03Lifecycle and segmentation
Droice Labs
- Role
- Marketing Strategy Consultant, Team Lead
- Dates
- September 2023 – April 2024
- Location
- New York
I led this engagement. Droice Labs is a healthcare data analytics company, and I owned customer lifecycle and segmentation strategy for the go-to-market motion behind Droice Hawk, with an 8-person cross-functional team reporting into it.
Healthcare AI is a crowded market in which everyone describes themselves in roughly the same language. That was the constraint the whole strategy had to work against: not a lack of prospects, but a lack of any signal separating one cohort of them from another.
+40%
Digital engagement, alongside brand awareness up 30%
−25%
Customer acquisition cost, with marketing spend down 20%
20
Competitors benchmarked on positioning, in a converged market
Who the buyers were
Lifecycle and segmentation strategy
I led an 8-person cross-functional team through identifying and prioritizing the high-value healthcare cohorts — not just which segments existed, but which ones were worth the motion, so prioritisation was an output of the analysis rather than an opinion applied afterwards.
The go-to-market motion was addressing healthcare buyers as one audience, which meant every message had to be general enough to fit all of them and was therefore specific enough for none.
Digital engagement rose 40% and brand awareness 30%. Both followed from the same shift: messages written for a cohort rather than for an average.
+30%
brand awareness
Read the approach
Problem
Segmentation is the part of a go-to-market motion that is easiest to skip and most expensive to skip. Treating healthcare buyers as one audience produces messaging that is technically true of everybody and compelling to nobody, and the symptom is not silence — it is engagement that never rises no matter how much is spent on it.
Approach
I ran the lifecycle and segmentation work with an 8-person cross-functional team, identifying the high-value cohorts and then ranking them, which is the half that usually gets dropped. A list of segments tells you the market has structure; a ranked list tells you where the motion should go first.
Each cohort was defined by what it actually does — how it buys, on what cycle, answering to whom — rather than by the firmographics that are easy to collect and weak at predicting anything.
What it produced
Digital engagement rose 40% and brand awareness 30%.
Neither number came from spending more. They came from the same shift: once the cohorts were ranked, a message could be written for one of them and take the risk of not fitting the others, which is the only way a message in a crowded market gets noticed at all.
Testing the messaging
A/B testing on messaging and funnel performance
Messaging decisions were being argued rather than measured, and acquisition cost was the number carrying the consequence.
−20%
marketing spend, while acquisition cost fell 25%
I ran A/B tests on messaging and on funnel performance in Google Analytics, so a claim about what worked had a result behind it instead of a preference.
Customer acquisition cost came down 25% and marketing spend 20% — cheaper acquisition on a smaller budget, which is the harder of the two to achieve.
Read the approach
Problem
Every organisation has opinions about which message works. Very few have evidence, and in the absence of evidence the opinion that wins is the one held by whoever is most senior in the room. That is a fine way to run a meeting and a poor way to run acquisition.
Approach
I ran A/B tests on the messaging itself and on funnel performance, in Google Analytics, so that each claim about what worked had a measurement behind it.
The funnel tests mattered as much as the message tests. A message that wins on click-through and loses further down the funnel has not won; it has moved the problem somewhere less visible, and only instrumenting both ends catches that.
What it produced
Customer acquisition cost fell 25% and marketing spend fell 20%.
Those two together are the result worth stating carefully. Acquisition cost usually falls because spend rises — buying better placement, more volume, more frequency. Here it fell while the budget was also cut, which means the gain came from the messaging rather than from the money behind it.
The eight-step strategy
Digital strategy from a 20-competitor benchmark
In a crowded healthcare AI market, a digital strategy written without knowing how twenty other companies already position themselves is a guess.
+35%
website traffic, from the eight-step strategy
I benchmarked 20 competitors' positioning and ran a communication-pattern analysis across them, then built an eight-step digital strategy on what that left unclaimed.
Website traffic rose 35%. The eight steps were sequenced, not parallel — each one assumed the previous had landed.
Read the approach
Problem
Healthcare AI is a market where the language converges. Companies describe themselves in the same register, claim the same categories of benefit, and compete for the same search terms, which means a strategy built only on what you want to say lands on top of what twenty other companies are already saying.
Approach
I benchmarked 20 competitors on positioning and ran a communication-pattern analysis across them — not to copy what worked, but to find what the market had left unsaid, because in a converged market the unclaimed position is the only one available.
The eight-step strategy was built on that gap, and sequenced rather than parallel: each step assumed the one before it had landed, which is what makes a strategy testable at eight points rather than only at the end.
What it produced
Website traffic rose 35%.
The sequencing is the part I would carry to the next engagement. A strategy that runs its steps in parallel produces one verdict at the end and no way to know which part earned it. A sequenced one tells you where it worked and where it did not, which is the difference between a result and a repeatable method.
Built with
- Google Analytics
- Segmentation and lifecycle analysis
- A/B testing
- Competitive benchmarking