Bid Floors, Bid Shading, and Auction Dynamics in Healthcare Programmatic

A common moment in a pharma media review: the HCP campaign CPM is several times the DTC campaign CPM, and someone asks why. The honest answer usually involves auction mechanics. It helps to understand them, because the instinct to "just lower the bid" can quietly push delivery toward the worst inventory on the plan.

How first-price auctions work

When a page loads, the SSP sends a bid request to many DSPs. Each DSP decides whether to bid and how much. In a first-price auction, the highest bid wins and pays its own bid. In the older second-price model, the winner paid just above the second-highest bid.

The industry moved largely to first price around 2019. The change made pricing simpler to explain, but it shifted the job of finding the right price onto the buyer. Bid too high and you overpay on every win. Bid too low and you lose.

Bid floors: the publisher's minimum

A bid floor is the lowest price the seller will accept. Floors can be:

• Fixed: a set price for a placement or format
• Dynamic: adjusted by the SSP based on demand, time of day, and past bids
• Deal-specific: a negotiated price in a PMP or preferred deal
• Audience-informed: higher for impressions the seller knows are valuable, like logged-in HCP traffic on an endemic site

Floors are not visible to buyers in most open auctions. You infer them from lost bids and win rates. If your DSP reports many bids lost below the floor, your bids are under the seller's minimum on that inventory.

Bid shading: the buyer's correction

Bid shading is the DSP's way of not overpaying in a first-price auction. You set a maximum bid. The DSP estimates the lowest bid likely to win that impression and bids somewhere between that estimate and your max.

A hypothetical example: your max bid is $40 CPM. The DSP's model predicts the impression will clear around $28. It bids $30. If it wins, you pay $30 instead of $40. Across a campaign, that adds up.

Shading works on predictions, so it has failure modes:

• It can shade too hard on scarce inventory and lose impressions you needed
• It learns from history, so it can lag when competition changes, like during a competitor's launch
• Its logic is mostly not transparent to buyers

Check whether shading is on, and look at win rate alongside the average clearing price. Falling win rates with stable bids can mean shading is too aggressive for that audience.

Why narrow HCP audiences produce high clearing prices

Several forces push HCP prices up together.

So a high HCP CPM is often what a contested auction for a scarce audience looks like. Cutting bids in that setting does not make the audience cheaper. It makes you lose to competitors on the good impressions and win the ones nobody else wanted. The existing article on why a cheap CPM can produce an expensive pharma outcome goes into that tradeoff.

What buyers can control

1. Max bids by tactic. Set bids separately for the core HCP list, broader contexts, and retargeting. One bid for everything wastes money on easy impressions and loses hard ones.
2. Shading settings. Where your DSP allows it, review shading on scarce lines.
3. Deal types. PMPs and programmatic guaranteed deals with endemic publishers fix or cap the price and often give first look. The tradeoffs are in direct, PMP, or open auction for pharma.
4. Frequency caps. Tighter caps reduce bids on people already reached and spread spend across the list. See frequency caps in pharma programmatic.
5. Supply paths. Fewer, cleaner paths reduce duplicate bidding against yourself. The quality path optimization explainer covers how.
6. Timing. Pacing and dayparting affect which auctions you enter. An overpacing campaign can bid into expensive hours or cheap junk. See pacing in programmatic.

Reading auction data in a report

Ask for win rate, average bid, average clearing price, and bids lost to floor, by line item and by top publishers. Read them as a set. High win rate and falling CPM on unfamiliar domains suggests cheap supply, a quality question covered in the programmatic media quality guide. Low win rate and many floor losses on endemic sites suggests the bid is below the market for the audience you want.

Practical takeaway

For your main HCP line item, pull win rate, average clearing CPM, and bids lost below floor for the top 20 publishers last month. If the endemic sites show low win rates and high floor losses, consider a PMP with those publishers before lowering any bid.

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.