A new-location ramp curve is the month-by-month revenue path a site was expected to follow when the capital was approved: how much revenue in month one, how steeply it climbs, and the month it reaches the run rate the investment case assumed. Managing the first 90 days after opening means reading actual performance against that curve early enough to act on it, separating a demand problem from an execution problem, and deciding whether to fund, fix, or reforecast. Most platforms have the curve in a financial model and nowhere else, which is how a location ends up four months behind before anyone says so out loud.
This guide starts the day the doors open. Whether to commit to the market at all is a different question, and Service-Area Expansion: Validating Demand Before You Commit covers the demand evidence, the go/no-go memo, and the order in which to stand up local presence and tracking before any paid spend. The owned local assets themselves belong to Local Search as an Owned Asset, and review capture as a running metric belongs to Running Reviews as an Operating Metric. What follows is the part that tends to get improvised at every opening: measuring a site against the number it was underwritten on, and deciding what to do when it misses.
The investment case almost always contains a monthly revenue build for the new site. The operating team almost never receives it. What reaches the general manager is an annual target, which is the least useful form of the same information, because an annual number cannot be missed until it is too late to do anything about it.
Extract the curve before the site opens and write it down as three things. The revenue expected in month one. The slope between month one and maturity. The month at which the site is assumed to hit its run rate. Then write the assumptions that produce those numbers, because the assumptions are what you will actually manage: inquiries per month, the conversion rate applied to them, the average job size, and the delivery capacity available in each month.
This matters more as the footprint grows. Across Claymore engagements we have supported 82% footprint growth in 15 months, and a platform opening at that rate cannot afford to treat each launch as a one-off. The curve and its assumptions are the reusable part.
A location running at 60% of plan is not a finding. It is a prompt to go looking, and four very different problems produce that same number:
Each has a different fix, a different cost, and a different person responsible. None of them is visible in the revenue line, which means the instrumentation to tell them apart has to exist from the first day of trading. That is a pre-opening decision, and it is the most common thing a launch skips because it feels administrative next to getting the doors open.
Revenue in the first six weeks is noise. The job count is too small, the opening promotion distorts the mix, and seasonality has not been separated out. Four things are readable, and they are the ones to look at:
That last one carries most of the weight. In a services business, completed revenue lags the sale by the length of the delivery cycle, so a new site can look acceptable on revenue in month two while its forward book is empty, and the shortfall only appears in month four when it is a quarter too late to seed differently. The week-six question is not whether the site is at plan. It is whether the inputs that produce the plan are present.
On opening day the location has no local reputation, no review base, no referral flow, and no organic position. Every one of those is an asset that takes months of trading to build, which means early demand has to be bought, and bought demand is the most expensive kind. The true cost of bought leads is higher than the invoice suggests once shared, resold, and low-intent volume is accounted for.
So the curve should carry a declining cost per customer, not a flat one. If the model assumes the same acquisition cost in month one and month nine, it is wrong in both directions: too optimistic at the start and too pessimistic later. Across Claymore engagements, shifting the mix toward owned acquisition has produced 60% CAC reduction alongside a 3x increase in owned share of acquisition, and roughly 38% of revenue in these businesses arrives through untracked word of mouth that a brand-new site does not yet have. The gap between a new location and a mature one is largely that missing base.
The practical form of this is two budgets rather than one. A seeding budget, which is temporary and has a written step-down schedule, and a steady-state budget, which is what the site runs on once its owned assets are producing inquiries. If the seeding budget never steps down, the location has not ramped. It has bought its revenue, and the economics in the investment case were never actually tested.
Ownership during a launch is usually split without anyone deciding to split it. Marketing runs the seeding. Operations runs delivery and hiring. A general manager takes the profit and loss on a date that was never agreed. The result is that a miss belongs to everyone, which means it belongs to no one, and it gets escalated a month or two after it became visible.
Name one owner for the ramp curve when the capital is approved. Give that person a weekly review of 20 minutes covering the four inputs and the forward book, not a monthly revenue read. Give the sponsor a single view: actual against curve, the four inputs, the current diagnosis, and the call being made. S&P Global found in 2026 that 71% of GPs and 53% of LPs now prioritize operational value creation, so site-level execution gets looked at in a way it did not a few years ago. A clear one-page answer is better than a defensive one.
Decide in advance what each kind of miss causes you to do. Writing the rule down before opening is worth more than waiting longer for data, because by the time a revenue shortfall is unambiguous the cheap responses have usually expired.
Reforecast the site at named checkpoints, month three and month six, rather than continuously. A curve that moves every month stops being a test and becomes a description of whatever happened, which is useless for deciding anything. The company-level version of this question, what to lock and what to reforecast across the whole plan, sits in the annual commercial plan.
There is also an honest conclusion that launch teams avoid. If inquiry volume is wrong by a wide margin and stays wrong after the seeding has been tested properly, the market read was wrong. That finding belongs in the next go/no-go decision, not in a harder push on the current one.
A launch should end on stated criteria rather than drifting into business as usual. Four are enough:
At that point the location joins the normal operating review on the same exhibits as every other site, and the launch owner hands the number over formally. Before then it needs its own weekly rhythm, because the failure modes are different from those of a mature site.
The last step is the one that compounds. Write down what the launch package contained, what it cost, and how the actuals compared with the curve, and the next opening starts from evidence rather than from memory. A platform opening its fifteenth location should not be re-deriving its seeding budget. That kind of accumulation is how commercial work produces results like the 36% revenue increase over 24 months on flat marketing spend we have seen in portfolio engagements: not from one clever launch, but from not paying the learning cost fifteen times.
If you want the commercial side of a platform assessed before the next opening, our 8-week commercial audit covers the acquisition economics, the data needed to read a ramp, and the operating rhythm around it.
There is no universal answer, and any number quoted without reference to your own business is guesswork. The defensible figure is the one in the model the capital was approved against, and the best check on it is your own history. A platform that has opened sites before has actual ramp data, and those actuals beat any external benchmark. If this is the first opening, treat the modeled ramp as an assumption under test rather than a target, and say so to the board before the site opens rather than after it misses.
Four inputs, not revenue. Inquiry volume against the assumption in the model, speed to first contact, conversion rate on the inquiries received, and work booked forward rather than work completed. Revenue in the first six weeks is too thin to read, and in a services business it lags the sale by a full cycle, so a site can look acceptable on revenue in month two and be empty in month four. Booked-forward work is the earliest honest signal you have.
Budget it as two numbers rather than one. A seeding budget, which buys demand the location has not yet earned and is deliberately temporary, and a steady-state budget, which is what the site should run on once its owned assets are producing. Write the step-down into the plan with dates. If the seeding budget never steps down, the location has not ramped, it has bought its revenue, and the unit economics in the investment case were never tested.
Not from the revenue line, and not before the inputs have been diagnosed. Decide the checkpoints and the triggers before the site opens: what you will look at in month three, what in month six, and what each kind of miss causes you to do. A conversion miss and a demand miss can produce the same revenue shortfall and call for opposite responses, so a decision rule written in advance is worth more than a longer wait for data.
One named person, from opening day, with the ramp curve and its four inputs in front of them every week. Launch ownership is usually split by default between the marketing team running the seeding, the operations team running delivery, and a general manager who inherits the profit and loss on a date nobody agreed. That split is why misses get noticed a quarter late. Name the owner when the capital is approved, not when the site opens.
Have a revenue problem the board is asking about? Start a conversation.