From Hyperlocal to Blockbuster: Why Precision Strategy Scales Like a Startup

“Find me solutions, Ben. Be an entrepreneur. That’s why you’re on the team. Time to shine.”

Fair enough.

What follows is the result of that challenge, and years of work that came before it. At DOT I/O Health, my team and I learned how to dig for signals in data that others missed. We built high-definition patient populations that actually drove engagement and proved our methods could scale. Those lessons and approaches are the foundation of what’s now being taken even further at Adelo.

“We need to do more with less.”

That was the mission set by the head of a liver franchise at a major pharma company launching a drug in infectious disease. As the lead consultant, I dived deep with my team, commercially and scientifically.

The clues have always been in the data. They sit in the signals. They hide in the inferences. You just have to know where to look.

Over five years of working with teams from global to local level, launching brands or finding ways to drive growth, I noticed something fundamental shifting. Pharma companies were being forced to think more creatively about how to engage HCPs, KOLs, patients, and payers on the ground. Health systems were changing. Budgets were tightening. The old playbook was breaking down.

It was starkly different from the era where you could launch a drug with highly scalable plans and watch it become a blockbuster if the efficacy was strong, you had the investment, and you were early in class. That era is gone. Between 2007 and 2011, 80 drugs representing roughly 25% of US drug sales lost patent protection. Development costs now average $2.3 billion per new drug. The numbers don’t work the way they used to.

Pharma companies are getting more creative. But creativity isn’t scalable. And scalability without intelligence is just expensive activity.

Today’s real challenge is different: How do you differentiate on access and engagement when everyone is chasing the same treatment endpoints, the same patient populations, the same clinical pathways?

But here’s the thing: everyone’s not actually chasing the same patient populations. They think they are. But they’re not.

Looking for Clues

I’ve done this most of my career. Looking into data. Finding the connections that others miss.

Whether it’s identifying a value proposition for a product, understanding the real opportunity in a deal, figuring out what to emphasise in a platform pitch, or crafting a strategy that actually resonates in the market, it always starts the same way. You dig into the data. You look for the patterns. You ask the uncomfortable questions that the surface analysis doesn’t answer.

The data tells stories. But you have to know how to listen.

When we examined the infectious disease, my team and I combed through vast amounts of academic data: systematic literature reviews, lifting out every noteworthy signal. Risk factors for this disease (hepatitis) were well known. But this particular variant had less diagnosis, less understanding. The disease was hidden.

As I searched for clues, I saw the main at-risk areas. One was migrants. That made sense; their native countries sometimes had higher prevalence rates. But I kept digging.

What interested me was what the research showed: much of it focused on testing within a single country (US, Europe, UK and other large economies) and categorising patients by demographics, including place of birth.

And then the real question: where was the research not looking?

Hepatitis was often invisible, undiagnosed. There was potentially a large population who didn’t realise they had it. People unlikely to be in any screening programme.

The deeper I dug into migrant populations (and migrant movements are growing globally), the more the numbers startled me. Research found that between 1.04 and 1.61 million migrants in the US alone may have chronic hepatitis B infection, nearly twice previous estimates. Prevalence rates vary dramatically by region of origin: 10% from Africa, 7% from Asia, 5% from Oceania. Despite these elevated rates, migrants face substantial barriers to testing and treatment: cultural norms, language barriers, fear of discrimination, administrative obstacles.

And here’s what bothered me: the science stopped. Guidelines acknowledged migrants as a risk factor. That was it. No deeper dive. To be fair, that’s partly because the science wasn’t focused on quantifying the actual numbers or translating those insights into actionable strategy. The epidemiological approach, valid as it is, doesn’t naturally lead to the next step: where are these populations, and what do we do about it?

That’s where the value was. In what wasn’t being asked.

Building Our Own Model

With the company launching in major markets, I started thinking differently. Not about typical epidemiology studies, but about assembling our own intelligence model.

If we could focus our analysis properly, we’d at least know which countries to prioritise for launch. Or even more valuable: which regions within a country. Which cities. Which neighbourhoods.

My team and I assembled:

• UN migration data that gave us patterns of human movement.
• A meta-analysis paper examining hepatitis prevalence across countries that gave us regional disease burden.
• Migration patterns from high-prevalence countries to destination markets that connected the two.

From here, we calculated potential at-risk populations. At first, national level. The numbers were high, high enough that they demanded action. This wasn’t an epidemiological study in the traditional sense. It was a strategic intelligence tool designed to identify hidden populations so we could allocate resources where they’d have maximum impact.

Then I wanted to go finer. Census data. Granular geographic information. We drilled down from national to local levels, using geospatial analysis to map neighbourhood-level social determinants. The precision became actionable.

This is what I’ve always done. Connect dots others don’t see yet. Find the leverage point in the data. The insight that changes how you think about the problem.

The Patient Population Illusion

Here’s something most pharma companies don’t see: when they say “patient populations,” they usually mean clinical definitions. Inclusion criteria. Disease stage. Comorbidities. It’s low-definition. All patients with hepatitis B who haven’t been treated. That’s a patient population.

But that’s not precise. That’s not actionable. That’s not enough.

Our patient populations were way more precise. Way more relevant to everyone: HCPs, payers, patients themselves.

We didn’t just say “undiagnosed hepatitis B patients.” We said “migrants from West Africa living in the Hackney ward of London with low transportation access and no prior hepatitis screening, who work in precarious employment, and have language barriers to healthcare access.”

That’s high-definition. That’s specific. That’s actionable.

And it makes all the difference to effective implementation.

When a clinician sees a patient matching that profile, they know what they’re looking at. When a field team targets those neighbourhoods, they understand the context. When a payer designs a patient support programme for that population, they’re not guessing. They’re designing for the actual barriers those patients face.

Most pharma companies are still working with low-definition patient populations. It’s clinical, it’s broad, it’s safe. But it doesn’t drive engagement. It doesn’t change behaviour. It doesn’t move the needle on access.

High-definition patient populations—SDOH-informed, geographically specific, contextually rich—are the real differentiator. They’re the difference between saying “target hepatitis patients” and saying “these are the exact patients we’re trying to reach, here’s where they are, here’s what matters to them, and here’s how to engage them.”

Why Local Data Engages

Here’s what became clear through this work, and what the research increasingly confirms: people respond more powerfully to information that’s relevant to their specific geography.

That really surprised me. But maybe it shouldn’t have.

It’s not just about precision. It’s about relevance. Information near home gets a reaction.

Research examining geolocated news consumption found that residents who access local news proximate to their physical location are significantly more likely to engage with that news and their community. They consume more information, seek out more opportunities to engage, and participate more, not just passively, but behaviourally. People in geolocated communities voted more, volunteered more, acted more.

This connects to what psychologists call construal level theory. The closer something is to us, in space, time, or social distance, the more concretely we think about it. National statistics feel abstract. Data about your neighbourhood feels real. When you show a clinician data about their own catchment area, something shifts. It becomes their problem to solve.

Research in the Journal of Marketing demonstrates that geographic proximity strengthens social influence. People cooperate more with those they believe are near them. They’re more easily persuaded by proximate sources. For healthcare professionals specifically, physicians are more influenced by geographically and socially close prior adopters when making decisions about new approaches.

And in 2025, HCPs expect this. They expect content that reflects their unique clinical environment, their cultural context, their regulatory landscape. Generic national messaging doesn’t cut it anymore. Local precision is table stakes.

But here’s the difference: traditional patient population definitions don’t enable that. You can’t show a clinician “your hepatitis B patients” if you’ve only defined the population clinically. You can show them “undiagnosed hepatitis B patients of West African origin with low healthcare access in your practice postcode.” That’s different. That’s their patient population. That’s real.

SDOH: Environment Determines Health

This is where the picture becomes powerful. Social determinants of health (SDOH) aren’t just demographic variables. They define the environment in which health happens. And environment impacts health outcomes across virtually every disease and condition.

Research shows that SDOH factors (health behaviours, clinical care, social and economic status, physical environment) account for 30% to 55% of health outcomes. A comprehensive study across 29,126 census tracts in 499 US cities found that SDOH variables have statistically significant impacts on population health outcomes, though the magnitude varies by local context. Four risk factors alone (asthma, kidney disease, smoking, and food stamps) significantly affected outcomes across all city groups.

Poor housing links to poor health. Food insecurity links to chronic disease. Transportation barriers link to missed care. These aren’t correlations you can afford to ignore in the real world. They’re the lived realities that determine whether treatments actually work.

For pharma, understanding these environmental factors is no longer optional. Regulatory agencies and pharmaceutical companies increasingly utilise real-world evidence to understand how treatments perform across diverse populations in actual practice, not just controlled trials. Market access strategies in 2025 are built on demonstrating real-world value with stories that focus on meaningful outcomes and broader impact on healthcare systems.

When you understand the SDOH factors affecting your patient population, you can design engagement strategies that work with those factors, not against them. A patient population without transportation access isn’t reached through geographically distant screening clinics. A patient population with language barriers isn’t reached through English-only materials. A patient population with economic instability isn’t managed through expensive private support programmes.

High-definition patient populations tell you what barriers exist. Then you can actually solve for them.

Activity Data: Behaviour, Not Just Intent

Here’s what changes everything. Adding activity data (exercise, fitness, movement patterns) frames actual behaviour, not stated intentions.

A large-scale study from Israel’s largest healthcare organisation found that physical activity app users had significantly lower risk of cardiovascular disease (HR 0.95), stroke (HR 0.91), and type 2 diabetes (HR 0.82) compared to non-users. This wasn’t self-reported. This was objectively measured behaviour integrated with electronic health records over multiple years.

Wearable technology data predict clinical outcomes including readmission and mortality. A Fitbit model using step counts and patient activity predicted hospital readmission with 88.3% accuracy despite small sample sizes. This is precision that captures what people actually do, not what they say they’ll do.

For pharma, this is transformative. Activity data provides signals of disease progression, treatment adherence, and real-world effectiveness that traditional claims data will never capture. It shifts the conversation from efficacy (what happens in a trial) to effectiveness (what happens in the real world).

But more importantly for patient populations: activity data helps define which patients are likely to engage with which interventions. A patient population showing high digital health engagement is different from one showing low digital engagement. Same clinical criteria. Completely different engagement strategy required.

This is where the work my team and I have built at Adelo comes in. We’ve designed a platform specifically to integrate activity data with clinical understanding and SDOH context, creating what you might call “intent signals.” Behavioural intent, not just stated intent. This is the foundation for understanding real patient populations and how they’ll actually respond to engagement.

The Scientific Literature: Closing the Loop

Layer in scientific literature (the published research, the meta-analyses, the epidemiological signals) and you have something genuinely powerful. The science connects the data to clinical meaning. It validates the patterns you’re seeing in SDOH and activity data.

When we connected migration patterns to published hepatitis prevalence data, we weren’t inventing a hypothesis. We were operationalising existing science at a granular level that made it actionable. The literature told us migrants were at risk. Our data modelling showed us where those populations actually lived, in which neighbourhoods, in which postal codes.

Instinctively, this combination works. SDOH data reveals the environmental context. Activity data reveals behavioural patterns. Scientific literature provides the clinical validation. Together, they create a picture that’s both rigorous and relevant.

And together, they define a patient population that’s actionable.

The Fertile Land: Signals as Strategy

Many companies layer treatment data, scripts, and ICD-10 codes to build longitudinal patient pathways. That’s standard now. But how do you differentiate if everyone is doing the same thing?

The fertile land is in these upstream signals, the predictors and signposts that exist before diagnosis and treatment. Migration patterns. Census demographics. Activity levels. Environmental exposures. Social vulnerability indices. These signals don’t just inform strategy. They become the real-world evidence.

This is how you differentiate on access and engagement. Not with more data. With smarter data. Data that tells a local story grounded in science.

And most crucially: data that transforms vague clinical populations into precise, actionable, high-definition patient populations.

The Solution: What We Built at Adelo

What started organically at DOT I/O Health (piecing together migration data, census information, and scientific literature to find hidden hepatitis populations) evolved into something bigger. A framework. A toolset. An approach that could scale across diseases, across markets, across therapeutic areas.

At Adelo, we’ve taken that work further. We’ve built a platform designed to operationalise this exact approach.

The Intelligence Layer

The foundation is geospatial SDOH intelligence: mapping social determinants at the neighbourhood level, not just zip codes. Our approach integrates SDOH data with clinical information to create high-definition patient populations that are geographically specific and contextually rich. We combine:

• SDOH geospatial data at neighbourhood/census tract level.
• Migration and demographic patterns from census and population databases.
• Activity and behavioural data from wearables and digital health sources. This is where our platform adds real value. We’re not just collecting activity data; we’re translating it into intent signals.
• Scientific literature connecting risk factors to clinical outcomes.
• Claims and prescription data for treatment pathway validation.

This isn’t just aggregation. It’s integration, creating a unified view that reveals where hidden populations live, what environmental factors affect their health, and how behaviour patterns predict outcomes.

The Engagement Layer

Intelligence without engagement is just expensive reporting. The data has to flow somewhere actionable. The engagement layer connects the intelligence to the people who need it:

• HCP-facing tools that deliver local, relevant data at the point of decision.
• Field team enablement with geographic insights specific to their territories.
• Payer conversations grounded in local population health evidence.
• Medical affairs materials that reference neighbourhood-level risk factors.

In 2025, the most effective HCP engagement is personalised, timely, and contextually relevant. Real-time, hyperlocal data isn’t optional. It’s the foundation. The ability to show a clinician exactly what’s happening in their catchment area, with high-definition patient population data informing every conversation, transforms abstract marketing into actionable intelligence.

The Evidence Layer

Here’s where it gets strategic. The same signals that inform engagement become the real-world evidence that supports market access.

By converting varied social, economic, and behavioural data into practical insights (geospatial dashboards illustrating how social determinants differ across neighbourhoods, understanding which patient populations are most likely to engage, and designing interventions that work with actual behaviour patterns) organisations can identify high-value opportunities before traditional clinical endpoints emerge.

The signals become the proof. The engagement generates the outcomes. The outcomes justify the investment.

The Architecture

What does this look like in practice?

Layer 1: Data Integration
SDOH indicators (housing, food security, transportation, economic stability) • Geographic demographics (migration, ethnicity, language, age distribution) • Behavioural signals (activity data, digital health engagement, intent indicators) • Clinical validation (scientific literature, meta-analyses, epidemiological studies)

Layer 2: Intelligence Generation
Neighbourhood-level risk mapping • High-definition patient population identification (clinical + SDOH + geographic + behavioural) • Predictive modelling for disease burden and engagement likelihood • Local context scoring

Layer 3: Engagement Activation
Territory-specific HCP materials tailored to precise patient populations • Field team briefings with local intelligence and population insights • Payer evidence packages by region and patient population segment • Medical affairs content grounded in local data and population characteristics

Layer 4: Evidence Capture
Real-world outcomes tracking by patient population segment • Engagement effectiveness measurement • Population health impact documentation • Continuous model refinement

The Outcome

When this works, when the intelligence flows through to engagement and the engagement generates evidence, you get something powerful:

• HCPs engage because the data is about their patients in their geography. High-definition, specific, relevant. It’s not generic. It’s theirs.
• Field teams perform because they have context, not just talking points. They understand the local health landscape and the precise patient populations they’re trying to reach.
• Payers listen because the evidence reflects their population. You’re not asking them to extrapolate from national averages to low-definition patient categories.
• Marketing teams act because they have clear targeting priorities grounded in high-definition patient population data. No more spray-and-pray.
• Medical affairs validates because the local signals connect to published science and the patient population definitions are clinically sound and contextually rich.

Why This Will Work

The reason I’m confident this works for pharma is straightforward: I’ve done it. My team and I built it. We’re about to be published on a paper validating this approach. It’s not theoretical. It’s proven.

What we’ve learned at Adelo is that this framework scales beyond pharma. The principles that work for identifying hepatitis populations in London apply equally to fitness engagement, health and wellness, even sports performance. Activity data becomes intent. Geography becomes context. SDOH becomes the environment for behaviour change.

I can see how this works for the whole of health, fitness, and even sport.

The Differentiation

Everyone has claims data. Everyone can build patient pathways. That’s not differentiation anymore.

The differentiation is in seeing the difference between low-definition and high-definition patient populations. Between “hepatitis B patients” and “undiagnosed hepatitis B patients of West African origin in low-access neighbourhoods.” Between clinical definitions and SDOH-informed precision.

The differentiation is in the upstream signals: the SDOH factors, the migration patterns, the behavioural data, the scientific connections that reveal hidden populations and predict who needs intervention before they show up in the healthcare system.

The differentiation is in activity data that becomes intent. In geomapping that becomes engagement strategy. In platforms that turn data into action.

The companies that win will be those that see the signals others miss. That understand that patient populations aren’t just clinical categories. That connect strategy to implementation with high-definition patient data running all the way through. That turn local intelligence into local engagement into local evidence.

Scaling the Approach: What Works in Hackney Works in Toronto

Here’s what we discovered once we’d identified the target populations, mapped their locations, and developed the engagement method: the approach scales.

Once you’ve found the hidden population in one borough of London (a high-definition patient population, specific and contextual and SDOH-informed) you have a playbook.

That playbook works in a ward in Toronto. It works in a neighbourhood in Paris. It works in a district in Sydney.

Not because the populations are identical. They’re not. But the framework is the same:

• Identify the high-definition patient population using SDOH and migration data.
• Locate them at the neighbourhood level using census and geographic intelligence.
• Engage them using locally relevant messaging and culturally appropriate channels.
• Measure the response and refine.

Each location still has its own dataset. Its own demographic profile. Its own healthcare system nuances. Its own high-definition patient population characteristics. But the methodology transfers. The principles hold.

It was like scaling a startup. Who knew.

You prove the model in one market. You document what works. You identify the variables that matter. Then you replicate, adjusting for local context but keeping the core approach intact.

This is what makes the solution entrepreneurial, not just analytical. It’s not a one-off insight. It’s a repeatable system that gets smarter with each deployment.

The first time we ran the analysis took months. The second time took weeks. By the fifth market, we had it down to days, because we knew what to look for, we knew how to source it, and we knew how to turn it into engagement.

That’s the difference between a consulting project and a scalable capability. One gives you an answer. The other gives you a machine for generating high-definition patient populations that actually work.

Beyond Pharma: High-Definition Targeting for Consumer Health, Wellness, and More

What’s exciting about the approach my team and I have developed is how naturally this methodology extends beyond pharmaceuticals to consumer health, wellness, and the modern “health experience” marketplace.

When budgets are tight, penetration and engagement matter more than ever. High-definition audience insights, built from SDOH, geography, and behavioural intent, are the solution not just for patient targeting, but for getting the right message to the right consumer without wasted spend.

Here’s how this framework plays out outside of pharma:

Skincare/Beauty: Precision for K Skin or Similar Brands
Rather than generic influencer outreach or broad social ads, brands like K Skin can identify micro-communities: urban neighbourhoods where SDOH data shows a higher prevalence of air pollution or UV exposure, combined with social listening that flags rising skin health conversations. Campaigns become locally relevant (“How London’s air affects your glow,” “SPF for Dallas city runners”), driving both product trial and genuine engagement.

Health Insurance: Building Trust and Community
An insurer looking to promote preventive health products or digital platforms can use this model to spot neighbourhoods underserved by traditional clinics but active on fitness apps. Outreach leverages local gyms, faith-based groups, or neighbourhood apps, not just blast emails. This builds not just conversion, but community loyalty and higher NPS.

Sports Partnership and Consumer Clothing (e.g., ALO)
A sportswear company like ALO can use local engagement analytics to identify city districts with high social media yoga engagement and low gym membership, then launch pop-up fitness experiences, offer digital wellness content, or even use census SDOH data to guide where new “ALO Experience” studios might thrive. For sponsors, localising activation (using micro-influencers whose personal stories match the SDOH profiles of a community) creates authentic resonance rather than shallow reach.

Community Health and Wellness Startups
Wellness startups focused on nutrition, physical activity, or digital communities can geo-target programs based on where food deserts or low activity scores cluster together. Neighbourhood ambassadors can lead the charge, and measurement of impact is built in. For example, launching a virtual healthy eating challenge just for high-risk postcodes, then tracking participation and wellbeing outcomes using wearable data.

In all these cases, the secret is the same: Moving from vague, broad mass-marketing or outreach to high-definition, hyperlocal, relevance-driven engagement. The data signals (the same signals we used for hepatitis launches) guide message, channel, timing, even community-building strategy.

And the benefit is twofold:

• Stretch every marketing £ or $: No wasted mass campaigns. Each piece of spend goes further.
• Build authentic community: Customers feel seen. Engagement rates go up. Loyalty and word-of-mouth follow naturally.

What began as a tool for smarter drug launches now helps brands, sponsors, and health innovators actually break through, even when the noise is high and the budget is limited. It scales up, but even more importantly, it scales down to where relationships and results are real.

What This Means

The solution that started organically at DOT I/O Health (from a directive to do more with less) is now a scalable framework embedded in a platform at Adelo.

SDOH geomapping isn’t a one-off project. It’s an operating model. A way of seeing healthcare markets (and wellness, fitness markets) that reveals what’s hidden and makes the invisible actionable.

But here’s what makes it different: data connects strategy to implementation.

That sounds obvious. It isn’t. The strategy-execution gap is one of the most persistent problems in business, and pharma and healthcare are no exception. Research shows that around 70% of hospital strategic initiatives fail. In pharma, the disconnect between headquarters strategy and field execution remains a “yawning gap,” with studies showing 73% of brand interactions with HCPs are unsynchronised. Strategy gets developed at the top. It cascades down. Context gets lost. The field team receives assets without understanding how they fit together or how to adapt them locally.

For years, the conventional wisdom was that you couldn’t boil the ocean. You had to pick your battles, narrow your focus, accept that comprehensive was impossible. A McKinsey consultant told me that once. It was true at the time.

It’s not true anymore.

With today’s technology (data infrastructure, AI, geospatial analytics) you can boil the ocean. You can process migration patterns across dozens of countries, overlay census data at neighbourhood level, integrate activity signals from millions of devices, and connect it all to scientific literature in real time. What used to take a team months now takes days. What used to be impossible is now operational.

The traditional approach treated strategy and implementation as separate phases, separate teams, separate problems. But they’re not. When data runs through both, when the same intelligence that informs the strategy also powers the engagement tools, when the signals that identify hidden populations also guide the field conversations, the gap closes. Strategy and implementation become one continuous flow.

This is what we’ve built. Intelligence that doesn’t stop at the PowerPoint. Data that travels from insight to action to evidence. A system where the same high-definition patient population that shaped the strategy is the same one the field team sees when they walk into a meeting.

That’s the difference between a strategy that lives in a document and one that lives in the market.

This is what I do. I look for the signals. I find the connections. I turn data into strategy. And then I figure out how to scale it, with strategy and implementation connected all the way through.

The difference between low-definition and high-definition patient populations is the difference between guessing and knowing. Between broad engagement and precise engagement. Between activity and impact.

The fertile land is there. The signals are waiting. The question is whether you’re looking.

The signals have always been in the data. They’ve always been there. In migration patterns. In census records. In activity streams. In scientific literature connecting those dots.

The companies that will win aren’t the ones with the biggest budgets or the most salespeople. They’re the ones that learned to see what’s already there: to find the hidden, high-definition patient populations, to understand the local context, to connect strategy to implementation with precision and speed.

And then scale it. Market by market. Borough by borough. Ward by ward.

That’s the real convergence. Not of technologies or datasets. Of clarity and execution. Of intelligence and action. Of low-definition thinking transformed into high-definition results.

Build from the ground up. Scale like a startup. That’s how you create a blockbuster now.