How to measure a fractional CMO
Fractional CMO KPIs
Built the growth engine that took cnvrg.io to its acquisition by Intel.
The metrics that matter, the vanity metrics to reject, and how to verify the numbers yourself.
The most common way a fractional CMO engagement fails is not that the CMO underperforms. It is that the company has no measurement architecture to know whether they are performing or not.
This page gives you the KPI framework I use with every client, and - more importantly - how to verify those KPIs yourself without taking anyone's word for it. Measurement should be independently verifiable. If the only person who can produce the numbers is the CMO you are evaluating, the measurement is not working.
Before reading the KPI framework: if you have not hired yet, see how to hire a fractional CMO first. If you are trying to decide whether to hire, see when to hire a fractional CMO.
Attention and demand metrics: what they can tell you
A useful KPI review names the decision, the relevant evidence and what work the operator will deliver.
Yaniv Goldenberg, fractional growth operator
Read each measure for the decision it supports. Attention, demand, acquisition, conversion and retained value can matter at different points in the customer journey.
Impressions and reach
Impressions and reach describe exposure. Review audience quality, placement, frequency and the intended brand or demand objective, then compare the relevant downstream evidence over an appropriate window. Exposure alone does not establish revenue or causal impact; lack of an attributed sale does not by itself establish zero effect.
MQLs without pipeline connection
Define an MQL with sales, then follow eligible leads into accepted opportunities and customers over the relevant buying cycle. Report volume, quality, handling and conversion together. A recent lead that has not matured is not a failed outcome, and a label alone does not establish commercial value.
Social engagement rate
Likes, shares and impressions measure attention. They can be useful for an agreed brand objective, but they do not by themselves establish trial signups or revenue. Report the attention metric beside the matching downstream outcome and observation window, rather than treating engagement as sales or declaring it worthless.
Branded search volume
Branded search can signal changing awareness or demand. Compare it with customer research, program timing, category conditions and downstream outcomes before assigning an explanation. Report the observation without treating it as proof that one program caused revenue.
The revenue-accountability standard
5 KPIs that actually measure a fractional CMO
CAC by channel, reconciled to eligible new customers
Customer Acquisition Cost is the agreed acquisition cost divided by new customers in the matching cohort or period. Reconcile campaign conversion definitions and attributed value with eligible CRM or billing records before interpreting channel performance. Report explained differences, unmatched records, margin and payback.
Attributed ROAS, contribution and incremental effect
ROAS compares attributed conversion value with advertising spend under the report’s definitions. A contribution calculation subtracts the agreed product and delivery costs from relevant revenue; define the numerator before dividing by spend. Incremental effect asks what changed because of the advertising and needs an appropriate comparison or experiment. An attribution model alone does not remove an organic baseline or establish that causal effect.
Attribution coverage percentage
Attribution coverage records the share of eligible confirmed revenue that can be matched to an acquisition source under a documented rule. Report the denominator, deterministic matches, modeled credit and unknowns separately. A high match percentage alone does not establish causality or make a budget decision defensible. Agree the relevant improvement work and observation window.
Funnel conversion rate by stage (with cohort isolation)
Conversion rate from acquisition to activation to revenue, measured by cohort, not by aggregated monthly totals. Aggregated totals hide mix shifts: if you acquired 3x more users in a low-converting channel last month, the aggregate conversion rate drops even if nothing in your funnel changed. Cohort-isolated conversion rates show whether the funnel itself is improving or degrading, separate from the channel mix effect. See the fractional CMO overview for how funnel ownership connects to attribution.
Pipeline contribution (for B2B) or LTV:CAC ratio (for consumer/SaaS)
For B2B, define sourced opportunities, stage, eligible accounts and closed-won outcomes separately; pipeline value is not realized revenue. For consumer and SaaS, compare acquisition cost with observed or explicitly modeled cohort contribution over a stated horizon. A 3:1 ratio can be a planning assumption, but no ratio or twelve-month point establishes sustainability for every margin, retention curve and cash position.
How to verify your fractional CMO's numbers without asking them
Measurement should be independently verifiable. Here is how to check each KPI without relying on the CMO's own reporting.
Verify CAC: cross-reference your payment processor
Reconcile eligible new customers, purchases and recorded events as different objects. For illustration, 100 eligible customers with 60 deterministic source matches leaves 40 unmatched under that rule; it does not prove 40 lost events. Check consent, identity, timing, duplicates and refunds before explaining the gap. My Glammie case reports capture moving from 59% to 100% in 30 days; that historical result is case context, not a target for every implementation.
Verify ROAS: design an appropriate incrementality test
A holdout can help estimate incremental effects when assignment, exposure, outcome definitions, sample size and the observation horizon support a meaningful comparison. Plan the experiment before changing spend, check contamination and uncertainty, and distinguish an inconclusive result from evidence of no material effect. Two weeks and ten percent of users are not universal design requirements; payment data alone does not establish valid assignment.
Verify collection, identity and cohort definitions
Client-side and server-side collection can each lose, duplicate or misclassify events. I inspect consent behavior, triggers, identifiers, delivery logs and reconciliation to eligible CRM or billing records before choosing the implementation. Server-side delivery does not guarantee a capture percentage or complete attribution. We agree scope, access, validation cases and delivery schedule.
A 90-day fractional CMO measurement plan
First phase: establish the measurement baseline. Inspect the event, customer and revenue definitions, access, consent and current records. Personally implement the agreed repairs and retain a baseline with eligible populations, matched and unmatched records, validation cases and open questions. Choose browser or server work from the actual problem.
Next phase: implement and review. Use the eligible evidence to select and implement the campaign, conversion, lifecycle or measurement changes. Record the baseline, cost, owner and observation horizon. If an incrementality test fits the decision and available volume, agree its assignment and acceptance before running it.
Review phase: decide the next action. Compare the agreed work and eligible observations with the baseline. Document the business decision, uncertainty, unresolved records and the next review. A cohort may still be maturing; neither an 80% match rate nor a calendar date proves tracking quality or improved performance.
For cost context on what this engagement costs relative to the value it produces, see whether a fractional CMO is worth it. For PE-specific measurement standards, see fractional CMO for private equity portfolios.
FAQ
Frequently asked questions about fractional CMO KPIs
How do I know if my fractional CMO's KPIs are real?
Reconcile the same eligible transactions, cohort, currency and period across analytics, CRM and billing. Inspect delivery, duplicates, identity matching, refunds and missing records; report deterministic matches, modeled credit and unknowns separately. A match percentage does not certify tracking. To estimate incremental impact, design a suitable comparison or controlled experiment with agreed assignment, outcome, sample and observation horizon. Report uncertainty and inconclusive results.
What KPIs should a fractional CMO report in their monthly review?
Select metrics for the actual business decision: eligible acquisition cost and new customers, attributed conversion value, gross contribution, source coverage, cohort transitions and qualified pipeline or retained value. State definitions, sources and windows. Attention metrics can support a brand objective; they are not substitutes for the relevant commercial outcome.
What is a good CAC for my stage?
Use your margins, retention, sales or purchase cycle, cash position and growth plan to set a useful acquisition-cost and payback target. Separate observed cohort contribution from a forecast. Neither 3:1 LTV:CAC nor six-month payback is a universal standard that decides whether your business can scale.
How quickly should a fractional CMO show measurable improvement?
Agree delivery milestones and a separate observation window. A released measurement repair can be checked before customer economics mature. I personally implement the work in scope and review what the eligible evidence supports; a fixed 30-, 60- or 90-day schedule does not guarantee better visibility or financial return.
What does a 90-day fractional CMO measurement plan include?
Use a first-quarter framework to inspect the evidence, personally implement the agreed priorities, then review eligible observations and the next decision. Define access, owners, costs and validation cases for each workstream. Actual timing follows the work and customer journey; pending cohorts remain pending.
Are you available to take on a new client right now?
I agree the scope, time commitment and availability for each client or project before work starts. The fastest way to find out where things stand is to contact me with your stage, your current growth, and the one metric you most need to move. I will reply within one day.
Which companies are the best match for a fractional CMO engagement with you?
The companies I serve best have product market fit and real traction: paying customers, revenue and a product people keep using. Pre-PMF, the first step is to establish whether a scoped engagement fits your funding, product readiness and growth problem. Growth work compounds on a real product and cannot create demand for a broken one.
What is attribution coverage and why does it matter?
Attribution coverage is the share of eligible confirmed revenue matched to an acquisition source under a documented rule. The report should identify the denominator, deterministic matches, modeled credit and unknowns. A match-rate threshold alone does not prove causality or establish that the remaining budget decisions are blind.
Why is platform ROAS not enough to judge a fractional CMO?
Platform ROAS is attributed conversion value divided by spend under that platform’s rules. It can differ from reconciled revenue and contribution because definitions, windows, eligible records and modeled credit differ. A contribution calculation and an incrementality study answer different questions. Reconcile the sources, then use an appropriate comparison to investigate causal lift; attribution alone does not establish it.
Which vanity metrics should I ignore when evaluating a fractional CMO?
Separate attention, demand and commercial outcomes. Impressions, engagement and branded search can inform an agreed brand objective; they do not alone establish revenue or causal impact. Evaluate qualified leads through opportunity and customer conversion, with sources, eligible populations and observation windows.
How should event collection be verified?
Client-side and server-side collection can each lose, duplicate or misclassify events. I inspect consent behavior, triggers, identifiers, delivery logs and reconciliation to eligible CRM or billing records before choosing the implementation. Server-side delivery does not guarantee a capture percentage or complete attribution. We agree scope, access, validation cases and delivery schedule.
What term do I sign up for with a fractional CMO, and how is it ended?
Contract terms, including duration and notice period, are agreed per engagement. Ending a fractional engagement involves no severance. Scope and exit terms are agreed before we start. If progress stalls, we review the scope and next steps under the agreed contract terms.
Next step
Want a measurement plan built for your business?
Start with a scoped diagnosis of the customer, event and revenue records, then the implementation needed for the next decision. I personally build and verify the agreed work. We set deliverables, access requirements and timing before starting, and you keep the documented findings.
Read research on the CMO role: Spencer Stuart CMO research · My track record to check: Yaniv Goldenberg on LinkedIn
Work with Yaniv
Fractional Head of Growth for B2B SaaS, B2C, and e-commerce. Revenue that compounds, not spikes. Any channel, any motion, any stage.
Explore the Fractional CMO Services hub