KPI Dashboards That Survive an Investor Review: The Twelve Metrics PE Sponsors Actually Read

Updated for 2026, The twelve metrics are unchanged, but two new lines now show up on every PE operating review: agent licensing exposure and SaaS-with-AI-feature spend. Pair this guide with our agent licensing meter field guide and the agent licensing pre-renewal audit briefing before the next package goes to the sponsor. The dashboard the engineering team built and the dashboard the operating partner reads are different artifacts In the engagements we have run with PE-backed mid-market portfolio companies, the most consistent finding is that the dashboard the company is most proud of, the one with thirty visualizations, drill-down hierarchies, and live database connections, is also the dashboard the operating partner does not open after the second monthly review. The dashboards that get read are sparser, slower-changing, more procurement-grade, and more obviously tied to the audited financials. The pattern is not subtle. Engineering teams build dashboards optimized for exploration; PE operating partners read dashboards optimized for accountability. Exploration favors interactivity, granularity, and breadth; accountability favors fixed layouts, period-over-period comparison, and depth on a small number of metrics that connect directly to the investment thesis. When the dashboard tries to serve both audiences, it serves neither, and the operating partner reverts to the monthly PDF the controller assembles by hand. The twelve metrics we describe below are the metrics we have observed PE sponsors actually engage with across operating reviews, and the architecture we describe is the architecture that survives an LP-level question about how a metric was calculated. Metric one: revenue and revenue retention Revenue at the top of the dashboard is the only line every reader will read, and the discipline is in what sits adjacent to it rather than the headline number itself. The headline is monthly revenue (or quarterly for non-subscription businesses), shown as actual versus prior-year, actual versus budget, and actual versus the last reforecast, with the variance commentary explicit and the variance attribution decomposed into volume, price, mix, and FX where applicable. For subscription or recurring-revenue businesses, the line below revenue is gross revenue retention (revenue from the prior-period cohort, excluding upsell and cross-sell, divided by the prior-period cohort's revenue) and net revenue retention (the same calculation including upsell and cross-sell). The PE sponsor reads gross retention as a churn signal and net retention as a land-and-expand signal, and the two together describe the underlying revenue quality in a way the headline does not. The reconciliation discipline is what separates a defensible retention metric from a marketing one. Gross retention should reconcile to the booked revenue in NetSuite or Sage Intacct under ASC 606 with a documented bridge from contract value to recognized revenue, and the cohort definition (calendar month of cohort entry, customer-level versus contract-level, treatment of multi-year contracts) should be written down and consistent period to period. We have seen retention metrics swing fifteen points across a single board meeting because the cohort definition changed between two analysts; this is the kind of credibility loss that does not recover. Metric two: gross margin and contribution margin Gross margin is the cleanest signal of operating leverage and product-mix shift, and the discipline is in the cost-of-revenue definition. For software businesses, COGS includes hosting, customer success allocated to fulfillment, third-party data and licensing fees, and the directly-attributable portion of professional services. For services businesses, COGS is fully-loaded delivery cost including burdened labor. For mixed-model businesses, the gross margin should be reported by revenue stream, with a consolidated weighted-average gross margin shown for context but not as the headline. Contribution margin sits below gross margin and adds the variable selling costs, sales commissions, sales allocations, marketing variable costs (digital spend, partner referral fees), to produce a margin that reflects the unit economics of revenue acquisition rather than only the unit economics of revenue delivery. Contribution margin is the metric the PE sponsor uses to evaluate whether a growth investment is creating value, and it is the metric that most often goes uncalculated because the cost allocation requires a working FP&A model rather than a query against the GL. The systems angle: gross margin should source from the GL (NetSuite, Sage Intacct, Workday, or the closing-system of record) with the COGS definition written into the chart of accounts. Contribution margin requires an allocation layer, Adaptive Planning, OneStream, Vena, Cube, or a maintained Excel model, with the allocation methodology documented and reviewed. The reconciliation from contribution margin back to gross margin and back to GAAP revenue and COGS is the bridge the auditor will eventually ask for in an audit-committee or readiness-for-audit review. Metric three: EBITDA and the adjusted EBITDA bridge EBITDA is the metric most central to PE investment decisions and most subject to definitional drift across periods, which is why the bridge from GAAP net income to EBITDA to adjusted EBITDA must be on the dashboard, not in a footnote. The bridge runs: net income, plus interest, plus taxes, plus depreciation and amortization, equals EBITDA; plus or minus stock-based compensation (depending on credit agreement and reporting convention), plus management fees, plus transaction expenses, plus restructuring expenses, plus the defined other-adjustments (typically capped by the credit agreement), equals adjusted EBITDA. Each adjustment line should be documented with a defined trigger (what qualifies as a "transaction expense", only a closed transaction, or also failed deal expenses?), a defined cap (some credit agreements cap permitted add-backs at a percentage of trailing-twelve-months EBITDA), and a defined approver (typically the CFO with audit-committee visibility for amounts above a defined threshold). The credit agreement is the controlling document for adjusted EBITDA in PE-backed companies, and the dashboard's adjusted EBITDA must tie to the credit-agreement definition or the lender will catch the mismatch in the next compliance certificate. The mistake we see most often is the "run-rate" adjustment, the add-back for hires not yet fully ramped, the synergy not yet fully captured, the contract not yet fully implemented. These add-backs are the ones that draw scrutiny from the next round of buyers, the ones that get written down in a quality-of-earnings review, and the ones that turn a board meeting into a discussion of whether the company has been telling itself a true story. We recommend treating run-rate adjustments as memo items below the bridge rather than as bridge components, with the underlying drivers tracked separately and revisited when the projection is realized. Metric four: working capital cycle (DSO, DPO, DIO) The working capital cycle is the metric most under-reported in mid-market dashboards relative to its importance to cash generation, and the three components, days sales outstanding, days payables outstanding, and days inventory outstanding, should appear together rather than in isolation. The cash conversion cycle (DSO plus DIO minus DPO) sums the three into a single days-of-cash-tied-up number that is the most direct read on working-capital efficiency. DSO calculation is more contested than the headline number suggests. The simple formula (AR / annualized revenue × 365) understates DSO for businesses with seasonal revenue and overstates DSO for businesses with high deferred revenue. The countback method (rolling backwards through revenue cohorts until cumulative revenue equals AR) is the more defensible calculation, and it is the one the audited financials' MD&A will eventually require. The dashboard should show DSO under both methods if the discrepancy is material. DPO and DIO follow analogous logic. DPO has a quality dimension, the team that lengthens DPO at the cost of strained vendor relationships is generating illusory cash improvements, and the dashboard should track vendor concentration and any disputes or holds as a counter-metric. DIO requires a working inventory accounting system (NetSuite Advanced Inventory, Sage Intacct with the inventory module, Fishbowl, NetSuite WMS, or a dedicated WMS) and is most often broken in mid-market businesses growing through acquisition where the acquired entity has a different inventory system that has not been integrated. Metric five: unit economics, CAC payback, LTV:CAC, and the customer cohort triangle Unit economics are most relevant for subscription and recurring-revenue businesses but increasingly relevant for any business with a defined customer-acquisition cost and a defined customer revenue stream. The two headline metrics are CAC payback (the months of contribution margin required to recover the customer-acquisition cost) and LTV:CAC (the customer lifetime value, calculated under defined retention and expansion assumptions, divided by the customer-acquisition cost). CAC definition is the input that most teams get wrong. CAC should include sales compensation (base, commission, and benefits), sales management, marketing program spend, marketing labor, sales operations, and any directly-attributable customer-success-on-implementation labor. The mistake we see most often is using only marketing program spend as the numerator, which produces a CAC that looks impressive and does not reconcile to GAAP operating expense. The cohort triangle (sometimes called the cohort retention curve or the cohort revenue triangle) is the underlying view that supports both metrics: each row is a cohort defined by acquisition month or quarter, each column is a tenure month or quarter from cohort entry, and each cell is the cohort's revenue (or active customer count, or net revenue retention) at that tenure. The triangle is what allows the PE sponsor to see whether retention is improving or deteriorating across cohorts, and it is the artifact that should sit one click below the headline LTV:CAC. The reconciliation: cohort revenue across all rows should sum to total recognized revenue, which should tie to the GL. We have seen cohort triangles that sum to a different number than the GL by ten or fifteen percent because of how renewals and upsells were attributed; this is the kind of error a quality-of-earnings analysis surfaces in week one of a sale process. Metric six: cash conversion cycle and the cash-from-operations tieback The cash conversion cycle (DSO + DIO − DPO) was introduced under metric four but warrants its own line because it is the headline number the PE sponsor reads to evaluate working capital management at a glance. The tieback to cash from operations is what makes the metric defensible: the change in cash conversion cycle, multiplied by the average daily revenue, should approximate the change in working capital component of the cash flow statement. Where the tieback fails, the cause is usually one of three: revenue is not being recognized on the same basis as the AR (deferred revenue mechanics under ASC 606), inventory turns are being calculated against an incorrect cost-of-revenue base, or DPO is being calculated against a different vendor population than the AP balance reflects. Each of these is a controllership question, not a dashboard question, and each warrants a discrete investigation rather than a footnote. Metric seven: leverage and covenant headroom Leverage and covenant headroom are the metrics most directly tied to the credit agreement and most directly relevant to the PE sponsor's downside management. The headline leverage metric is total funded debt to trailing-twelve-months adjusted EBITDA, calculated under the credit agreement's definitions of both numerator and denominator. The covenant headroom is the percentage gap between the actual leverage ratio and the covenant trip ratio, with the headroom on each defined covenant displayed. The dashboard should also show the projected covenant headroom at the next test date, which requires a forward look from the rolling forecast or the 13-week. Cross-link to /blog/13-week-cash-flow-operational-rhythm for the covenant projection mechanics. The PE sponsor reads this metric to evaluate whether a permitted-acquisition or distribution is feasible under current debt capacity, and the answer to that question is often the most important answer the dashboard provides in any given month. Metric eight: capex run-rate and capital allocation discipline Capex run-rate is the metric that distinguishes operating performance from capital allocation, and it is the metric most often under-reported because mid-market companies frequently lump maintenance capex and growth capex together under a single budget line. The discipline is to split capex into maintenance (required to sustain current operating capacity), growth (incremental capacity expansion), and transformation (one-time platform or technology investments), with each tracked against budget and against prior-year run rate. The PE sponsor reads capex to evaluate whether the underlying business is generating free cash flow at the rate the headline EBITDA suggests; a business with EBITDA growth funded by escalating maintenance capex is not generating the cash flow the multiple implies. Cross-link to /blog/capital-allocation-governance-board-framework for the capital allocation governance structure. Metric nine: headcount and headcount efficiency Headcount is the metric that most directly drives the operating expense base in services and software businesses, and the discipline is in the breakdown, by function (sales, marketing, R&D, G&A, customer success, fulfillment), by employee versus contractor, by location, and by full-time-equivalent versus full-time-headcount. The PE sponsor reads headcount to evaluate operating leverage; revenue per employee and EBITDA per employee are the headline efficiency ratios. The reconciliation: headcount in the dashboard should reconcile to the headcount in Workday, ADP, Rippling, BambooHR, or whatever HRIS is the system of record, and the reconciliation should explicitly handle contractors (typically not in the HRIS), open requisitions (sometimes counted, sometimes not), and pending separations (counted until last day of work). The mistake we see most is the dashboard reflecting headcount as of a different date than the financial statements, which produces a revenue-per-employee that does not reconcile. Metric ten: pipeline coverage and forecasting accuracy Pipeline coverage (open pipeline divided by remaining-period quota) and forecast accuracy (forecasted revenue divided by actual revenue, by period) are the metrics that connect the sales operating system to the financial operating system. The PE sponsor reads pipeline coverage to evaluate whether the next-period revenue forecast is supported, and reads forecast accuracy as a credibility metric on the sales leadership. The discipline: pipeline definitions (stage definitions, weighted versus unweighted, conversion-rate assumptions) should be consistent across periods, and the dashboard should track both the headline coverage ratio and the underlying funnel conversion rates so the reader can distinguish a pipeline-coverage problem from a conversion-rate problem. Pipeline that is not supported by recent conversion-rate evidence is pipeline that has not yet been earned. Metric eleven: customer concentration and revenue concentration Customer concentration is the metric that surfaces revenue-quality risk and is the metric most often under-reported on the headline dashboard despite being one of the metrics that most affects valuation. The headline is top-five-customer-concentration (the five largest customers' revenue divided by total revenue), with top-ten-customer-concentration as the secondary line. For businesses with channel partners, the dashboard should also show channel concentration. The PE sponsor reads concentration to evaluate downside exposure from customer loss; an exit-quality business has top-five concentration below twenty-five percent and top-ten concentration below forty percent, with no single customer above ten percent. Concentration above these thresholds does not preclude an exit, but it does change the buyer pool and the multiple, and the dashboard should surface the trajectory rather than only the snapshot. Metric twelve: free cash flow and cash conversion Free cash flow (operating cash flow minus capex) is the metric the lender, the next buyer, and the LP each ultimately read, and the cash conversion ratio (free cash flow divided by adjusted EBITDA) is the metric that bridges the operating performance narrative to the capital-structure narrative. A business with eighty percent cash conversion is generating cash at a rate that supports debt service, growth investment, and distributions; a business with forty percent cash conversion is funded by working capital expansion or external capital. The dashboard should show free cash flow on a trailing-twelve-months basis (to smooth seasonality), the cash conversion ratio, and the underlying drivers, operating cash flow, capex by category, and the change in working capital. Cross-link to /blog/management-vs-financial-reporting-boundary for how non-GAAP cash metrics interact with GAAP reporting. The tooling: Power BI, Tableau, Looker, native FP&A, or specialist platforms The platform decision is genuinely sizable and depends on the team's existing data infrastructure, the level of self-service expected, and the integration burden the team can sustain. We see four practical patterns. Power BI is the most common choice in mid-market portfolio companies running Microsoft 365, with the advantage of native integration with NetSuite (via API), Sage Intacct (via API), and the FP&A platform (Adaptive, OneStream, Vena), and the disadvantage of a steeper modeling curve than the marketing materials suggest. Tableau is the choice when the team already has Tableau Server, when the visualization needs are richer, or when the existing analytics team is Tableau-trained; the disadvantage is licensing cost at scale. Looker (now Looker Studio under Google) is the choice when the team is on Google Cloud Platform or has invested in dbt and a modern data warehouse (Snowflake, BigQuery, Databricks). The advantage is the LookML modeling layer, which enforces metric definitions across the dashboard; the disadvantage is the integration burden if the underlying data warehouse is not yet built. Native FP&A platform dashboards (Adaptive Planning's dashboards, OneStream's dashboards, Vena's dashboards, Mosaic, Pigment, Cube, Datarails) are the choice when the FP&A platform is the source of truth for the metrics and when the team wants to avoid maintaining a separate BI layer; the disadvantage is the limited visualization flexibility relative to Power BI or Tableau. Specialist platforms, Mosaic, Cube, Pigment, Datarails, sit between FP&A and BI and are built specifically for finance dashboards. The advantage is opinionated metric definitions and pre-built integrations to NetSuite, Sage Intacct, QuickBooks, Workday, ADP, and the major HRIS systems; the disadvantage is platform lock-in and the maturity gap relative to Power BI or Tableau. The architecture mistake we see most: dashboards built by engineers for engineers, ignored by the operating partner The most consistent failure pattern across the dashboards we have audited is the following: the dashboard is built by an internal analytics or data-engineering team optimizing for technical excellence, with thirty visualizations on a tabbed interface, drill-down hierarchies that go three levels deep, and live database connections that refresh every five minutes. The operating partner opens the dashboard once, recognizes that the layout requires twenty minutes to navigate to the answers needed in two minutes, and reverts to the monthly PDF the controller assembles by hand. The fix is not technical. The fix is to put the FP&A leader (or the controller, in smaller orgs) in the role of dashboard product owner, with the analytics team as the engineering function rather than the design function. The FP&A leader writes the layout based on what the operating partner reads, defines the twelve metrics and the period-over-period comparisons, and signs off on every visualization. The analytics team builds and maintains; the FP&A team owns the contract with the reader. The reconciliation discipline is the second part of the fix. Every metric on the dashboard should have a documented definition, a documented data lineage from source system to dashboard, and a documented owner who is responsible for the metric's accuracy. When the PE sponsor asks "how was this calculated," the answer should be available in two clicks rather than two days. We have audited dashboards where the answer was not available at all; this is the failure mode that loses the operating partner's trust permanently. What we recommend Pick the twelve metrics first, the layout second, and the platform third. The metric selection drives the layout, and the layout drives the platform requirements, but most teams reverse the order and end up with a sophisticated platform displaying metrics the operating partner does not read. Put a finance leader in the role of dashboard product owner. The analytics or data-engineering team builds, but the FP&A leader or controller signs off on every metric, every layout decision, and every period-over-period convention. Engineering ownership of the dashboard layout is the most consistent root cause of the dashboards-that-go-unread problem. Document every metric's definition, data lineage, and owner. The operating partner will eventually ask how a metric was calculated, and the answer should be two clicks deep, not two days deep. The dashboard's credibility is its reconciliation discipline, not its visual sophistication. Cross-link to /blog/board-reporting-decisions-not-status for the board package structure, /blog/management-vs-financial-reporting-boundary for the GAAP-vs-non-GAAP boundary, and /blog/investor-reporting-cadence-pe-portfolio for the broader PE reporting cadence.