Healthcare Programmatic Targeting Options Compared

A targeting option is a bet on a signal. NPI targeting bets that a matched device belongs to the prescriber on your list. Contextual bets that someone reading about psoriasis treatments is more likely to have psoriasis. Each bet has a different failure mode, and in healthcare the failure modes include legal ones. The healthcare programmatic guide explains where targeting sits in the bigger picture. This piece compares the options directly.

The eight healthcare programmatic targeting options at a glance

The ratings are judgement calls, not benchmarks. A specific vendor's condition audience could be better or worse than the row suggests, which is why provenance checks matter more than category labels.

NPI targeting: precision with a denominator problem

NPI targeting starts from a list of prescribers, usually ranked by decile, and asks a data partner to match them to cookies, mobile IDs, emails, or other identifiers. When the match is good, this is the cleanest HCP tactic available, and it supports NPI-level measurement. The weak points are the match itself and the way reach gets reported. A 70 percent match rate says nothing about how many of those matches are right, and reach reports that use matched NPIs as the denominator flatter the result. The NPI targeting guide covers the mechanics and the questions to ask.

Condition audiences: useful, and the riskiest data you will buy

Condition audiences are consumer segments built from purchase data, surveys, modeled signals, or other sources. They give DTC brands addressability outside endemic sites. They also put the brand in the path of state consumer health data laws, platform health ad policies, and plain reputational risk. Before using one, ask where the signal came from, whether the source collected consent for this use, how old the data is, and which states are excluded. Condition audience targeting for DTC goes through those questions in detail.

Contextual and endemic: targeting the place, not the person

Contextual targeting picks impressions based on what is on the page or in the video. Endemic buying picks publishers whose whole audience skews medical. They are related but not the same, and the difference matters for budget. Contextual can run across the open web at open-web prices; endemic is a narrower, pricier set of placements with stronger audience composition. For consumer campaigns, contextual has become the default first layer for many brands because it avoids most person-level data questions. The how-to is in pharma contextual advertising, and the budget tradeoff is in endemic vs. non-endemic media.

Geography, first-party, lookalikes, and retargeting

Geographic targeting

Geography alone is a weak audience signal, but it is the backbone of hospital marketing and a useful layer for pharma when formulary access, disease prevalence, or field coverage varies by region. It also makes the cleanest test designs, because ZIP codes and DMAs can be held out.

First-party data

Pharma first-party data usually comes from patient support programs, co-pay card enrollments, and website registrations. It is accurate and small, and using it for media depends entirely on what the consent language allowed. The common mistake is running acquisition media against people who are already on therapy.

Lookalikes

Lookalikes trade precision for scale. If the seed is small or sensitive, the model has little to learn from and may learn the wrong thing, such as age and device type rather than condition. Treat lookalikes as an experiment with a control, not a default.

Retargeting

Retargeting site visitors works well in other categories. On a health website, the pixel that builds the audience can itself be the compliance problem. For HIPAA covered entities, HHS guidance on tracking technologies makes this a legal question first. For pharma brand sites, the state laws and pixel governance questions apply.

How to choose a targeting mix

1. Name the audience unit: NPI, person, household, or ZIP code.
2. Name the outcome and how it will be measured. Pick targeting that can produce a usable exposure file for that method.
3. Rule out options that fail privacy or legal review for your advertiser type.
4. Pick one precision tactic (NPI, endemic, or first-party) and one scale tactic (contextual or geography).
5. Decide what you will compare against. A contextual baseline is a good control for a paid audience segment.
6. Set exclusions: current patients, non-target specialties, states where your data use is restricted.

Practical takeaway

Take your current plan and write each tactic into the table above with your own ratings for precision, scale, privacy risk, and measurement fit. Any tactic rated high on privacy risk and weak on measurement is the first one to cut or to test against a contextual control. For HCP plans, pair that review with an HCP target list audit before the next flight.

Editorial note. Analysis and frameworks are the author's own and do not represent Acxiom or any current or former employer, client, or named platform. Examples labeled hypothetical or illustrative are not results from real campaigns. Nothing here is legal, regulatory, or medical advice.