The kill-switch framework

Built directly on the KPI dashboard in the previous lesson. A kill switch only works if it's set before you need it — deciding the threshold in the moment a bad quarter is already underway defeats the entire purpose of having one.

---

The specific thresholds below are this course's own synthesis — a judgment call built on the sourced benchmarks from earlier modules, not an independently sourced claim in their own right, and therefore not confidence-tagged the way a factual claim is. Treat them as a reasoned starting framework to adapt against your own numbers, not a rule handed down from an external study.

#Why this model needs its own version of this framework

Every business model this platform covers needs an honest quit doctrine, but this one needs it more than most, for a specific reason traced directly from this module's capital lesson: the capital and time already spent by the point a real enterprise motion is even testable — a working product, initial compliance readiness, a first real sales cycle run to completion — is large enough that sunk-cost reasoning is a genuine, structural risk here in a way it isn't for a business that can be meaningfully tested in a month for a few thousand dollars.

#The thresholds

On the sales motion itself:

• Zero deals reach the finalist/shortlist stage of a real RFP process within two full sales-cycle lengths (per Module 1's benchmarks, roughly 8–24 months depending on target contract size) — a signal the product-market fit or the go-to-market motion itself is wrong, not that the team needs to try harder at the same approach.
• Win rate on deals that do reach the finalist stage stays near zero across multiple genuine attempts — distinct from the first threshold, because reaching finalist stage and then losing repeatedly is a different, more specific problem (Module 2's committee dynamics, or a real product gap) than never reaching that stage at all.

On the compliance and delivery layer:

• SOC 2 readiness — or the specific compliance bar your target buyers require — is not achievable within your actual runway — per this module's capital lesson, this is frequently the real gating cost, and discovering it can't be funded in time is a legitimate reason to stop or pivot down-market rather than a temporary setback to push through.
• A signed account's implementation repeatedly and materially exceeds its committed timeline or scope — per the previous lesson's failure mode, this is a leading indicator of the exact renewal-risk problem that undermines the entire retention thesis this business model depends on.

On the unit economics:

• Gross revenue retention runs meaningfully below what the switching-cost mechanism (Module 1) predicts, across more than one renewal cycle — if the theoretical stickiness isn't showing up in the actual numbers, the mechanism this course's whole economic case rests on isn't operating for your specific product, and that's worth taking seriously rather than assuming it just needs more time.
• Customer acquisition cost, fully loaded against the real sales-cycle length (not a hoped-for shorter one), doesn't clear a credible payback period against realistic retention — the specific payback horizon that's healthy varies by how well-capitalized the business is, but the calculation itself, done honestly with real, not assumed, cycle-length and retention numbers, is non-negotiable before committing further capital.

#What the kill switch is not

None of the above is a reason to quit after one lost deal, one slow quarter, or one difficult implementation — every threshold above is specified across multiple cycles or multiple attempts precisely so a single bad data point can't trigger it. The purpose of setting these thresholds in advance is the opposite of pessimism: it's what makes it possible to commit fully to the model, for the genuinely long runway this module's capital lesson describes, without either quitting prematurely on ordinary early-stage friction or persisting indefinitely past the point where the honest evidence says the mechanism isn't working for this specific attempt.