What Is Conversion Signal Optimisation (& Why It Matters)
First thing to say: we made this phrase up. We're big believers that Conversion Signals are not a one-and-done setup, but rather needs continuous monitoring and optimisation. Read our thinking and see if we can convince you!
Date
29 Oct 2025
Topic
Google Ads
Reading time
5 Minutes
Wait, What's A Conversion Signal?
Signals describe the various different data that inform AI decision making (or to be more precise: machine learning). As digital advertising largely operates using automated algorithms, they optimise through crunching massive datasets and improving thier performance through incremental learning. The basis of this is the optimisation goal, or the conversion. When selecting a conversion and bid strategy, it's like the advertiser is selecting an algorithm. Once the conversion starts to collect data, that algorithm is observing a multitude of conversion signals, testing often thousands of iterations of bid, ad placement and creative, then putting all of this together to optimise for performance. These algorithms are hard workers!
Why Conversion Signal Optimisation?
Conversion Signal Optimisation is the process of maximising campaign performance through the selection, implementation and ongoing improvement of advertising conversions. It provides an opportunity for competitive advantage, as much of advertising is now automated; for example bid management and creative testing. We find many of our clients are less concerned about A/B testing thier ad copy than they are about their conversion signal quality. We believe this trend in automation will only continue, for example as advertising platforms are increasingly using conversion signals and on-site data to create ad copy and personalise messaging to different users.
How Does Conversion Signal Optimisation Work?
We see the process as having three parts:
Data quantity - i.e. are there enough conversion events to hit the minimum or ideal threshold (Google advises at least 30 conversions account-wide per month, or 50+ for ROAS bidding)
Data match quality - i.e. the presence of metadata within each conversion, such as email or click ID, that are used for matching
Prospect match quality - i.e. the relevance of the conversion user signal to the ideal prospect.
The first two steps are more technical, as they require improvement in data collection. This requires knowledge in data collection and quality, such as utilising Google Tag Manager and ad platform data management tools, such as Google Ads' Data Manager and Meta's Event Manager.
The last step is more creative as it involves the ongoing decision around what data to mark as the primary conversion goal of a campaign, plus what value to give it. It's a surprisingly manual process and something that is much harder for a machine to understand. We find the best process is to observe the outputs carefully and continuously - meaning observe what the characteristics of the conversions are and decide if they match what your ideal outcome - and adjusting your conversions appropriately. This is something we cover in our Audience Optimisation process.
Still Don't Get It?
We'd love to give you an explanation of how your conversion signal decisions could be holding you back. Please reach out to us and we can arrange a consultation.
