ad delivery systems, I understand that precise technical implementation of your website is essential to avoid wasting resources. Some critical technical aspects that influence ad algorithms include:
- Server-side API integration (especially important post-iOS 14)
- First-party cookie integration
- Advanced matching criteria
- Custom conversion events
- Real-time event tracking
Moreover, it is vital to monitor every significant user interaction on the server side. My experience at Meta indicated that campaigns performing server-side event tracking saw a 3-4 times improvement compared to those that used only client-side tracking.
Collecting First-Party Data
This data is the backbone of effective ad algorithms. Key data points you should track include:
- User behavior trends
- Conversion pathways
- Time taken to convert
- Signals from cart abandonment
- Metrics on feature utilization
As a tip, log these events promptly on your servers, as averaged data from client-side only methods often leads to a 30% loss in the data. Hence, it’s beneficial to use your own first-party tracking pixel or an intelligence application instead of relying on third-party options from advertising platforms like Meta and Google.
Optimizing Algorithms
From my background in constructing these systems, the following factors are crucial:
- Event quality evaluations are more reliable when tracked server-side compared to third-party pixels.
- Server-side conversion matching
- Alignment of bidding strategies
- Signals regarding creative performance, which are the most apparent.
The algorithm tends to value signals sent from the server 2-3 times more than data from pixels.
Engaging Through Email
I firmly believe that combining paid strategies with email marketing yields the best results. When effectively integrated, they provide high-quality engagement signals for retargeting purposes. A typical customer journey to consider includes:
- Cart abandonment for e-commerce
- Abandoned interest signals for B2B
It’s important to differentiate between abandoned cart/intent and abandoned checkout: at the checkout, you’ve already captured the email and the user’s intent is clear. For shops, the point of cart addition is critical, while for B2B, it might be browsing the pricing page.
Unfortunately, many overlook these strategies since they often require additional effort. However, if you can connect user sessions through their history, you can leverage cookies to recognize past interests and craft targeted email flows for those users. This approach is particularly underrated.
Bonus Tip: Integrate email engagement insights back into ad platforms through server events, which enhances targeting efficacy by around 25-30%.
In short, quality first-party data is essential for feeding into the algorithms of various platforms. With the right implementation, I’ve routinely witnessed a 2-3 fold increase in ROAS. Feel free to reach out for guidance on technical aspects; I might share a detailed post on this topic should there be sufficient interest!
, When you talked bout combining email marketing with paid ads, it reminded me how neglected email campaigns can be. Even in my expereince, simple email strategies boosted conversion by 20%. What are some commom pitfalls ppl fall into while trying to integrated their paid and email campaigns?