The assumptions behindthe Reality Check
Any model that fills in a number you did not give it is making a claim on your behalf. These are all of ours, in one place, with what each is for and what changes if it is wrong for you.
These are rules of thumb, not research. They come from patterns across early-stage B2B companies, and they are set conservatively on purpose: where a default could flatter you or caution you, it cautions you.
None of them override you. Every number you provide is used as you gave it, and the result labels which figures were yours and which were assumed. If a default is wrong for your business, the tool will tell you it was used, and this page tells you what it cost you.
If one of these does not match what you see in your own numbers, that is worth a conversation. It usually means either the benchmark is wrong or something in your business is unusual, and both are interesting.
These only apply where you told us you did not have the number. Anything you did give us is used as you gave it, and the result says which is which.
Deliberately below what a healthy B2B business should hold, because a company that does not yet measure retention usually has more churn than it thinks, and an optimistic default here quietly inflates every number downstream.
If this is wrong for you: If your real retention is higher, the forecast understates you, and by roughly the difference applied to your whole book.
Low on purpose. At this stage expansion is usually something that happened rather than something anyone runs, and revenue that arrived without a motion behind it is a poor basis for planning next year.
If this is wrong for you: If you have a real expansion motion, this is one of the cheapest places to beat the forecast.
A middling early-stage B2B rate. It is used only when neither a win rate nor six months of opportunity and win counts were available, and the result flags it as assumed every time.
If this is wrong for you: This is the single most sensitive number in the model. If you do not know it, measuring it is worth more than anything else in the report.
A company that has never taken a rep from hire to quota should not be modelled as though it will do it perfectly the first time. The longer default is the cost of that inexperience.
If this is wrong for you: Shorter ramps move revenue earlier inside the period rather than creating more of it.
The tool splits ARR three ways: what exists, what will still be here, and what actually proves your motion repeats. These are the rules that do the splitting.
An account several times the size of a normal deal cannot be pointed at as proof that the next customer looks like it. It stays in your ARR in full and it may well renew. It simply carries almost no weight as evidence about how the next sale gets made.
If this is wrong for you: If that account genuinely is your future ICP, won through the motion you plan to scale, the follow-up questions give most of its evidence value back.
Above that line the book starts describing a handful of relationships rather than a market, so the excess is counted at half weight as evidence. The revenue is still fully counted as revenue.
If this is wrong for you: Nothing, if those three are genuinely representative. The follow-up questions are where you say so.
When your target needs a number to move, the tool checks how far it has to move before calling the plan credible. These bands are where that judgement comes from.
Win rate moves in points, not percentages, and it moves slowly. A plan that needs it to jump more than twelve points is usually describing a different company, not a better quarter.
If this is wrong for you: A genuinely different motion, such as moving from outbound to inbound or narrowing the ICP hard, can beat these bands. The tool will still flag it, and it should.
Raising price is the cheapest growth there is when it works, and the most over-assumed line in early forecasts. Anything above the largest deal size you have closed more than once is treated as aggressive at best.
If this is wrong for you: One large deal never validates a new price. Two starts to.
Judged after subtracting whatever your planned investment actually pays for. Asking for more pipeline with nothing behind it is a very different claim from asking for it while doubling a channel that already works.
If this is wrong for you: A channel genuinely finding product-market fit can outrun these bands. That is rare enough to be worth flagging when a plan depends on it.
The most common error in a growth plan is assuming pipeline scales with spend. It does not, and how badly it does not depends on where the money goes.
Pipeline scales with spend raised to a power below one. Doubling a proven channel gets you roughly 74% more pipeline, not 100%. Doubling into experiments gets you about 23%, and mostly it buys learning rather than pipeline.
If this is wrong for you: Early in a channel's life the curve can be steeper. Later it flattens further. These are mid-life numbers.
New budget can only scale the part of your pipeline that marketing actually creates. If most of your pipeline comes from the founder's network, more budget has less to multiply.
If this is wrong for you: If you know your real sourced split, it is worth telling us, because it changes what spending more can do.
Money spent in month one does not close revenue in month two. Inside a twelve-month horizon, lag plus cycle is often why a large budget increase moves the number far less than expected.
If this is wrong for you: Fast channels can be under a month. The follow-up questions let you say so.
Capacity is the other half of every month in the model. Whichever of pipeline or capacity is smaller is what you actually close.
Derived from sales cycle length, because a longer cycle means fewer concurrent deals a seller can genuinely carry. If you know your own number, it overrides this entirely.
If this is wrong for you: Teams with strong sales support or a simple product beat these. Complex enterprise motions fall below them.
Used to delay revenue rather than to reduce it. A lever pulled in month one cannot close anything until a full cycle has passed, which is why short horizons punish late changes.
If this is wrong for you: Nothing structural. It mostly shifts when revenue lands inside the period.
Tell me. That is a more useful conversation than most.
If a default here does not match what you see in your own business, one of two things is true: the benchmark needs changing, or something about how you sell is genuinely unusual. Either is worth half an hour.
Book a fit callModel version 1.0.0. These values are read directly from the tool's configuration, so this page cannot drift from the numbers actually in use. Run the Reality Check.