Beyond the Funnel: Redefining Growth Metrics in the AI Era

Chief Growth and Marketing Officer Programme

Explore Programme

For years, the marketing funnel has provided a useful way to organise growth: create awareness, generate consideration, drive conversion and measure what happens at each stage. But AI is making the customer journey less linear, while giving marketing teams more ways to influence it.

Customers can now discover brands through AI-generated answers, compare options without visiting websites and receive increasingly personalised recommendations. At the same time, AI can analyse customer signals and change campaigns while they are running. The implication is significant: growth metrics for marketing leaders can no longer focus only on what happened at each stage of a predefined funnel. They must increasingly capture how demand is forming, where influence is occurring and whether that influence creates incremental business value.

Why the traditional funnel is becoming harder to measure

The marketing funnel is still useful, but it no longer captures the entire customer journey. AI-powered search and answer tools are changing how people discover and evaluate brands, often without visiting a company website. As a result, traditional signals such as traffic, clicks and conversions may not tell the full story. A drop in website visits, for instance, may not mean demand has fallen. Customers could simply be finding information elsewhere. For CMOs, the focus is shifting from tracking conversions to understanding where customer influence happens and how it contributes to business growth.

AI in marketing analytics changes the role of measurement

AI is giving marketing teams a much faster way to work with customer data. It can pick out patterns from large datasets and help teams make decisions sooner. But more data and better tools do not necessarily mean better measurement. BCG’s 2025 survey of more than 200 CMOs found that measurement ranked lowest among several investment priorities for improving digital customer experience and marketing technology. It also found gaps in linking marketing efforts to incremental sales. The bigger question, then, is not how much teams can measure, but whether they are measuring what really matters for growth.

From funnel metrics to connected growth signals

Going beyond the marketing funnel does not mean throwing out familiar metrics such as conversions, revenue or customer acquisition cost. These numbers still matter. The shift is towards understanding what happens around them. CMOs can look at brand searches, AI-generated mentions, customer engagement, retention, repeat purchases and lifetime value to get a fuller picture of growth. They can also track how quickly teams spot changing customer behaviour and act on it. The goal is not to add more numbers to the dashboard, but to understand which signals are actually driving business growth.

What should CMO growth KPIs measure next?

The next generation of CMO growth KPIs will need to look beyond immediate results. Marketing leaders must understand what drove growth, why it happened and how quickly teams can respond when customer behaviour shifts. AI can help uncover these patterns, but decisions still require human judgement. The real change is moving from fixed reporting to continuous learning. For CMOs, the challenge is connecting marketing activity with business outcomes while staying adaptable as technology, channels and customer expectations evolve. These capabilities are central to the ISB Chief Growth and Marketing Officer Programme, designed for leaders shaping growth strategies.

FAQs

What should marketing leaders measure beyond conversions?

They should look at demand, customer behaviour, brand influence, incremental impact and long-term value.

How is AI changing marketing measurement?

AI makes it easier to spot patterns, predict behaviour and optimise campaigns in real time.

Should CMOs stop using traditional KPIs?

No. Revenue, conversion, acquisition cost and retention still matter. The focus is on connecting these metrics with newer signals.