Most pharma campaigns are built to reach the entire market of potential prescribers. They use broad audience catalogues, which capture every health care professional (HCP) who prescribed in their pharmacological class in the last two years, regardless of whether they have a relevant patient today.
At best, the result is diluted impact. At worst, this leads to massive budget waste.
By analyzing electronic health record (EHR) and claims data, Wrango enables pharma marketers to reach HCPs who are actively managing therapy-relevant patient cases. Instead of targeting everyone who has prescribed in a category, marketers can focus on physicians currently diagnosing and treating qualified patients through Soon-to-Prescribe Audiences.
What are Soon-to-Prescribe Audiences?
Soon-to-Prescribe Audiences identify healthcare practitioners (HCPs) who are likely to prescribe a specific therapy class in the near term based on recent clinical activity from patients under their care.
These audiences are built using real patient-level clinical signals (recent diagnostic tests, procedures, and diagnoses) that indicate active management of conditions associated with a given pharmacological class. By analyzing the recency, combination, and clinical relevance of these events, patterns emerge that are strongly correlated with near-term prescribing behavior.
By linking these clinical signals directly to the treating provider, Soon-to-Prescribe Audiences surface HCPs who are actively managing patients whose clinical profiles align with the use of specific therapies. This shifts targeting from historical prescribing patterns or broad specialty assumptions to present clinical momentum.
This distinction becomes clear when you look at real-world prescribing volume. For example, we witness over 3,000,000 GLP-1 claims every month across our dataset. This is the outcomes data from which we reverse engineer Soon-to-Prescribe Audience targeting.

More than 3 million GLP-1 claims are processed monthly. By reverse engineering claims data, we identify the behavioral patterns and patient signals that precede prescribing decisions.
Why Historical Prescribing Data Falls Short
Prescribing behavior is uniquely event-driven. Contrast this with other types of consumers: someone who’s purchased running shoes from Nike before is likely to purchase from them again next time they’re in the market. But a physician doesn’t prescribe because they did so 18 months ago. They prescribe because a patient sitting in front of them today has a specific clinical need.
Take GLP-1 therapies as an example. Patients with overlapping aspects of metabolic syndrome—such as type 2 diabetes, hypertension, hyperlipidemia, obesity, and insulin resistance—have a strong association with GLP-1 prescriptions. A physician treating patients who exhibit several of these signals is significantly more likely to initiate GLP-1 therapy in the near term.
Timing sharpens this further. If a patient’s most recent Hemoglobin A1C is greater than 9.0%, they are, on average, roughly 45 days away from a GLP-1 prescription.
The result is an in-market audience built around current clinical momentum. Soon-to-Prescribe Audiences enable pharmaceutical brands to prioritize the HCPs most likely to prescribe next—improving campaign efficiency, accelerating therapy adoption, and driving measurable impact.

AI reveals strong overlap between metabolic syndrome indicators and GLP-1 prescribing. Hypertension, Type 2 Diabetes, hyperlipidemia, and obesity consistently lead confirming that prescribing is driven by active clinical signals, not historical scripts.

AI also surfaces non-obvious signals, like sleep apnea, that correlate with GLP-1 prescribing beyond traditional metabolic indicators.

Elevated A1C levels signal immediate treatment momentum. Patients with A1C ≥9.0% are significantly closer to GLP-1 initiation, allowing models to prioritize HCPs at the point when prescribing decisions are actively forming.
Why Soon-to-Prescribe Audiences deliver measurable impact
Soon-to-Prescribe Audiences fundamentally change how pharma marketers engage, measure, and scale. Key benefits include:
- Engaging prescribers at the moment of opportunity. Reach HCPs who are actively treating patients whose clinical profiles align with your therapy class during the window when treatment decisions are forming.
- Improving efficiency and relevance. Focus investment on providers with current clinical signals indicating near-term prescribing potential, rather than broad historical segments.
- Accelerating adoption among new prescribers. Identify future prescribers before the first prescription is written by analyzing the clinical patterns that precede therapy initiation.
- Powering targeting and measurement from the same source of truth. In pharma, claims data is outcomes data. By building audiences and measuring performance directly against real-world prescribing activity, campaigns are optimized to the business results that matter.
- Moving faster with centralized intelligence. When outcomes data is native to audience modeling, insights can be refreshed and distributed across channels without manual stitching or delayed validation.
The result is a marketing strategy aligned with real prescribing behavior, engaging the right HCPs, at the right moment, with measurable clinical impact.
In our GLP-1 example, a two-year prescriber lookback yields ~435,000 HCPs. That audience includes many physicians who simply do not have a patient need right now.
We tested applying:
- Recent patient-level clinical signals
- Diagnosis overlap and comorbidity patterns
- Severity indicators (e.g., elevated A1C levels)
- Recency-based timing windows
- New prescriber definitions tied to therapy initiation
In doing so, we reduced that pool to ~118,000 HCPs actively treating patients likely to initiate therapy in the near term. In this case, that meant avoiding approximately $285,798 in monthly spend directed at physicians who were not in-market..

Without patient-level signal refinement, most budget goes to historical prescribers. Using real-time diagnosis triggers shifts spend toward physicians actively managing therapy-relevant cases.
The Takeaway
Audience-based targeting flips the model.
Stop doing this:
- Targeting everyone who prescribed in the last two years
- Refreshing audiences twice a year
- Measuring CTR as a proxy for success
Start doing this:
- Defining audiences based on real-world clinical timing
- Modeling severity and recency signals
- Refreshing segments automatically
- Aligning spend to who is in-market now
The difference between “the market” and “in-market” is the difference between waste and efficiency.
Learn More
If your display budget is fixed each month, the question isn’t how many HCPs you can reach, but whether you’re reaching the right ones at the right moment.
Connect with our team to explore in-market HCP modeling powered by real-world claims data.










