Introduction

Our conversion optimization algorithm ensures that advertising budgets are allocated to the most effective placements, where they are likely to drive meaningful engagement and conversions. Our default bidding model focuses solely on click-through rate (CTR), but our conversion algorithm follows a multi-step funnel approach, progressively optimizing toward conversions based on real user engagement. This document explains:

• How our conversion optimization works
• The key event tracking methods available
• Best practices for setting up an effective conversion funnel

How our conversion optimization algorithm works
The optimize based on conversions feature is designed to gradually refine ad targeting and budget allocation by analyzing a combination of click data, engagement signals, and conversion events. The algorithm follows a progressive learning process to improve performance over time.

Funnel optimization steps

CTR optimization
The system prioritizes ads that drive high click-through rates to ensure engagement.

Engagement filtering
Time spent on site and other quality signals help identify valuable traffic.

Soft conversion goals
Events such as scrolling and form starts indicate user interest and purchase intent.

Final conversion goals
The algorithm shifts budgets toward placements driving actual sign-ups, purchases, or other primary goals.

Setting up a funnel-based conversion approach
To make full use of optimize based on conversions, advertisers should implement a multi-stage conversion funnel that feeds valuable data to the algorithm. This approach ensures optimization begins early, even before hard conversions occur.

Conversion goal types on our platform

How the algorithm uses them

• Soft goals (time spent, scroll depth, button clicks) help refine traffic early.
• Primary goals (sign-ups, purchases) are the final optimization stage.

Example: using a funnel approach for lead generation
A company running a lead generation campaign wants to optimize toward sign-ups for a free trial. Instead of optimizing directly for sign-ups (which may take time), the campaign follows a multi-stage funnel:

Optimization logic

• The algorithm prioritizes ads that drive CTR, focusing on creatives with high engagement.
• It then filters for placements that drive quality traffic, identifying high-value media sites.
• If users interact (scrolling, form starts), the campaign shifts budget toward those placements.
• Finally, it allocates more spend to sites that drive sign-ups.

Best practices for optimizing conversions

Use a funnel approach
Don’t just track final conversions. Optimize for soft goals like time spent and scroll depth. This ensures the algorithm learns early in the campaign.

Set multiple conversion goals
Define both engagement-based and final conversion goals. Arrange them in priority order, with soft goals first and the final goal last.

Allow the algorithm time to learn
Avoid frequent manual changes. The system needs data to optimize. Let campaigns run for at least 7–14 days before adjusting budgets.

Enable server-side tracking (S2S)
If browser-based tracking is unreliable due to cookies, Intelligent Tracking Prevention (ITP), or ad blockers, server-to-server (S2S) tracking can be used as a fallback or alternative to ensure more reliable event measurement.

Conclusion

Our conversion optimization algorithm follows a structured funnel-based approach, ensuring that ad spend is directed toward the most effective placements rather than just users who are most likely to convert. By setting up multi-stage conversion goals and allowing the algorithm to learn over time, advertisers can achieve higher engagement, better ROI, and lower acquisition costs. If you have any questions regarding this topic, reach out to us through the platform chat.