Andrew Crossley

    What KPIs should a CPO track?

    SHORT ANSWER

    A CPO should track one primary metric per quarter plus a small guardrail set. Pre-PMF: activation rate, week-four retention and time-to-value. Post-PMF: net revenue retention, expansion rate and revenue per active user. Always: cycle time and rework rate as health metrics. If your product dashboard has more than seven numbers on it, nobody is accountable for any of them.

    Answered by Andrew Crossley, Fractional Chief Product Officer · Updated 2026-08-01

    Why it matters

    Metric sprawl is the most reliable symptom of an unowned product function. Ten tracked numbers means each team optimises the one that flatters it.

    The metric a company optimises explains its behaviour better than its roadmap does — that was the clearest lesson from marketplace product work at scale.

    How it works in practice

    1. 1

      Choose one primary metric per quarter

      It must be movable by the product team within the quarter and connected to revenue within two.

    2. 2

      Name one owner

      One human, not a team. Shared ownership of a metric is no ownership.

    3. 3

      Add guardrails, not more goals

      Two or three counter-metrics that must not degrade — support contact rate, latency, churn.

    4. 4

      Instrument time-to-value

      Minutes or days from signup to first real outcome. It predicts retention better than almost anything else pre-PMF.

    5. 5

      Track delivery health separately

      Cycle time and rework rate tell you whether the product function itself is working, independently of market outcomes.

    Common mistakes

    • Reporting signups and page views to the board as product performance.
    • Changing the primary metric mid-quarter because it is not moving.
    • Tracking NPS as a primary metric — it is slow, noisy and rarely actionable at small n.
    • Measuring feature usage without measuring whether the feature changed retention or revenue.

    FROM EXPERIENCE

    Support queues as a product KPI

    At Sage and Echo-U the highest-signal product metric was not in the product analytics at all — it was contact rate per hundred active accounts, broken down by reason code. Every recurring reason was a product defect wearing a support costume.

    Churn was decided in the first thirty days, so time-to-value plus thirty-day contact rate predicted revenue retention far earlier than the retention metric itself.

    Frequently asked

    How many KPIs is too many?

    More than one primary plus three guardrails plus two health metrics. Beyond that, focus dissolves.

    Should a CPO own revenue?

    Post-PMF, yes — net revenue retention is a legitimate product KPI. Pre-PMF, revenue is too lagging to steer by.

    What about AI feature metrics?

    Add task success rate and human-override rate. Model quality means nothing if users correct the output every time.

    IN SHORT

    • One primary metric per quarter, one named owner, two or three guardrails.
    • Pre-PMF: activation, week-four retention, time-to-value. Post-PMF: NRR, expansion, revenue per active user.
    • Track cycle time and rework as product-function health, separate from outcomes.

    Explore the Crossley Method

    Seven stages from first idea to first revenue, with one output each.

    Explore the Crossley Method

    THE FRAMEWORK

    The Crossley Method: idea to first revenue in seven stages

    See the full method
    1. STAGE 1DiscoverWeek 1
    2. STAGE 2ValidateWeek 2
    3. STAGE 3PrototypeWeek 3
    4. STAGE 4Build MVPWeeks 3-4
    5. STAGE 5LaunchWeek 5
    6. STAGE 6First RevenueWeek 6
    7. STAGE 7ScaleOngoing

    MORE ANSWERS

    Fractional CPO

    What does a fractional CPO do?

    A fractional CPO is a senior product executive who owns product strategy, discovery and delivery for one to three days a week instead of full-time. They set the product direction, decide what gets built and what gets cut, run the operating cadence with engineering, own the product metrics reported to the board, and coach or hire the permanent product team that eventually replaces them.

    How much does a fractional CPO cost?

    A fractional CPO costs roughly £4,000–£10,000 per month in the UK, or $6,000–$15,000 in the US, depending on days per week. One day a week sits at the bottom of that range, two-plus days at the top. There is no employer's National Insurance, pension, equity or recruiter fee, so the loaded cost is close to the headline number.

    When should a startup hire a fractional CPO?

    Hire a fractional CPO when product decisions have become the bottleneck but you cannot yet justify a full-time executive. In practice that is after you have engineers building and before you have product-market fit: usually pre-seed to Series A, three to fifteen people, with a roadmap that keeps growing and a founder who no longer has time to run discovery properly.

    Is a fractional CPO worth it?

    A fractional CPO is worth it when product decisions are costing you more than the retainer — which, with four engineers on payroll, happens fast. One quarter of misdirected engineering costs £50,000–£80,000 in salary alone. The retainer pays for itself if it prevents a single wrong quarter. It is not worth it if you lack build capacity, or will not give the role decision rights.