Respond to a Utilization Baseline Request

A benefits leader or plan director sometimes calls asking for a savings number before anyone has agreed on the utilization data behind it. As a market access manager, you are the one who has to slow that request down without sounding evasive or unhelpful. How you handle this exchange decides whether the account trusts your numbers later, or starts treating every estimate you give as a guess dressed up as a fact.

Why This Conversation Goes Wrong

The pressure in this call is real. The leader has a finance meeting coming up. They want one clean figure to bring into that room, not a discussion about data sources. If you hand over a number without a baseline attached to it, that number gets repeated in a slide deck, then in a budget request, then in a renewal conversation a year later, long after anyone remembers it was a placeholder.

The failure mode is almost always the same. A manager wants to be responsive, so they run a model using a generic or national utilization rate because the plan has not pulled its own claims data yet. The output looks precise because it has a dollar sign in front of it. Nobody in the room questions a specific number the way they would question a range. Once that figure exists, it becomes the plan's expectation, and any later number that differs from it looks like backpedaling, even if the later number is the accurate one.

The fix is not to refuse the conversation. It is to change what the conversation is about. Instead of defending a savings figure, you walk the plan through what a figure would need to rest on for their specific population.

How To Prepare Before The Call

Before you get on the phone, get clear on a few things so you are not improvising under pressure.

1. Know what utilization data actually exists for this plan right now, and what stage it is in. Claims data that is still being adjudicated is not the same as a finalized baseline, and you need to be able to say which one you have.
2. Know who owns that data on the plan side. If the plan has a data or analytics lead, name them on the call. Redirecting to a person is more concrete than redirecting to a process.
3. Know whether this is an exploratory conversation or a formal budget review. The two calls sound similar but call for different levels of caution. A formal review with real dollars attached deserves more rigor than a first exploratory chat.
4. Have your approved economic model or materials ready to reference, with the assumptions clearly labeled as assumptions, not as this plan's results.
5. Decide in advance what you can commit to today, such as a scheduled review once claims data is available, so you are not inventing a timeline on the spot.

If you want to rehearse this kind of pressure before it happens with a real client, DealSpeak's Pharmaceuticals practice conversations let you run this exact scenario against an AI counterpart that pushes back the way a real benefits leader would.

What To Say

A benefits leader opens with, "Just give me the annual savings number so I can bring it to finance next week."

You respond, "I want to give you something finance can actually use, and a single number without your utilization baseline could send them in the wrong direction. I can walk you through the approved model and show every assumption we would need to plug in for your population. What utilization data does your finance team currently treat as reliable?"

The leader says, "Honestly, our claims data will not be finalized for another six weeks. I was hoping we could work around that."

You answer, "Then let us use these six weeks well. I will send over the model with the assumptions clearly marked as placeholders, so your team can see exactly what would change once real numbers come in. Once your data owner has that baseline, we schedule a short review together and put a number to it that will actually hold up in front of finance."

Notice what happened in that exchange. You never said no. You gave the leader something to bring back today, a documented framework with visible assumptions, while refusing to let a placeholder pass as a forecast for their plan. You also asked a direct question about their data instead of just stating a policy, which keeps the conversation collaborative rather than defensive.

Mistakes To Avoid

A few habits sink this conversation even when the manager has good intentions.

1. Do not fill a data gap with a convenient estimate just to keep the call moving. A friendly guess today is a disputed number in six months.
2. Do not present a general model as a forecast specific to this plan. Say plainly that the model is a framework, not a prediction, until real utilization numbers go into it.
3. Do not promise a savings figure or a percentage reduction. Promising an outcome puts you on the hook for a result you do not control.
4. Do not let the leader's urgency become your urgency. A finance deadline on their side is not a reason to skip a step on yours.
5. Do not end the call without a concrete next step. Vague reassurance without a scheduled review date reads as a dodge, even when your reasoning was sound.

This same discipline, being clear about what you know, what you do not, and where the line sits, shows up across other tough pharmaceutical conversations. The same instinct to protect a boundary rather than fill silence with a comfortable answer is covered in Set Scientific Boundaries in an Advisory Discussionand in Redirect an Identifiable Case Question to General Evidencewhich deals with a related pressure to answer with more specificity than the evidence actually supports.

How To Practice And Score It

Reading through the right answer is not the same as producing it live while someone is pushing you for a number. Practice this with a colleague or a manager who plays the impatient benefits leader and interrupts you with requests for certainty, such as "just round it for me" or "finance will not accept a range."

When you review the practice run, check for specific behaviors rather than a general impression. Did you name the missing baseline early, in the first response, rather than after being pressed twice. Did you label every assumption as an assumption instead of letting the model's output speak for itself. Did you ask who owns the data on the plan side, and did you get a real answer rather than a shrug. Did the call end with an actual date or trigger for the next review, or with a vague promise to follow up sometime.

DealSpeak scores this kind of exchange against those same behaviors, so you can see exactly where a real conversation would have drifted into a premature number or a vague answer, and rerun the scenario until the habit holds under pressure.

A savings number said the wrong way at the wrong time causes more damage than saying nothing at all. Hold the line on the baseline, give the plan something concrete to work with in the meantime, and the number you eventually deliver will actually survive contact with finance.