Pacing in Programmatic: Overpacing, Underpacing, and Even Delivery Explained

Pacing sounds like an operations detail. In practice it is where a lot of quality problems start. A campaign that overpaces often does so because it found cheap inventory nobody else wanted. A campaign that underpaces often gets "fixed" by loosening the controls that kept it clean. Knowing how to read pacing is the first step to not making either worse.

What overpacing and underpacing mean

Pacing compares two percentages: the share of the flight that has passed, and the share of the budget (or impression goal) that has been delivered.

• On pace: 50 percent of the flight is over and about 50 percent of the budget is spent.
• Overpacing: spend is running ahead. Fifty percent of the flight, 70 percent of the budget.
• Underpacing: spend is running behind. Fifty percent of the flight, 30 percent of the budget.

A pacing index puts this in one number: delivered percent divided by elapsed percent, times 100. In the overpacing example, 70 divided by 50 gives an index of 140. In the underpacing example, 30 divided by 50 gives 60. An index of 100 is on plan.

ASAP vs even pacing

Most DSPs offer at least two pacing modes.

Even pacing also usually smooths delivery within the day so the budget does not all go in the morning. Check the setting. Some platforms pace daily, some across the flight, and the difference shows up when a day or two is missed.

What causes overpacing

Overpacing usually means the campaign is winning more auctions than planned. Common reasons:

1. Bids set higher than needed. In first-price auctions, a high max bid with weak shading wins more often than intended.
2. Audience or inventory wider than intended. A targeting rule did not apply, a list expanded, or a broad PMP was attached.
3. Pacing set to ASAP by mistake. It happens more often than anyone admits.
4. A budget change mid-flight. The DSP recalculates and races to catch up on a new total.
5. Cheap supply showed up. A new path or a set of low-cost domains started clearing at your bid. This is the one to worry about, because it looks like efficiency.

The existing overpacing diagnostic for pharma campaigns walks through these in order. The main advice: check where the extra delivery went before cutting bids. If it all went to unfamiliar domains, a bid cut will not fix it. An inventory rule will.

What causes underpacing

Underpacing means the campaign is not winning enough. Usual suspects:

• The audience is small. A rare-disease HCP list of a few thousand NPIs may simply not generate enough daily impressions.
• Bids sit below what the auction clears at. Narrow HCP audiences are contested, and clearing prices tend to be high. The explainer on auction dynamics in healthcare programmatic covers why.
• Frequency caps are tight relative to audience size.
• Inventory or suitability rules exclude most of the available supply.
• Something is broken: creative not approved, a tag failing, a deal ID inactive, a flight date wrong.

The underdelivery diagnostic covers the order to check things. Setup errors come first because they are fast to rule out.

How to read a pacing report

A useful pacing report shows more than spend to date. Look for these columns, and ask for them if they are missing:

1. Spend and impressions to date versus plan, with the pacing index.
2. Daily spend trend for the last 14 days, alongside the total.
3. Win rate and average clearing CPM over the same period.
4. Spend by domain or app group for the last 7 days.
5. Reach and average frequency against the target list.

Read them together. A pacing index of 125 with flat CPMs and the usual domain mix probably means bids are a bit high. The same index with CPMs falling and new domains appearing means the campaign found cheap supply, and quality is the question. An index of 70 with a win rate under a few percent points to bids or audience size. An index of 70 with a normal win rate but few bid requests points to targeting being too narrow.

A hypothetical example: a 10-week HCP campaign with a $200,000 budget is at week 4 with $110,000 spent. Elapsed is 40 percent, delivered is 55 percent, index about 138. If the last week of domain data shows a jump in unknown sites, the fix is inventory rules. If the mix looks normal and CPMs rose, the fix is bids or pacing mode.

Why pacing ties back to quality

Pacing pressure changes behavior. Near the end of a quarter, an underpacing campaign tempts people to remove the blocklist, open the audience, or add an open exchange line. That can hit the number while buying the kind of inventory you spent the first half of the flight avoiding. The programmatic media quality guide lays out which controls are worth defending.

A cleaner approach is to agree upfront which levers can move to recover delivery (bids, caps, pacing mode) and which cannot (inventory and suitability rules). Write it in the brief.

Practical takeaway

Add one column to your weekly pacing report: spend by domain group for the last seven days. When pacing jumps or drops, that column will tell you whether the cause is price, audience, or inventory before anyone changes a 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.