The Data Behind the Goal: How to Pressure Test Targets Before Leadership Signs Off

Setting achievable goals is a foundational life skill. In a business context, it’s critical; the goals you set ripple out to other teams and stakeholders who plan their own work around them. Get them wrong and you’re sending people in the wrong direction.

I’ve seen what happens when a goal is set without real alignment on the data behind it. There is confusion when the quarter doesn’t go as planned, finger-pointing between teams, and a scramble to explain a miss that everyone could have seen coming if the goal had been pressure tested up front.

For medical device, healthcare, and other privacy-first brands, the stakes on that miss are higher than they are in other industries. Data is constrained by privacy regulations that limit what you’re allowed to measure in the first place. A goal that would be merely aggressive in e-commerce can be arithmetically impossible here, and it’s crucial to catch that difference before pressure testing.

Ideally, defining and pressure testing a goal isn’t something you do after leadership hands down a number. The strongest version of this process has data and analytics in the room before the goal is set, shaping your understanding of what’s realistic in the first place. Here’s what that looks like broken into six pillars, each one building on the data foundation you started with.

1. Definition: Name the North Star Metric, and Name What It Costs

Every goal needs one metric the team treats as the answer. That’s your North Star, and getting crisp on it is the first pressure test, because a goal built on a fuzzy metric can’t be tested at all.

Crisp means three things:

• One metric, stated in the exact terms it will be measured in. “Qualified leads” isn’t crisp. “MQLs that clear the current scoring threshold and are accepted by sales within 14 days” is.
• One owner and one source of truth. If you asked marketing, media, and sales for last month’s lead number right now, you might get three different answers, all defensible. One team is counting form fills, one is counting platform-attributed conversions, and one is counting records that reached the CRM with an owner assigned. Pick one, name who owns it, and write down where it comes from. Otherwise you don’t have a North Star yet.
• One honest statement of what it doesn’t capture. Say the North Star is new patient appointments booked. That metric says nothing about whether the patient shows up, what payer they came in under, or whether the appointment leads to the procedure the service line is actually trying to grow. None of that makes it the wrong North Star. It just means the things it can’t see need to be named now.

Naming a North Star is a decision to optimize something, and that is always also a decision to accept pressure somewhere else. Push hard on lead volume and you loosen the top of the funnel; close rate and cost per acquisition feel it immediately. Push hard on pipeline efficiency and volume flattens. The time to say it out loud is before leadership signs off, not in the quarterly review when someone notices that leads doubled while conversion halved.

So we name secondary metrics alongside the North Star, every time, and we state what we expect to happen to them. Some are guardrails: metrics we’re willing to see move, but only so far, with a floor written down in advance. Others are simply critical in their own right, things like revenue quality, patient acquisition cost, or branded search volume, and they get tracked with the same discipline as the North Star even though they aren’t the number on the wall.

The output of this step is a sentence anyone in the organization can repeat: we are optimizing X, we expect Y to move in this direction by roughly this much, and if Y crosses this line we stop and revisit. If you can’t write that sentence, the goal isn’t ready to pressure test.

Privacy-first brands have an extra consideration here. A North Star you can only measure partially puts the entire approach on estimated ground. Sometimes the right call is a metric slightly further from the business outcome you ultimately care about, but one you can measure cleanly under your consent and tracking constraints, with the outcome metric held as a critical secondary.

2. Historical Data: Build the Forecast, Then Flag Which Parts Are Estimated

Goals are almost always tied to a forecast or projection, and a good forecast starts with the historical data you already have. You don’t need a PhD in statistics to build something useful. Projections based on historical data can range from a simple growth factor applied to last year’s numbers, all the way up to a full regression model accounting for a dozen variables. (If you’re deciding between measurement approaches for a forecast, we’ve written about how to choose between MMM and MTAdepending on the question you’re trying to answer and why that choice looks different when third-party tracking isn’t available to you.) Either approach will tell you what you want to know, what you’ve accomplished, and what’s realistic to expect on the horizon.

Again, there’s a wrinkle here that marketers outside of privacy-first industries don’t have to think about. The historical data privacy-first brands are forecasting from is often incomplete by design. Consent requirements, restrictions on what can be tracked on patient-facing pages, and platform-level limits on MedTech and healthcare audiences can all leave measurement gaps. If your baseline was assembled from a partially instrumented measurement stack, your forecast inherits every one of those gaps. That means you need to know which parts of your history are solid and which are estimated, and to say so out loud before anyone commits to a number built on top of them.

With the North Star and its secondaries defined, the forecast has something specific to point at. Build it for the North Star first, then build a rough expectation for each secondary metric too. You don’t need the same rigor there, but you do need a number because “we expect close rate to dip” is not something you can measure against later. “We expect close rate to dip from 22% to somewhere around 18%” is.

Whatever data and assumptions you use to build the forecast, write them down. That sounds obvious, but it’s the piece that gets skipped most often, and it’s the piece you’ll need most later. Three months from now, if performance is off track, the assumptions you wrote down will help you explain why instead of just guessing.

3. Context: Test Whether the Goal is Physically Possible in Your Market

Historical data tells you where you’ve been. Context tells you whether it’s possible to go where you’re headed.

Context always matters, but it matters even more when you don’t have historical data to lean on, especially in the case of a new product launch, a market you haven’t entered before, or a channel with no track record. In these instances, there’s no trend to forecast from, and step one simply isn’t available to you. Context becomes the substitute: what do we know about this market, this competitor set, and this audience, even without our own numbers to point to?

Early in my career, I worked with a marketing manager at an industrial cleaning solutions company. His firm had recently started servicing petroleum refineries, and he’d been handed a large budget with a mandate to double marketing revenue. At the time, I was working at a company that sold data on industrial plants and refineries, and he was a client who genuinely valued what that data could tell him.

We found his firm was already working with 90 of the roughly 130 operating petroleum refineries in the United States. There simply weren’t enough net-new refineries left to hit the number he’d been given, no matter how well the campaign performed. Armed with that context, he went back to his stakeholders and told them the goal just wasn’t achievable.

The same type of questions can be asked of the context of any goal you’re setting:

• What does market concentration actually look like?
• How have competitors been performing?
• What else must be true for you to hit this number, and who outside marketing controls whether it comes true?
• What is the actual buying committee here? For MedTech and healthcare industries, a goal built on reaching physicians looks very different once value analysis committees and supply chain enter the process.
• Where is the product in its regulatory pathway, and does this goal assume a clearance date that hasn’t happened yet?

It’s often worth bringing in an outside perspective at this stage. Someone even one step removed from the day-to-day will usually spot the thing everyone closest to it has stopped questioning.

Lastly, context isn’t only there to tell you a goal is impossible. It can just as easily hand you an underserved segment nobody’s pursued, a competitor pulling back their spend, or a market shift that makes an aggressive number more achievable than it looks on paper. The point is, don’t use context as a way to talk goals down. Make sure the goal reflects the contextual reality, whichever direction that points.

4. Time Frame: Compare Like to Like, at the Same Resolution as the Goal

Every goal lives inside a time frame: quarterly, annual, even daily, and that time frame is where your historical data earns its keep. A number in isolation doesn’t tell you much. That same number, placed against last year’s performance for that exact period, and the periods around it, tells you a lot more.

Say leadership wants 20% growth over a set time frame. Is that aggressive or conservative? You can’t answer that using the language of the goal alone. You need data. What did this same time period look like last year? Did the periods before it trend up or down? And what made last year’s number what it was? Seasonality is the obvious one, but it’s rarely the only factor. Others include a promotion you ran last October, a competitor who went dark for six weeks, or a tariff change that pulled purchasing forward into a quarter that now looks like organic growth. Any of these can leave you comparing against a number that was never really the baseline. Know what’s inflating or deflating the comparable period before you agree to grow on top of it.

These types of considerations matter just as much on the short end. Daily or weekly goals get judged against daily or weekly history, not a monthly average smoothed over. Pressure testing the time frame of your goals means making sure you’re comparing “like to like,” and that the historical data backing your goal was pulled at the same resolution as the goal itself.

5. Communication: Make Every Check-In a Data Check

Setting a goal and then forgetting about it is one of the most common ways misalignment happens. Communication means every check-in should be a data check, not just a status update. Is actual performance tracking with what the historical trend predicted? Has anything happened in the market that changes the context you built the goal on?

Every check-in reports the North Star and every secondary metric, in the same order, every time. It is the single cheapest safeguard in the process. A North Star reported alone will look like success right up until someone asks what happened to the metric nobody was showing. Reporting them together means a guardrail getting close to its floor is a visible event in week three, not a discovery in week twelve.

There’s a coordination problem specific to privacy-first industries too. In most organizations, a goal check-in involves marketing and leadership. Here it frequently also involves legal, compliance, regulatory restrictions, etc., and any one of them can change what you’re permitted to do in service of the goal. Building that routing into your check-in cadence is what keeps a constraint from going unnoticed until it’s too late.

This is also where the data you gathered up front pays off a second time. If you did the work in steps one and two, you already have a baseline to measure against and a set of assumptions you can revisit. This helps make a check-in concrete instead of vague. You can point to what the trend looked like a few weeks ago, what it looks like now, and which specific assumption, if any, changed in between. Markets shift, budgets change, and a goal that made sense at the start of the period can be stale before it ends. Data is what tells you it’s stale, and by how much, instead of leaving you to guess.

6. Reassess: Go Back to the Data That Built the Goal

You won’t always hit the goal. That’s true in business the same way it’s true anywhere else. Whether a miss shows up as a surprise at the finish line, or as something everyone already saw coming, says a lot about how well a team is set up to handle it. This is where reassessment comes in.

Reassessing isn’t a one-time to-do item you tack on at the end of the period. It means going back to the same data that built the goal in the first place and asking if it still holds up. Has the historical trend shifted? Has new context emerged that changes what’s realistically achievable in the time frame you set? Reassessment is really just steps one through three, run again with fresh data instead of the numbers you started with. This is what makes reassessment useful, as opposed to just a postmortem.

A calendar date isn’t a great trigger for reassessment on its own, since it means you find out about a problem whenever the calendar says to look, not when the data already told you. A better approach is to reassess whenever one of the following happens:

• Performance drifts from the trend. Say your forecast, built off last year’s trend, predicted 500 leads for the month. Three weeks in, you’re sitting at 350, well off the pace the trend predicted. That gap is your signal to reassess right then, not to wait and see what the final number looks like at month’s end.
• An assumption breaks. If one of the assumptions you wrote down in step one turns out to be wrong, treat that as its own trigger, independent of whatever the numbers say that week.
• New context emerges. A competitor move, a market shift, anything that would have changed the goal if you’d known it going in is worth revisiting the goal for, even if performance still looks fine on paper.
• The regulatory picture shifts. This is a change that can invalidate a goal in MedTech and healthcare industries overnight. You don’t control them and you often don’t get much warning. Treat any material movement here as an automatic trigger to go back to the forecast, even if performance to date looks fine.
• A guardrail is breached. If a secondary metric crosses the floor you set in step one, that’s a reassessment trigger on its own, even if the North Star is pacing perfectly.

Build this habit of reassessment from the start. Then when something needs to change, you’re adjusting on purpose, and backed by evidence, instead of explaining why the number didn’t hold after the fact.

Notice the thread running through all six pillars: data isn’t one step in the process, it’s the foundation each one sits on. Your North Star defines which data the goal actually lives or dies on, and your secondary metrics define what you refuse to trade away to get there. Historical data tells you where you’ve been. Context tells you what the data means in the real world. Time frames tell you what data comparisons to make. Communication turns data into an ongoing conversation instead of a single bet. And reassessment is just the willingness to go back to the data when the world changes, instead of defending a number that was set based on assumptions that no longer hold.

Pressure testing a goal isn’t about lowering expectations. It’s about making sure that when leadership commits to a number, everyone in the room is standing on the same data. That’s what keeps a stretch goal a stretch, instead of a setup.

If your growth targets are being set against measurement you’re not fully confident in, that’s a fixable problem, and it’s one we’ve spent more than a decade solving. Let’s talk about what a pressure test could look like for your next planning cycle.