Building a DTC Acquisition Engine That Scales Beyond Legacy Channels
Several record-setting new-business months, on a broader and better-forecast acquisition engine.
Case study
Cost per acquisition down 30% and growth rates up 32%, on a proprietary cross-platform intelligence layer.
A valuable customer data set the business was not using to guide growth, and acquisition concentrated in lower-funnel search.
A specialty insurance organization held a large and valuable customer and prospect data set, and was not using it to guide growth decisions. Performance ran on traditional search marketing, broad targeting assumptions, and manual analysis. That worked to a point, but it kept the business concentrated in lower-funnel demand capture: measurable and efficient, and capped.
Reaching the next stage of growth meant a broader performance marketing model: better data, sharper segmentation, and predictive insight to reach higher-value audiences earlier in the journey. The challenge was never simply adopting AI. It was using AI and analytics to improve real decisions.
We served as an embedded executive growth partner, connecting strategy, data, analytics, media execution, and executive decision-making rather than treating AI as a standalone innovation project.
The centerpiece was a proprietary growth intelligence layer that brought performance signals together across search, paid social, and additional programmatic and audience activation platforms, so the team could compare signals instead of reading each platform in isolation. Around it, we gave hands-on guidance to the external media activation partner executing programmatic and paid social targeting.
Customer and policyholder data defined commercially meaningful segments rather than broad demographics.
AI and predictive analytics identified which audiences were most likely to convert and which carried the highest value.
Forecasting and scenario planning moved reporting conversations from backward-looking summaries to forward-looking decisions about where to invest, test, scale, or pull back.
Cost per acquisition fell 30% through AI-enabled segmentation and targeting, while growth rates rose 32% as the business expanded beyond search.
Cost per acquisition fell 30% while growth rates rose 32%, without losing accountability for results. Each is drawn from zero against where it started, at one scale.
The lasting asset is the intelligence layer itself: consolidated learning across platforms, owned internally rather than dependent on platform-specific reporting or agency-managed optimization.
The organization can now see which audiences are working, where the next growth pockets are, and how to move higher in the funnel, where attribution is harder and confidence is normally the constraint.
The business expanded beyond search, without losing accountability for results.
AI creates value only when it improves decisions.
The opportunity for a growth organization is not adopting AI tools: it is using AI to make better choices about customers, channels, creative, investment, and measurement. Applied that way, it sharpened segmentation and targeting, expanded the business beyond search, cut CPA, and raised growth: strategic judgment, operating discipline, and analytical rigor turning AI from a concept into a measurable growth advantage.
RelatedAI sales accelerationDemand generationFinancial services & insurance
Keep reading
Several record-setting new-business months, on a broader and better-forecast acquisition engine.
The book of business doubled during the engagement, run as a strategic growth channel.
Clear ownership of capabilities and of markets, and one way to prioritize the work.
Apply it to your business
Bring us the growth problem you cannot staff for.
Every month, one strategic idea from the operators who run our engagements — what's working in the field, what isn't, and the numbers behind it. Written for CEOs, not marketers.
The next issue lands the first Tuesday of the month. In the meantime, the archive is open — start with the one your pipeline needs.
Sent to
Read the archive