← Back to glossary
Audiences & Targeting
3 min read
Predictive audience targeting uses machine learning to reach people before they show obvious intent — forecasting who is likely to convert, churn or become high value from patterns in data. It moves targeting from reacting to past behaviour to anticipating future behaviour.
Learn from outcomes. Models study who converted and who did not.
Find leading signals. They spot early patterns that precede action.
Score new users. Each person gets a likelihood of the outcome.
Target the likely. Spend concentrates on high-probability audiences.
Likelihood to convert. Who is close to buying.
Churn risk. Which customers may leave.
Predicted value. Who is likely to be worth the most.
Next action. What a user is likely to do next.
Anticipation. It reaches people before they self-identify.
Efficiency. It avoids spending on unlikely prospects.
Value focus. It can prioritise future high-value customers.
Prediction is probabilistic. Forecasts are odds, not certainties.
Data hunger. Good predictions need rich, clean data.
Bias. Models can inherit skew from the past.
Transparency. Predicted audiences can be hard to explain.
Keep exploring
Browse all 61 advertising terms
→
OpenAds connects your assistant to every major ad platform, so you can plan, launch and optimise campaigns in plain language — approving the moves that matter.
Explore OpenAds