A scenario
What happens if we cut our price 10 percent
A price cut is the only decision on this list where the cost lands in full and the benefit is a hypothesis. The margin goes in the month you decide. The volume has to arrive, from a market that has competitors in it who can answer.
Run this on your own numbersAll scenarios
Ships as a starter scenario: one lever, price_pct at minus 10, landing in month one.
Mechanics
What the model does with it
Same lever as the rise, same order inside the month, and an asymmetry that catches people out: cost of goods does not move with price.
Both price indexes fall together
The list price and the new business price are both multiplied by 0.9 in the month the lever fires. New business gets the whole cut immediately, because new business pays today's quoted price by definition.
The cost line does not follow
Revenue is accounts times revenue per account times the price index. Cost of goods is accounts times revenue per account times one minus gross margin times the supplier index. Price appears in the first and not the second, which is correct and is the whole problem. On the invented Harborline twin at 31 percent gross margin, ten percent off price takes the margin on a unit from 31 cents in the dollar to 21, so volume has to rise about 48 percent to stand still.
Existing customers get the cut at renewal pace
The same repricing rule applies in both directions. A segment closes the gap to the new list price by one over the average contract months times its contract multiplier each month. So the revenue you give away arrives gradually while the new business discount is immediate, which makes the first quarter look better than the steady state.
New business responds harder than churn does
Forty percent of the elasticity is spent on churn, the rest on new business, at 1.25 times and with a damping term that pulls the response in as the gap gets large. The reasoning is that a buyer with no switching cost to pay is the most price sensitive person in your market. So a cut buys logos faster than it buys loyalty.
Competitors follow you down
The matching rule has no sign in it. A negative move produces a negative target: reaction times your move times 0.7 plus 0.6 of aggression, landing after the lag. When they land, the relative price advantage you bought partly closes, and the volume assumption behind the decision quietly changes.
Capacity is the thing that bites back
Volume is served by people. Utilisation is revenue over headcount times revenue a head can serve. Above 1.12 the strain counter starts, quality falls toward a target that includes it, satisfaction trails quality, and churn takes the difference. If runway allows it, the operations lead hires about five percent more heads after two strained months, on a four month cooldown.
The finance lead notices
Two or three consecutive months of negative operating profit reach the investor, after which every plan is judged against cash rather than growth. Below the runway threshold the finance lead cuts between 4 and 20 percent of headcount depending on their patience, and takes thirty percent out of marketing at the same time. That is the path where a price cut ends in a smaller company.
The ecosystem
Which agents move, and why
A cut is answered by fewer agents than a rise, and the ones who answer matter more.
Nobody regulates a price cut. Partners do not drift, because the drift rule only fires above a 10 percent price index. The channel likes you more and the model does not give you credit for it.
- Price led customers, 1.7 times the base elasticity, are the ones the cut is aimed at and the ones most likely to leave again when somebody undercuts you next year
- Competitors, objective to take share when an opening appears at weight 0.45, follow a share of the cut once, after their lag
- Salespeople, objective to hit quota at weight 0.55, book more and are measured against the same quota, so attainment rises and the discounting rule stays asleep unless you granted discount authority as a separate lever
- The operations lead, objective to deliver what was sold at weight 0.5, hires into strain when there is runway for it
- The finance lead, objective to protect margin at weight 0.45, is the agent most likely to end this scenario early
The arithmetic, on an invented company
What ten percent costs Harborline Components
Honestly
Where this is weakest
This is the scenario where the model is most likely to be optimistic, because the things that make a price cut work in real life are exactly the things it does not carry.
- No volume discount on the way in. The supplier index does not fall when you buy more, so scale economics are absent
- No fixed cost absorption. Operating cost is headcount times cost per head, so a busier factory does not get cheaper per unit
- No capacity ceiling other than the strain feedback. Nothing says you physically cannot make 48 percent more
- The response is symmetric by construction. Real markets often punish a rise harder than they reward a cut, and one elasticity cannot hold both
- Nothing models the customer who was going to renew anyway and now pays less. The gradual repricing catches part of that, not all of it
- Competitors match once per gap. A genuine price war, where each side answers the other repeatedly, is not in the engine
The band, not the line
One year, ninety times, with the price down ten
What people ask about this one
How much volume do I actually need?
Divide your current gross margin by the margin after the cut. At 31 percent falling to 21 percent, that is 1.48, so 48 percent more units. The engine computes it the same way, which is why the answer is stable even when everything else in the run is noisy.
Does this start a price war?
Not in this engine. Competitors answer a gap once and then mark it answered. They do not answer your answer.
If a spiral is the risk you are actually worried about, build a custom scenario with a competitor price move scheduled after yours and see what the second round does. The model will run it. It will not invent the third round for you.
Do my existing customers get the cut automatically?
Only at renewal pace, and only if you use the lever that moves the list price. There is a separate lever that moves the new business price on its own, which is closer to what most companies actually do and is a much cheaper experiment.
What if my sales team is already discounting?
Then your realised price is lower than your list price and the twin should be built on the realised one. Check revenue per customer against list in the ledger. If they disagree, the discount is already happening and cutting the list price cuts it twice.
Next
The questions this one leads to
Find the price rather than testing one
A sweep runs the whole Monte Carlo at every step of a range. Twenty five settings at eighty replications is two thousand full simulations and returns a response curve with a best point on it.