How do you define exit criteria for ending a fractional CRO engagement?

The Series B compliance company that could not let go

Picture a Series B SaaS company selling enterprise compliance software — the kind of platform that automates evidence collection for SOC 2, ISO 27001, GDPR, and CCPA obligations. The founder closed the first twenty customers personally. Every one of those deals was a bespoke negotiation with a chief compliance officer who trusted the founder specifically, not the company generically. The board funded a Series B on the strength of that traction and immediately asked the obvious question: what happens when the founder stops being available for every deal?

That is the moment a fractional CRO gets hired. The engagement starts with enthusiasm and a vague mandate — "build us a real sales org" — and roughly nine months later it enters the phase nobody planned for, which is ending it. This is where most fractional engagements go sideways. Not at the start, where the diagnostic work is obvious and the wins are fast, but at the exit, where the question "are we done?" has no agreed answer and both sides are guessing.

The failure mode is predictable. The fractional CRO believes the work is complete because the playbook exists, the manager is hired, and the last two quarters looked fine. The founder believes the work is incomplete because they still get pulled into the hairy deals and it still feels like the company would wobble without the outside operator. Both are reasoning from vibes. Neither has a written standard. So the engagement either drifts for another two quarters at full retainer, or it ends abruptly, the manager gets overwhelmed, the founder steps back into late-stage deals, and six months of process work quietly unwinds.

The fix is unglamorous: define the exit before you define the work. In the same document that specifies scope and compensation, write the conditions under which the engagement is finished. Make them measurable enough that a board member reading the CRM can verify them without asking either party for an interpretation. If the criteria require a conversation to adjudicate, they are not criteria — they are opinions with a number attached.

There is a second reason to write exit criteria early that has nothing to do with the ending. Exit criteria are a scope contract in disguise. A fractional CRO who has agreed that the engagement ends when a first-line manager is ramped has implicitly agreed that hiring and ramping that manager is in scope. A founder who has agreed that founder sales involvement must drop below a threshold has implicitly agreed to change their own behavior. The negotiation over exit criteria surfaces the disagreements that would otherwise show up in month seven as resentment. It is the cheapest alignment exercise available, and it happens before anyone has spent money.

The same logic applies well beyond the CRO seat. Fractional CFOs, fractional CTOs, interim VPs of marketing, and RevOps consultants all face the identical structural problem — an outside operator hired to build capability that must eventually live inside the company. The criteria differ by function, but the shape does not. Capability transferred, not tasks completed.

How the gate mechanism actually works

Exit criteria work as a gate system, not a scorecard. A scorecard averages — you can be strong on process documentation and weak on founder detachment and still show a decent composite. A gate system does not average. Every gate must pass independently, because the gates are not measuring the same thing from different angles. They are measuring different, non-substitutable dependencies that each individually keep the company reliant on the fractional operator.

Here is the practical structure that survives contact with a real engagement.

Gate 1 — Process exists and is followed, not just written. The distinction matters enormously. Every fractional CRO writes a playbook. Almost none get it adopted. The measurable version is CRM instrumentation: required fields for the stage transitions that actually predict outcomes. In enterprise compliance, that means a "security questionnaire sent" date and a "legal review start" date on every opportunity, populated on at least 80% of open deals. If reps are not filling those fields, the process is a document, not a practice, and the moment the fractional CRO stops asking about it in the weekly pipeline review, it stops entirely.

Gate 2 — A first-line manager is hired and independently ramped. Ramped means something specific: managing a qualified pipeline of a defined size — call it $1M in a business with $75K-plus ACVs — without the fractional CRO in the deal reviews. Hired is not ramped. A manager who has been in seat 90 days and still forwards every pricing exception upward is a coordinator, not a manager.

Gate 3 — Forecast attainment within 10% for two consecutive quarters. One quarter is luck. Two quarters is a signal that the qualification criteria and the stage definitions actually correspond to buyer behavior. Measure the 60-day-out forecast against actual closed-won, because a forecast produced in the final two weeks of a quarter is just a status report.

Gate 4 — Founder sales involvement below 10% of total sales hours. This is the gate that decides whether the engagement genuinely worked, and it is the one people most want to skip because it requires the founder to log their own time. A simple weekly self-report is enough — calls, emails, meetings, anything sales-touching. Precision is not the point; direction is. A founder going from 60% to 12% is a transformed company. A founder stuck at 35% is a company that has hired a very expensive deal desk.

Gate 5 — Deal cycle compression. In enterprise compliance, a 120-150 day cycle is normal at the start. Getting the controllable portion under 90 days — excluding buyer-side security review pauses, which you do not control — proves the qualification gate is killing bad deals early instead of letting them die slowly in legal.

Gate 6 — Pipeline coverage holds without the operator's network. Coverage of roughly 3.5x quarterly target, sustained two quarters, matters less as a number than as a source test. If a meaningful share of pipeline traces back to the fractional CRO's personal relationships, coverage collapses the week they leave.

The branching matters as much as the gates. A single gate missing by a small margin is a fixable problem with a short deadline, not a reason to re-buy another quarter of full-rate work. A gate missing badly is a diagnosis: something structural is wrong, and extending without a root-cause analysis just buys the same result again at the same price.

One nuance that gets lost: the gates should be measured continuously from month three onward, not evaluated once at the end. A dashboard reviewed monthly turns the exit from a negotiation into an observation. By the time month nine arrives, everyone already knows which gates are green, because they have watched them trend for six months. Surprise at the exit is a symptom of bad instrumentation, not bad performance.

Real numbers, ranges, and what they actually indicate

Numbers in a fractional engagement are only useful if you know what they are proxying for. Here are the ones worth instrumenting and what each one is really telling you.

Engagement duration. Most fractional CRO engagements at Series B run somewhere in the six-to-twelve-month band before either converting to full-time, ending, or dropping to advisory. Under six months is usually too short to hire and ramp a manager, which means the capability did not transfer — you rented execution. Beyond twelve to eighteen months without a clear conversion path, you are paying fractional rates for what has become a permanent part-time seat, which is a legitimate arrangement but should be named as such rather than pretended to be a transition.

Retainer and structure. Series B fractional CRO retainers commonly land in the $15K-$25K per month range for a two-to-three-day-per-week commitment, with wide variation by market, equity component, and scope. The structural choice that matters more than the number: how much of it is tied to gate achievement. A meaningful performance component — say a bonus per gate met — aligns the operator toward finishing rather than extending. A pure retainer with no completion incentive quietly rewards duration.

Ramp time for new AEs. In a regulated-buyer category, six to nine months to first closed deal is realistic, because the rep must learn a regulatory landscape on top of a product. This directly constrains exit timing: if you hire AEs in month four of a nine-month engagement, none of them will have closed independently by the exit, and Gate 3 will be measuring a team that has not been tested.

Deal size and its second-order effects. A single-region compliance deployment might run $75K-$150K ACV; multi-jurisdiction coverage across several frameworks can push it several times higher. The exit-criteria implication is that a company with a wide ACV spread needs segmented gates. A manager who can run the $75K motion independently may still be unable to run the $400K multi-framework deal, and a blended metric hides that gap completely.

Where deals actually die. In compliance and security-adjacent categories, a large fraction of late-stage losses trace to security review and legal terms rather than to product fit or discovery quality. Liability caps and termination-for-convenience clauses are the recurring flashpoints — procurement wants a longer termination window and a higher aggregate cap than standard terms offer. This is why a pre-qualification gate matters: sending a security requirements checklist before the demo kills unqualified deals in week one instead of month four, and it is one of the highest-leverage process changes a fractional CRO can install.

Forecast reliability threshold. Below roughly ten active enterprise opportunities, forecast variance is dominated by single-deal outcomes. One slipped deal swings the quarter enough that attainment percentage tells you nothing about process quality. Do not measure Gate 3 until the pipeline is deep enough for the number to mean anything, and say so explicitly in the engagement letter rather than discovering it at measurement time.

Discount discipline. A discount approval matrix — written justification above 15%, founder sign-off above a higher threshold, quarterly pattern review — is a small mechanism with an outsized effect on exit readiness. Founders discount to win logos. That is rational early and corrosive later, because it sets the baseline the sales team inherits. If discounting is still ad hoc at the exit, the pricing authority has not actually transferred.

Founder involvement, tracked honestly. The 10% threshold is a defensible line, but the trend matters more than the threshold. Track it weekly from month one so you have a curve, not a single reading. A founder whose involvement declined steadily and plateaued at 12% is in a different situation than one who sat at 45% for eight months and then dropped to 9% in the final month because everyone knew it was being measured.

Trade-offs, alternatives, and adjacent structures

The six-gate model is not the only way to end an engagement, and it is not always the right one. Understanding the alternatives clarifies when to use it.

Time-boxed vs. milestone-boxed. A time-boxed engagement — "nine months, then we reassess" — is simple, predictable for budgeting, and easy to sell to a board. Its weakness is that it decouples payment from capability transfer entirely. Milestone-boxed engagements tie the ending to outcomes but invite scope disputes and can drag if the milestones depend on factors outside the operator's control, like a delayed funding round or a market shift. The practical answer for most Series B companies is a hybrid: a time-boxed outer bound with milestone gates inside it, so the engagement cannot run forever but also cannot end with the work half-transferred.

Convert to full-time vs. clean exit vs. advisory tail. Conversion is tempting when things are going well, but it tests a different skill set. Building a playbook and coaching one manager is not the same job as running a 25-person org through a hiring plan and a comp redesign. Ask directly whether the operator wants the full-time seat; many fractional practitioners deliberately do not. A 60-day full-time trial with a board review at the end is a reasonable de-risking structure. A clean exit is the right call when the capability transferred and the company's next phase needs a different profile. An advisory tail — a reduced retainer covering a weekly call and escalation support for six months — is the middle path, and it exists mostly to prevent the expensive failure mode where the founder panics after one bad quarter and re-hires at full rate.

Replacement vs. extension when gates fail. If the same gate fails twice with a specific coaching plan in between, extension is usually the wrong move. Either the operator is mismatched to the stage, or the company is not structurally ready for a process-driven motion — and a third quarter at full retainer will not resolve either.

Sequencing against adjacent functions. A fractional CRO exit rarely happens in isolation. The RevOps function, the marketing engine, and the CS motion all interact with whatever the CRO built. If RevOps is still owned by a part-time contractor when the CRO leaves, the CRM instrumentation that Gate 1 depends on has no permanent owner and will decay. Sequencing the RevOps hire before the CRO exit is often more important than any single gate.

Financial mechanics of the ending. Write the money into the same document as the gates. A transition fee equal to roughly one month's retainer, paid on founder sign-off that gates are met, gives the operator a reason to finish cleanly rather than let the engagement fade. An advisory retainer at a fraction of the original rate should be negotiated at the start, not after the exit, when leverage has shifted and both parties are negotiating under pressure. Board approval of the exit criteria before the engagement begins removes the single most common late-stage dispute: whether the gates were actually met.

Renegotiation triggers. Exit criteria written in month zero can become wrong by month six. A bridge round, a new board member pushing for faster growth, a pivot in ICP, or a major competitive shift all change what "done" means. Build in a clause: material changes trigger a criteria review within 30 days. The alternative is measuring the engagement against a definition of success nobody believes in anymore.

Pitfalls that quietly break the exit

The founder who cannot stop closing. This is the dominant failure mode in founder-led companies, and it is not a discipline problem — it is an identity problem. The founder built the company by being the person the buyer trusted. Asking them to stop is asking them to give up the thing that made them successful. The mechanism that works is a written founder exit playbook, authored around month three, that specifies by deal type exactly when the founder is brought in. A "no founder zone" below a certain ACV threshold, where the founder joins only on written request from the manager, converts a judgment call into a rule. Two consecutive months of following the playbook is a reasonable proof standard.

Vanity gates. "Playbook delivered" and "team trained" are not gates. They are deliverables, and deliverables can be complete while capability remains untransferred. Every gate should be answerable by looking at a system — the CRM, the forecast history, the time log — rather than by asking someone whether it feels done.

Measuring the manager on the operator's pipeline. If the fractional CRO sourced the pipeline the new manager is running, the manager has been tested on inherited demand. A useful stress test before exit: pick a live deal, explicitly bar both the founder and the fractional CRO from any contact, and see whether it closes within the normal cycle. That single test is worth more than a quarter of dashboards.

Hitting the gates early and leaving immediately. Early success is often a masking effect — the founder is still the real closer, now operating through the manager. Honor the full transition period even when gates pass in month five. Use the extra weeks for stress tests rather than banking the time.

No named owner for the instrumentation. Gate 1 depends on CRM fields staying populated. If nobody owns that after the exit, adherence decays within a quarter and the process reverts to tribal knowledge. Name the owner — typically RevOps — before the operator leaves.

Ignoring the team's read on the transition. Reps notice when leadership is ambiguous. An exit that is announced late, or announced as a demotion of the manager's authority, produces attrition at exactly the wrong moment. Communicate the transition plan to the team when the transition period starts, not when it ends.

Skipping the root-cause analysis on a failed gate. Extending a quarter without diagnosing why a gate failed reproduces the failure. If forecast attainment missed, was it qualification, stage definitions, or a single anomalous deal? The answer changes what the extra quarter should focus on.

Related questions

How long should a fractional CRO engagement run before you evaluate the exit gates?

Start measuring from month three, evaluate formally around month nine. Earlier evaluation is premature because AE ramp in complex categories runs six to nine months; later evaluation risks the engagement drifting into a permanent part-time seat without anyone deciding that deliberately.

Should exit criteria go in the engagement letter or a separate document?

The engagement letter. Criteria in a side document get treated as aspirational. In the contract, alongside scope and compensation, they are binding and reviewable — and the board can approve them before any money is spent.

What if the founder refuses to accept the criteria mid-engagement?

Treat it as a signal the founder is not ready to delegate revenue leadership. Offer a reset focused entirely on coaching the founder out of deals. If the resistance persists, ending the engagement is cheaper than continuing one that cannot succeed.

Do the same exit criteria apply to a fractional CMO or fractional CFO?

The shape does; the gates do not. Every fractional executive engagement should end on transferred capability rather than completed tasks, but the measurable proxies differ — pipeline generation independence for marketing, close-process reliability for finance.

How do you keep the process from decaying after the operator leaves?

Assign a permanent internal owner for the instrumentation, usually RevOps, before the exit date. Schedule a 90-day post-exit review against the same gate metrics so decay is visible early rather than discovered two quarters later.

FAQ

How do I know whether to extend the engagement or replace the fractional CRO?

Look at whether the same gate failed twice. If the manager has not reached independent pipeline ownership after nine months, or founder involvement has not meaningfully declined across two quarters, extend once with a specific written coaching plan and a named root cause. If that same gate fails again, the issue is fit — either the operator is mismatched to the stage or the company is not ready for a process-driven motion. A third quarter at full retainer rarely changes either.

What happens if we hit the exit criteria early?

Honor the transition period anyway. Early gate achievement often masks a hidden dependency — the founder may still be the real closer, simply routing through the manager. Use the extra weeks to stress-test: run a deal where the founder and the fractional operator are both explicitly barred from contact, and see whether it closes on a normal cycle. If it does, the capability genuinely transferred. If it stalls, you found the dependency before you paid for the discovery in lost revenue.

Should we convert the fractional CRO into a full-time hire when the gates pass?

Only if they have demonstrated hiring and team development, not just process design. Those are different jobs, and many fractional practitioners deliberately prefer the variety of fractional work. Ask directly. If yes, structure a 60-day full-time trial with a board review at the end. If no, start the full-time search immediately and retain them as a paid advisor through it — they know the process better than any candidate will on day one.

How do you measure founder involvement without it becoming a policing exercise?

Keep it lightweight and self-reported. A weekly log of sales-touching hours, filled in by the founder, is sufficient. Precision does not matter; the trend does. Frame it as instrumentation for the transition rather than as a performance metric, and review it in the same meeting where you review pipeline coverage so it reads as one more operating number rather than a judgment.

What is the right length for the advisory tail after the exit?

Commonly around six months at a substantially reduced rate, covering a short weekly call and email escalation support. The purpose is not ongoing work — it is preventing the expensive rebound where a single bad quarter triggers a panic re-hire at full rate. Negotiate it at the start of the engagement, when leverage is balanced, not at the end.

Do exit criteria need board approval?

If the board is funding the engagement, yes. Board approval before the engagement begins eliminates the most common late-stage dispute — whether the gates were actually met — and forces the criteria to be specific enough that a director who was not in the room can verify them from the CRM and the forecast history.

Sources

• https://hbr.org/2017/03/what-sales-leaders-can-learn-from-the-best-sales-reps
• https://www.saastr.com/how-to-hire-a-great-vp-of-sales/
• https://openviewpartners.com/blog/
• https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights
• https://www.gartner.com/en/sales
• https://www.bain.com/insights/topics/sales-and-marketing/
• https://a16z.com/enterprise-sales/
• https://www.salesforce.com/resources/research-reports/state-of-sales/
• https://www.bridgegroupinc.com/blog
• https://firstround.com/review/

Related on PULSE

• [How do you structure compensation for a fractional CRO?](/knowledge.html)
• [When should a founder stop leading sales directly?](/knowledge.html)
• [How do you build a sales playbook that reps actually follow?](/knowledge.html)
• [What pipeline coverage ratio should an enterprise SaaS company target?](/knowledge.html)
• [How do you hire and ramp a first-line sales manager?](/knowledge.html)
• [What does a RevOps function own after a fractional engagement ends?](/knowledge.html)