How do I structure a fractional CRO's first 90-day plan when the sales team has no historical data to work from in 2027?

Signals you actually need this

The clearest signal is that nobody can answer basic questions about the revenue engine with a number. Ask three people on the team what the win rate was last quarter and you get three different answers, or three shrugs. Ask what pipeline coverage looks like for next quarter and the response is a feeling. Ask how long deals take from first touch to close and someone says "it depends." When every answer is qualitative, you do not have a revenue system — you have a set of individual habits that happen to produce some revenue.

A second signal is a CRM that exists but is not trusted. This is subtly different from having no CRM at all. The tool is installed, reps log in, opportunities have stages — but the stages mean different things to different people. One rep moves a deal to "Proposal" when they send pricing; another moves it there after a verbal yes. One rep closes lost deals; another leaves them open forever so the pipeline looks healthy. The data is technically present but structurally meaningless, which is worse than an empty database because it creates false confidence. A fractional CRO walking into this needs to treat the existing records as raw material to be re-qualified, not as a baseline to be trusted.

A third signal is that the team is new. If you hired three or four reps in the last two quarters and none of them have been through a full annual cycle, you genuinely have no historical data — not because you failed to collect it, but because the observation window is shorter than the sales cycle. A company selling a 90-day enterprise cycle that started selling nine months ago has, at best, one and a half cycles of evidence, which is not enough to establish a reliable conversion rate. The fractional CRO's job in that situation is to compress the learning loop: shorten the feedback cycle artificially by instrumenting leading indicators, so you can make decisions before you have twelve months of lagging data.

A fourth signal is a pivot. If the company changed ICP, pricing model, or product packaging in the last two quarters, all prior data is contaminated. Deals closed under the old motion do not predict outcomes under the new one. Founders often resist this conclusion because the old data is the only data they have, and it feels better than nothing. It usually is not. A fractional CRO should be willing to say plainly that the historical record is not usable and that the plan has to start from a clean observation window.

The fifth signal is the one founders notice first: revenue is flat or declining and nobody can explain why. When the diagnosis is "we need more leads" but nobody has tested whether the problem is lead volume, lead quality, conversion, or retention, you are guessing. A fractional CRO's first job is to replace the guess with a measurement. That measurement does not require history — it requires instrumentation going forward, which is exactly what the first 90 days should build.

What good looks like vs. bad

The single biggest fork in a no-data 90-day plan is whether the fractional CRO treats the absence of data as a reason to move slowly or as the primary problem to solve. The good version front-loads instrumentation and accepts that the first month produces almost no revenue impact. The bad version jumps straight to tactics — rewriting the pitch deck, changing the comp plan, firing the underperformer — because tactics feel like progress and instrumentation feels like delay.

The good path has a specific shape worth spelling out. Weeks one and two are pure instrumentation: every opportunity stage gets a written exit criterion, every rep gets a required-field discipline, every deal gets a next-step date. This is unglamorous and reps will resist it, because it feels like admin. The fractional CRO has to hold the line here, because the entire plan depends on the data being clean from this point forward.

Weeks three through six are the evidence window. The team sells normally, but now every stage transition is timestamped and every deal has a recorded outcome. At the end of six weeks you have something you did not have before: a real, if small, sample of how your actual buyers move through your actual funnel. It will be noisy. It will be a sample of maybe 20 to 60 opportunities depending on deal volume. But it is real, and it is yours.

Weeks seven through nine convert that sample into a baseline model. If 40 opportunities entered the funnel and 6 closed won, your observed win rate is 15 percent — with a wide confidence interval, which you should state honestly rather than pretending it is precise. If average cycle length across those 6 wins was 47 days, that is your working assumption, flagged as provisional. If pipeline coverage at the start of the window was 2.1x and you closed at 15 percent, you now know that 2.1x coverage is not enough and you need roughly 6x to 7x to hit target — a number that would have been pure guesswork three months earlier.

Weeks ten through twelve are where you finally get to act on real numbers. Comp plans get set against observed attainment distributions. Hiring plans get set against observed rep productivity. The forecast gets built from the model rather than from rep optimism. This is the payoff, and it only exists because the first six weeks were spent building the instrument instead of swinging it.

The bad path is common because it is what an experienced operator's instincts push toward. A fractional CRO who has seen twenty revenue engines will recognize patterns quickly and want to act on them. That instinct is valuable, but in a no-data environment it is dangerous, because pattern recognition from other companies is not evidence about this one. The discipline is to hold hypotheses loosely, write them down, and test them against the evidence window rather than acting on them in week one.

Real cost and ROI ranges

A fractional CRO engagement is typically priced as a monthly retainer against a defined time commitment, commonly one to three days per week. Because rates vary enormously by operator seniority, geography, and scope, the useful way to think about cost is not the headline number but the ratio of cost to the value of the decisions being made. A fractional CRO who costs a meaningful monthly retainer but prevents a six-figure mis-hire or a mispriced comp plan that caps growth for a year has paid for themselves several times over.

The concrete ROI math in a no-data situation looks like this. Suppose the company is doing $4M in annual recurring revenue and wants to reach $6M. Without a baseline model, the typical failure mode is to hire two reps at a fully loaded cost of roughly $150K to $200K each, assume they will each carry $500K to $750K in quota, and discover nine months later that the actual ramp is slower and the actual productivity lower than assumed. That is a $300K to $400K commitment made on assumptions. A fractional CRO's first 90 days exist specifically to replace those assumptions with observed numbers before the commitment is made.

The other ROI lever is forecast accuracy. Companies without historical data typically forecast by asking reps for a number and applying a haircut. That produces a forecast that is wrong by 30 to 50 percent in either direction, which cascades into hiring decisions, cash planning, and board communication. A baseline model built from six weeks of observed stage conversion will not be precise, but it will typically cut forecast error substantially, because it is anchored in behavior rather than sentiment. For a company raising capital or managing a cash runway, that improvement alone often justifies the engagement.

There is also a cost to not doing this. The most expensive outcome in a no-data environment is a confident wrong decision made early. Rewriting the comp plan in week one based on a hunch, then discovering in month four that it incentivized the wrong behavior, costs you a quarter of selling time plus the disruption of changing comp mid-year — which damages rep trust in a way that is hard to repair. The fractional CRO's willingness to spend six weeks measuring before acting is not slowness; it is the cheapest insurance available against that outcome.

On the contract side, structure the engagement in phases with a clean off-ramp. A short diagnostic phase, then a written 90-day plan with a scorecard, then execution with monthly review gating renewal. Insist on a defined day count in writing — "up to two days per week focused on CRM instrumentation, stage definition, and pipeline modeling" — so scope creep is visible in both directions. Avoid open-ended retainers with no graduation clause. The best outcome of the engagement is that the fractional CRO builds the system, documents it, and hands it to a full-time leader. A contract that plans for that graduation is healthier than one that maximizes the retainer's lifespan.

How it plugs into your workflow

The 90-day plan has to attach to the operating rhythm the company already has, or it will be ignored. The fractional CRO should not create a parallel process. They should modify the existing weekly pipeline review, the existing monthly business review, and the existing quarterly planning cycle so that each one starts consuming the new data.

The weekly pipeline review is the workhorse. Before the engagement, this meeting is usually a rep-by-rep recital of deal status. After instrumentation, it becomes a structured review against stage exit criteria: which deals moved, which stalled, which were disqualified, and why. The fractional CRO's role in this meeting is to enforce the criteria and to capture the reasons deals stall, because those reasons are the raw material for the baseline model. Over six weeks, patterns emerge — deals stall at the same stage, for the same reasons, in the same segment — and those patterns are the first real insight the company has ever had about its own motion.

The monthly business review is where the model gets refreshed and where the fractional CRO reports against the scorecard. The scorecard should be short: three to five metrics that define success for the engagement. In a no-data situation, the right early metrics are leading indicators — pipeline created, stage conversion rates, next-step discipline compliance — because revenue lags the work by a quarter or two. Tracking closed revenue in month one of a no-data engagement tells you nothing about whether the plan is working. Tracking whether every open deal has a dated next step tells you a great deal.

The quarterly cycle is where the model becomes decisions. Once you have a baseline conversion rate, you can set targets that are grounded rather than aspirational. You can size the sales team against observed rep productivity instead of a benchmark from a company that is not yours. You can build a coverage model that says "to close $X next quarter at an observed 15 percent win rate and a 47-day cycle, we need $Y in qualified pipeline entering the quarter." That is a fundamentally different conversation than "let's hire three reps and see."

The workflow also needs a documentation layer, because the entire point is capability transfer. Every stage definition, every scorecard metric, every model assumption should live in a written document that a future full-time leader can inherit. A fractional CRO who leaves behind a working system and a written playbook has done the job. One who leaves behind a set of habits that only they understand has not.

Related questions

How long before we have usable data?

With a functioning CRM and a defined evidence window, roughly six weeks of active selling produces a usable — if noisy — sample. If deal volume is very low, expect eight to ten weeks. State confidence intervals honestly rather than pretending the sample is precise.

What if our CRM is a spreadsheet?

That is common and workable. Move to a real CRM first, because stage timestamps and required fields are the mechanism. A spreadsheet cannot enforce stage exit criteria, and without enforcement the data stays unusable.

Should we pause selling during the evidence window?

No. The evidence window requires normal selling activity. Pausing defeats the purpose. The change is not less selling; it is more disciplined logging of what happens during selling.

Can we set quotas before the 90 days end?

You can set provisional quotas, clearly labeled as provisional and subject to revision at the 90-day mark. Setting firm quotas on no data and then changing them mid-year damages rep trust more than waiting.

What if the fractional CRO wants to change everything in week one?

Push back. Ask what evidence supports each change. A fractional CRO who cannot separate hypothesis from finding is likely to impose another company's playbook on yours, which is exactly the failure mode this plan is designed to avoid.

FAQ

Why not just use industry benchmarks instead of waiting for our own data? Benchmarks describe the median company, not yours. Win rates, cycle lengths, and ramp times vary enormously by segment, price point, and motion. Benchmarks are useful as sanity checks on your observed numbers — if you measure a 60 percent win rate, a benchmark tells you to check your stage definitions. They are not useful as substitutes for measurement, because acting on a benchmark that does not match your business produces confident wrong decisions.

What is the single most important thing to fix in the first two weeks? Stage exit criteria. Every opportunity stage needs a written, unambiguous definition of what must be true for a deal to enter it. Without this, every downstream number is meaningless, because "win rate" depends entirely on what counted as a qualified opportunity. Get this right before touching anything else.

How do we keep reps from gaming the new instrumentation? Make the data useful to them. If the pipeline review becomes a genuine coaching conversation grounded in real stage data, reps engage. If it becomes a compliance audit, they will enter minimum viable data and the model will be built on noise. The fractional CRO should spend as much time making the data useful to reps as making it accurate for leadership.

What if we discover the historical data was actually fine? Then you have saved six weeks and can move faster. The instrumentation work is not wasted — cleaner stage definitions and required fields improve any dataset. Run the evidence window anyway, but shorten it, and use the historical record to validate the model rather than to build it.

How do we know when the fractional CRO should graduate out? When the system runs without them. Concretely: the weekly review happens and produces useful output whether or not they attend, the monthly scorecard is populated and reviewed, and a full-time leader can step into the role with a written playbook. If the process collapses when they miss a week, the capability has not transferred.

Does this plan change if we are pre-revenue? Substantially. Pre-revenue, founder-led selling is the right motion and a fractional CRO is usually premature. The exception is when the founder is selling but cannot systematize what they are learning. In that case the fractional CRO's job is narrower: extract the founder's implicit motion into explicit stage definitions and a documented playbook, which becomes the baseline for the first sales hires.

Sources

• Harvard Business Review — The First 90 Days and Executive Onboarding
• First Round Review — Building a Repeatable Sales Motion
• SaaStr — Go-to-Market Leadership and Sales Operations
• Pavilion — Go-to-Market Community for Revenue Leaders
• RevGenius — RevOps and Sales Community
• Bessemer Venture Partners — State of Go-to-Market
• Wizards of Ops — RevOps Community

Related on PULSE

• The CRO Role Scope Breakdown
• Building a Repeatable Sales Motion
• Vetting Fractional Executives
• Sourcing Senior Revenue Leaders
• Designing a RevOps Scorecard