Why this ebook exists
Almost every go-to-market KPI a board asks for today was invented inside one kind of company: a venture-funded, subscription-priced, direct-sold software business between roughly 2008 and 2015. Rule of 40, magic number, burn multiple, LTV/CAC, NRR. Inside that world they are excellent instruments.
Then they escaped. They now sit in board decks of companies that sell capital equipment through dealers, carry a twelve month backlog, book half the year in two quarters, and make most of their gross profit on parts. In those companies the metric does not merely fail to help. It points the operator in the wrong direction.
I ran the Brazil business of Ag Growth International from roughly US$35M to more than US$150M in three years, then carried country and later regional P&L responsibility in North America. That is capital goods: long cycle, channel dependent, seasonal, backlog driven, aftermarket rich. I have sat in the meeting where somebody asks for the magic number of a grain handling business and everyone nods.
How to read each entry
- What it is. One self-contained sentence.
- The formula.
- Benchmark, with source. Organization and year. Where no credible public benchmark exists, it says so. Nothing here is invented.
- How this KPI lies to you.
- What to do with it.
- Does it transfer to a physical economy business?
A benchmark is a distribution, not a target. When a source reports a median, half the sample sits below it. Use benchmarks to ask better questions, never to set a bonus.
Part 1: The acquisition KPIs
1. Customer acquisition cost (CAC)
What it is. The fully loaded sales and marketing cost to win one new customer.
The formula. Sales and marketing expense divided by new customers acquired. The better board variant is the CAC ratio: that expense divided by new annualized revenue won.
Benchmark, with source. Benchmarkit, 2025 B2B SaaS Performance Metrics: the median company spends US$2.00 of sales and marketing expense per US$1.00 of new customer ARR, against a median expansion CAC ratio of US$1.00, with sales and marketing at 37% of revenue. For a physical economy business there is no reliable public benchmark; use peer SG&A as a percentage of net sales from their 10-K filings.
How this KPI lies to you. It lags, so a company that cuts marketing hard posts a beautiful CAC for two or three quarters before the pipeline collapses. And “fully loaded” is a decision, not a fact: include sales engineers and CAC jumps 30 percent, exclude channel margin and a dealer-led business appears to acquire customers for free.
What to do with it. Never look at one CAC. Look at it by segment and channel, with the cost definition written down and frozen so it cannot drift between board meetings.
Does it transfer? Yes, with an adjustment. In a channel business the true cost includes dealer margin plus dealer support (co-op, demo units, training, floor plan interest). Most manufacturers exclude channel margin, understating the real cost of a customer by a wide multiple.
2. CAC payback
What it is. How many months of gross profit from a new customer it takes to earn back what you spent to win them.
The formula. CAC divided by (new monthly recurring revenue multiplied by gross margin percentage).
Benchmark, with source. KeyBanc Capital Markets and Sapphire Ventures, 2024 Private SaaS Company Survey (104 companies, median ARR US$26M): median payback of 20 months. Benchmarkit, 2025: payback has lengthened 12.5% since 2022. David Skok’s “SaaS Metrics 2.0” (Matrix Partners, For Entrepreneurs, circa 2011) suggested 5 to 7 months, which shows how far reality drifted from the canonical rule.
How this KPI lies to you. It assumes the customer survives the payback window. A 20 month payback with 25 percent annual logo churn means a meaningful share of customers never pay you back at all, and the average conceals that completely.
What to do with it. Compare payback months against median customer life in months. If payback exceeds a third of median life, you do not have an acquisition efficiency problem. You have a retention problem wearing an acquisition costume.
Does it transfer? Yes, and it matters more in the physical economy. The first machine is often sold at thin margin to install the customer, with the return coming from parts, service, and the replacement cycle. Compute payback on installed-base gross profit, not on the equipment sale. A deal that looks like a loss becomes a 14 month payback with a ten year annuity behind it.
3. Lifetime value (LTV)
What it is. The total gross profit you expect from a customer across the whole relationship.
The formula. (Average revenue per account multiplied by gross margin percentage) divided by revenue churn rate.
Benchmark, with source. No reliable public benchmark exists for LTV as an absolute number, because it is not comparable across price points or segments. The inputs are benchmarkable: gross margin (Benchmarkit, 2025: median 77% for B2B SaaS; David Sacks, Craft Ventures, “The SaaS Metrics That Matter”, 2021: at least 75% long term) and retention, covered below.
How this KPI lies to you. LTV is a forecast dressed as a measurement: it observes churn over a short window and projects it forward for years, so a company with 18 months of history computes a customer life longer than the company has existed. The formula is a hyperbola, so small errors in churn produce enormous errors, always in the flattering direction.
What to do with it. Cap the horizon. Compute three year gross profit per customer instead. It is defensible, auditable against actuals, and it does not explode when churn is low.
Does it transfer? Yes, with a heavy adjustment. Lifetime value in capital goods is real and often enormous, but lumpy rather than smooth: machine, then parts and service for eight to fifteen years, then a replacement. Model it as an installed-base annuity. The churn analogue is losing the parts business to a will-fit competitor while your machine is still in the field.
4. LTV/CAC ratio (the most abused metric in go-to-market)
What it is. Expected lifetime gross profit per customer divided by the cost of acquiring that customer.
The formula. LTV divided by CAC.
Benchmark, with source. David Skok, Matrix Partners, “SaaS Metrics 2.0” (For Entrepreneurs, circa 2011): “the best SaaS businesses have an LTV to CAC ratio that is higher than 3, sometimes as high as 7 or 8.” That is the origin of the 3:1 rule. Two things get dropped whenever it is quoted: Skok described mature, steady-state businesses, and he explicitly warned that “there are always situations where it makes sense to break them.”
How this KPI lies to you. It is a ratio of a soft forecast to a definitional choice, and both move without the business changing. Drop the churn assumption from 12 percent to 9 and LTV rises a third; move sales engineering out of the CAC definition and CAC falls a fifth. Those two edits, neither touching a customer, take the ratio from 2.5 to 4.2. Then the structural lie: a high ratio is frequently a symptom of underinvestment, so the board celebrates the number that should have triggered the question.
What to do with it. Demote it. Use CAC payback as the primary efficiency metric, because months are auditable. If you must show LTV/CAC, print the churn assumption and the CAC definition on the same slide.
Does it transfer? No. The clearest non-transfer in the list. The numerator needs a ten to fifteen year installed-base forecast and the denominator is dominated by channel margin the ratio was never built to hold. Replace it with three year installed-base gross profit per unit placed, divided by the all-in cost to place that unit.
Part 2: The retention KPIs
5. Net revenue retention (NRR)
What it is. What happened to last year’s customer cohort this year, counting expansion, contraction, and churn, excluding new logos.
The formula. (Starting cohort revenue plus expansion minus contraction minus churn) divided by starting cohort revenue.
Benchmark, with source. Benchmarkit, 2025 B2B SaaS Performance Metrics: median NRR of 101%. SaaS Capital, 2025: median of 102% for annual contract values between US$25,000 and US$50,000, and for bootstrapped companies between US$3M and US$20M ARR a median of 103% with the 90th percentile at 117.9%. Note how far that sits from the 120 percent quoted in board meetings as if it were normal.
How this KPI lies to you. It nets two opposite things and reports one number. A company at 105 percent could have 98 percent gross retention with steady upsell (healthy) or 80 percent gross retention rescued by three enormous expansions (a fire, currently hidden). It also flatters seat-based pricing, because the customer’s own hiring reads as your retention.
What to do with it. Never show NRR without GRR beside it. The gap between them is the message: a wide gap means expansion is papering over churn.
Does it transfer? No, not as defined. A capital business has no contractual renewal event, so the arithmetic has no anchor: a customer who bought a US$400,000 machine last year and buys nothing this year is mid-cycle, not churned. The substitute is aftermarket revenue retention on the installed base (parts plus service from the same fleet, year over year), which has a real annual cadence.
6. Gross revenue retention (GRR)
What it is. The same cohort measurement as NRR with expansion set to zero, so it can never exceed 100 percent.
The formula. (Starting cohort revenue minus contraction minus churn) divided by starting cohort revenue.
Benchmark, with source. Benchmarkit, 2025 B2B SaaS Performance Metrics: median GRR of 88%. As a logo-retention cross-check, David Sacks, Craft Ventures (2021): 90 to 95 percent for enterprise, 85 percent for mid-market, 70 to 80 percent for small business.
How this KPI lies to you. It is the more honest of the pair, but it lies about timing. A customer who has already decided to leave but is contractually locked for eleven more months counts as fully retained, so with multi-year contracts GRR reports the churn decision one board cycle too late to act on.
What to do with it. Pair it with a leading indicator that moves before the contract does: usage decline, support sentiment, or in a physical business, a drop in parts order frequency from an account that still owns the machine.
Does it transfer? Yes, with an adjustment. Applied to the aftermarket line rather than the whole account, GRR works well and is genuinely diagnostic. An account whose parts spend fell 40 percent while its fleet stayed the same size is being served by somebody else, and you learn that roughly two years before you lose the replacement sale.
7. Revenue mix: new versus expansion versus renewal
What it is. The share of period revenue from brand new customers, from selling more to existing ones, and from customers continuing at the same level.
The formula. Three percentages of total period revenue (new logo, expansion, flat renewal or repeat) summing to 100 percent.
Benchmark, with source. There is no single public benchmark for the mix percentages themselves, but the economics are well sourced. Benchmarkit, 2025 B2B SaaS Performance Metrics: median new customer CAC ratio of US$2.00 per dollar of new ARR against a median expansion CAC ratio of US$1.00. Expansion revenue costs half as much to produce.
How this KPI lies to you. Mix is the KPI most exposed to definitional gaming, because whoever reports it usually owns the target: a price increase booked as expansion, a cross-sell to another division of the same parent booked as new logo. It also prompts the wrong conclusion, because boards see that expansion is cheaper and push for more of it, which in a finite installed base is a plan to run out of runway quietly.
What to do with it. Fix the definitions in writing, including the ambiguous cases. Track mix as a three year trend: a business whose new logo share falls every year is consuming its base, and it will look excellent right up until it does not.
Does it transfer? Yes, and it is one of the most useful metrics in the physical economy. Reframe the buckets as conquest (a machine placed where a competitor’s fleet used to run), fleet expansion, and replacement. A business growing only on replacement is growing on the market’s cycle, not on its own competitiveness.
Part 3: The efficiency KPIs
8. Rule of 40
What it is. A single test holding that growth rate plus profit margin should sum to at least 40.
The formula. Revenue growth rate (percent) plus profit margin (percent, usually EBITDA or free cash flow margin).
Benchmark, with source. Popularized in 2015 by Brad Feld and Fred Wilson in near-simultaneous blog posts. Bessemer Venture Partners, State of the Cloud 2022: companies at or above the Rule of 40 carried valuation multiples 65% to 75% higher than peers below it. Bessemer itself moved away from it in 2024, proposing a “Rule of X” that weights growth more heavily than margin.
How this KPI lies to you. It treats a point of growth and a point of margin as interchangeable. A company at 60 percent growth and negative 20 percent margin scores 40, and so does one at 5 percent growth and 35 percent margin. It also breaks in a cyclical business, swinging 40 points on the market’s rhythm rather than on management performance.
What to do with it. Use it to open a conversation about the growth and margin trade-off, never as a score. If the board wants one number, insist the two components appear beside it.
Does it transfer? With a heavy adjustment, and only smoothed. Cyclical input volatility makes the annual figure meaningless. Use three year rolling averages of both components, and set the threshold from your own industry’s history rather than importing 40 from software. A 40 target in a lower-growth business is a target to under-invest.
9. Magic number
What it is. How much new annualized revenue each dollar of sales and marketing spend produced in the following quarter.
The formula. (Current quarter ARR minus previous quarter ARR) divided by the previous quarter’s sales and marketing expense.
Benchmark, with source. The concept originated at Scale Venture Partners, where Rory O’Driscoll observed while evaluating Omniture that the company generated more than US$2 of first-year revenue per US$1 of go-to-market investment. It was popularized by Lars Leckie of Hummer Winblad Venture Partners in 2008. The conventional reading (above roughly 0.75 means spend more, below 0.5 means fix the engine first) is a circulated heuristic, not a published statistical benchmark.
How this KPI lies to you. It hard-codes a one quarter lag between spend and revenue, which is right for a self-serve motion and wrong for any sales cycle longer than 90 days. With a nine month cycle, this quarter’s revenue came from spend three quarters ago, so the metric divides an output by an unrelated input. It is also violently unstable under seasonality.
What to do with it. If your sales cycle is under a quarter, it is a fast and useful read. If it is not, do not use it. Use CAC payback and the CAC ratio, which cope with lag by construction.
Does it transfer? No. A capital equipment business has a cycle measured in quarters or years and revenue recognized against a backlog booked long before, so the lag assumption never holds. The closest defensible substitute is order intake per dollar of commercial spend on a trailing twelve month basis.
10. Burn multiple
What it is. How many dollars of cash the business burned to create one dollar of new annual recurring revenue.
The formula. Net burn divided by net new ARR.
Benchmark, with source. David Sacks, Craft Ventures, “The Burn Multiple” (April 2020) and “The SaaS Metrics That Matter” (October 2021): “a Burn Multiple of less than one is amazing, but anything less than two is still quite good.” Several more granular label tables circulate online attributed to Sacks; the sentence above is what the canonical posts actually say.
How this KPI lies to you. The denominator is recurring revenue, so any business whose growth arrives as one-time revenue gets an absurd result. More subtly, net burn includes working capital movements, so a company that simply delayed paying vendors posts an improved burn multiple while nothing improved.
What to do with it. Useful for a venture-funded, subscription-priced business judging whether growth is worth what it costs. Outside that context, use free cash flow conversion and cash conversion cycle, which handle inventory and receivables honestly.
Does it transfer? No. Two independent reasons, either fatal. There is no ARR denominator in a capital business, and a physical business burns cash into inventory and receivables as a structural feature of growing, so the metric flags healthy expansion as a problem. Substitute the cash conversion cycle plus free cash flow as a percentage of EBITDA.
11. Revenue per employee
What it is. Total revenue divided by full time equivalent headcount, used as a coarse productivity measure.
The formula. Revenue (or ARR) divided by average FTE count in the period.
Benchmark, with source. Benchmarkit, 2025 B2B SaaS Performance Metrics: median ARR per FTE of US$200,000 for companies between US$50M and US$100M ARR, rising to US$300,000 above US$100M. For manufacturing and distribution there is no reliable public benchmark comparable across subsectors, and that is a property of the metric rather than a research gap: vertical integration alone can move it threefold. Compare against named public peers, using the headcount disclosed in Item 1 of their 10-K.
How this KPI lies to you. It rewards outsourcing rather than productivity: move manufacturing to a contract manufacturer and the number doubles overnight with no change in how well the business runs. It also ignores margin, so a shift toward low-margin pass-through revenue improves it while destroying value.
What to do with it. Use gross profit per employee instead. It is immune to the pass-through distortion and largely immune to the outsourcing one, because outsourcing moves cost into COGS where it belongs.
Does it transfer? Yes, but only in the gross profit form and only against your own peer set. Never compare a manufacturer’s revenue per employee to a software company’s. The two numbers describe different physics.
Part 4: The risk KPIs
12. Customer concentration
What it is. How much of your revenue depends on a small number of customers.
The formula. Revenue from the top customer as a percentage of total, plus the same for the top five and top ten. Present all three.
Benchmark, with source. The anchor is regulatory, not statistical. FASB ASC 280-10-50-42 requires a public entity to disclose the fact and the amount when revenue from a single external customer reaches 10% or more of total revenue. There is no reliable public benchmark for private company concentration, though acquirers and lenders anchor to the same 10 percent in practice.
How this KPI lies to you. It hides common control. ASC 280 itself notes that entities under common control count as one customer, so a business reporting “no customer above 8 percent” may have four of its top ten owned by the same sponsor. It hides through the channel too: fifty dealers looks well distributed until eight of them are consolidated by one holding company.
What to do with it. Compute concentration on the ultimate parent, not the billing entity, then a second version on decision maker: if one buying group decides for twelve legal entities, that is one customer.
Does it transfer? Yes, formula unchanged, with an addition. Capital businesses need a second measure software does not: concentration by end market, crop, or commodity. A grain equipment business with 200 well distributed customers all exposed to the same corn price has one customer, and it is corn.
13. Gross margin by revenue line
What it is. Gross profit as a percentage of revenue, computed separately for each revenue stream rather than for the company as a whole.
The formula. For each line: (line revenue minus line cost of goods sold) divided by line revenue. Allocation rules must be written down.
Benchmark, with source. Benchmarkit, 2025 B2B SaaS Performance Metrics: median total gross margin of 77%; David Sacks, Craft Ventures (2021): at least 75% long term. For equipment dealers, the North American Equipment Dealers Association 2025 Cost of Doing Business Study (2024 results, as reported by Farm Equipment) gives, for dealers between US$200M and US$600M, a parts gross margin of 30.06% and a service gross margin of 67.38%; above US$600M, 26.98% and 52.18%. The largest dealers show lower margins on both lines.
How this KPI lies to you. Blended gross margin is one of the most effective concealment devices in business reporting, because it is not a lie in any auditable sense. A company at a healthy blended 42 percent can run one line at 65 percent and another at 8 percent, with the second growing faster. Each quarter the blend slips, everyone blames pricing pressure, and nobody names the line responsible.
What to do with it. Report every line separately, every period, with the allocation rule attached and unchanged. Report mix shift alongside, because most blended margin movement is mix, not price.
Does it transfer? Yes, and in the physical economy it is close to indispensable. The NAEDA figures show why: service runs at roughly twice the gross margin of parts, and parts at multiples of wholegoods. A business managing to a blended number is managing to an average of three businesses with different economics.
14. Forecast accuracy at board level
What it is. How close the number the CRO committed to at the start of the quarter came to what the company delivered.
The formula. 1 minus the absolute value of (actual minus forecast) divided by forecast. Use absolute error, so beating the forecast counts as a miss, because it is one.
Benchmark, with source. CSO Insights, 2017 World-Class Sales Practices Study: the win rate of forecasted deals is 46.9%, and only 48.2% of deals closed as originally forecast. At the median organization, a deal in the committed forecast is roughly a coin flip.
How this KPI lies to you. Netting. A quarter where one deal came in US$2M early and another slipped US$2M out reports as perfect accuracy and represents a process that knows nothing. The other lie is sandbagging: measure accuracy without penalizing overachievement and you have built an incentive to forecast low.
What to do with it. Report two numbers: total-level absolute error and deal-level absolute error. Penalize overachievement symmetrically. Track each forecaster across eight quarters, because forecast bias is a stable personal trait you can simply correct for.
Does it transfer? Yes, and it becomes more consequential. The forecast drives steel purchasing, production slotting, and dealer floor plan, so optimism converts directly into inventory you financed and cannot sell. Report accuracy separately for the backlog-covered part and the still-to-be-sold part. Conflating them is how a business comes to believe it has 90 percent accuracy when the accurate 90 percent was the backlog reading itself back.
Part 5: The KPIs the SaaS canon never wrote down
These three appear in no software metrics framework, because software has no equivalent. In a capital and channel business they carry more decision weight than anything above.
15. Backlog and book-to-bill
What it is. Backlog is orders received and not yet shipped. Book-to-bill is orders taken over orders shipped in the same period, telling you whether backlog is growing or being consumed.
The formula. Book-to-bill = orders booked divided by revenue billed. Backlog coverage in months = ending backlog divided by average monthly revenue.
Benchmark, with source. SEMI publishes a monthly Book-to-Bill Express Report for the North American semiconductor equipment industry: a three month moving average of bookings divided by a three month moving average of shipments, where 1.09 means US$109 of orders per US$100 billed and 1.0 is the parity line. For broad manufacturing, the US Census Bureau M3 survey (Manufacturers’ Shipments, Inventories, and Orders, monthly since 1957) publishes unfilled orders, and the derived unfilled orders to shipments ratio is FRED series UMTMUS. Pull the current reading rather than quoting a stale one. There is no cross-industry benchmark for what backlog coverage should be, because your production lead time sets it.
How this KPI lies to you. Backlog lies about quality: twelve months priced at last year’s steel cost is not an asset, it is a margin liability you have already signed. It lies about cancellability, because backlog a dealer can cancel without penalty evaporates exactly when the cycle turns. Book-to-bill lies through its denominator, spiking above 1.0 when shipments collapse for a supply reason.
What to do with it. Report backlog in three cuts every board meeting: total value, months of coverage at the current shipment rate, and margin embedded in backlog at current input costs. That third cut is the one almost nobody produces and the one that predicts the next four quarters of gross margin.
Does it transfer? It transfers in the other direction. This is a physical economy native metric, and its software analogue is contracted revenue not yet recognized (remaining performance obligation). Any capital business not running backlog as a primary board metric is flying with its most predictive instrument switched off.
16. Channel inventory turns
What it is. How many times per year the inventory sitting in your dealer or distributor network converts into end user sales.
The formula. Cost of goods sold divided by average inventory value. Parts turn rate, as used in dealerships: trailing twelve month parts sales at cost divided by average parts inventory.
Benchmark, with source. Farm Equipment, reporting on dealer operations, gives an industry benchmark of 2.5 to 3 turns for equipment inventory. Kelly Mathison of the Dealer Institute, in Equipment Dealer Magazine (August 2021), reports that “many high-performance dealers are happy with parts turn rates between two and three times.” The NAEDA 2025 Cost of Doing Business Study (2024 results, via Farm Equipment) found turns down across dealer size groups, with the above US$600M group falling 0.34 turns, and used inventory ranging from 9.42% to 25.2% of total.
How this KPI lies to you. The most dangerous lie in this document is here, and it is specific to manufacturers. Your revenue is recognized when the machine reaches the dealer; the dealer’s inventory turn measures whether it reached an end user. A manufacturer can post a record year, pay the bonus, and be measuring nothing except how much inventory it pushed into a channel that has not sold it. The correction arrives one to three quarters later as an order drought nobody forecast.
What to do with it. Insist on retail sell-through data from the channel, as a condition of the dealer agreement rather than a favor. Report dealer inventory in months of retail supply, by dealer, showing the distribution. A quarter where sell-in exceeded sell-through is a quarter you borrowed from, and the board deck should say so.
Does it transfer? No transfer needed: it is native to the physical economy and has no software equivalent. The closest analogue is unused licenses sitting on a customer’s shelf, which fails at renewal for the same reason: revenue was recognized on a purchase that never became consumption.
17. Parts and service revenue over equipment revenue (absorption)
What it is. How much of the business is the aftermarket annuity rather than the machine sale, and in dealer terms, how much of total fixed expense the aftermarket departments cover on their own.
The formula. Aftermarket ratio = (parts revenue plus service revenue) divided by equipment revenue. Absorption rate = aftermarket department gross profit divided by total dealership operating expense.
Benchmark, with source. North American Equipment Dealers Association, 2025 Cost of Doing Business Study (2024 results, via Farm Equipment): parts aftermarket absorption of 38.45% for dealers between US$200M and US$600M, and 26.68% above US$600M. Associated Equipment Distributors publishes a parallel Cost of Doing Business Report for construction equipment distributors. Currie Management Consultants defines absorption as the share of a dealership’s total expenses covered by its aftermarket departments.
How this KPI lies to you. It moves for the wrong reason. When machine sales fall in a downturn, absorption improves, so the dealer principal presents a rising rate in the worst year of the cycle and it reads as progress when it is arithmetic. It also hides warranty work booked into service revenue at internal rates.
What to do with it. Track absorption alongside absolute aftermarket gross profit dollars: if the ratio rises while the dollars stay flat, nothing improved. Also track aftermarket revenue per machine in the field, which is cycle-neutral and measures what you control.
Does it transfer? It is the native metric, and it transfers to software conceptually. The analogue is services and support revenue over subscription revenue, which most SaaS boards ignore. In a capital business this ratio is often the difference between surviving the cycle and not, because the aftermarket annuity keeps paying while machine sales are down 40 percent.
The three SaaS KPIs that break in a capital business, and what to use instead
Four of the seventeen were marked as not transferable. Three of them break in ways that actively cause bad decisions, and they are the three most likely to appear in a board deck somebody copied from a software company.
| SaaS KPI | Why it breaks | Use instead |
|---|---|---|
| LTV/CAC ratio | The numerator needs a ten to fifteen year installed-base forecast; the denominator omits channel margin. Both are soft in the flattering direction, and a high ratio usually signals underinvestment while getting celebrated as discipline. | Three year installed-base gross profit per unit placed, divided by the all-in cost to place it, dealer margin included. |
| Net revenue retention | There is no annual renewal event to retain. A customer mid-cycle on a fifteen year machine reads as churn, so NRR reports a healthy capital business as a disaster. | Aftermarket revenue retention on the installed base (parts plus service from the same fleet, year over year) plus replacement capture rate (share of customers whose replacement machine you won). |
| Magic number and burn multiple | Magic number hard-codes a one quarter lag, wrong by an order of magnitude when the cycle is a year. Burn multiple needs an ARR denominator that does not exist, and treats inventory investment (the normal cost of growing) as inefficiency. | Order intake per dollar of trailing twelve month commercial spend in place of magic number. Cash conversion cycle plus free cash flow as a percentage of EBITDA in place of burn multiple. |
The pattern is worth naming: each assumes revenue arrives smoothly, on a contractual cadence, from a direct relationship, with no goods in between. Break any one of those four assumptions and the metric does not degrade gracefully. It inverts.
The one-page dashboard the board actually needs
Seven numbers. Not twelve. A board meeting holds about twenty minutes of genuine attention for performance, and a dashboard demanding more gets skimmed, which is worse than a short one that gets read.
| # | The number | Why it earns a place |
|---|---|---|
| 1 | Order intake, trailing twelve months, versus prior year, with book-to-bill beside it | Revenue describes the past; orders describe the next four quarters. In a backlog business this is the only forward-looking number that is a fact rather than an opinion. |
| 2 | Backlog: months of coverage and margin embedded at current input costs | Coverage says how long the factory is safe. Embedded margin says whether those months are profitable, which is the question nobody asks. Together they predict gross margin one to three quarters out. |
| 3 | Gross margin by revenue line, with mix shift | Blended margin conceals. Line margin plus mix tells the board whether a movement was price, cost, or mix. The NAEDA spread (service near 67 percent against parts near 30 percent) shows how much the blend destroys. |
| 4 | Aftermarket gross profit dollars and aftermarket revenue per unit in the field | The annuity that carries the business through the down cycle. In dollars, not as a ratio, so it cannot improve merely because machine sales fell. |
| 5 | Channel inventory in months of retail supply, showing the distribution and not just the average | The best early warning of an order drought. If sell-in has run ahead of sell-through, this is where it shows before it becomes a surprise. |
| 6 | Customer concentration on ultimate parent, plus end-market concentration | The 10 percent line from ASC 280-10-50-42 is the anchor. Computing on the ultimate parent makes it honest, and the end-market cut catches the risk that 200 customers can still be one bet. |
| 7 | Forecast accuracy: absolute error, backlog portion and to-be-sold portion reported separately | Not about revenue: about whether the other six can be believed going forward. A CRO whose forecast holds has earned the right to be trusted on the plan. |
Deliberately absent: Rule of 40, magic number, burn multiple, LTV/CAC, revenue per employee, NRR. Not because they are worthless, but because at board altitude each is either a derivative of the seven above or a number that cannot be acted on in the room.
If your business is subscription software, swap items 1, 2, and 5 for net new ARR, remaining performance obligation, and NRR paired with GRR. Items 3, 4, 6, and 7 stay as written. More than half of what a board needs is identical in both worlds. The disagreement lives entirely in how demand arrives and whether goods sit between you and the customer.
The rule underneath all of it
Every KPI here is a compression of reality, and compression always discards something. The discipline is not picking the right metric. It is knowing, for each number on your dashboard, exactly what it discarded, and checking that thing separately.
NRR discarded the difference between churn and expansion, so you check GRR. Blended gross margin discarded the mix, so you check the lines. Absorption discarded the cycle, so you check the dollars. Backlog discarded the input cost, so you check embedded margin. Sell-in discarded the end user, so you check sell-through.
A go-to-market leader who can name what each number on the board deck is hiding is worth more than one who can recite twenty benchmarks. The benchmarks are here and they are sourced. The naming is the job.
Andre Magrini is a fractional Chief Revenue Officer based in Chicago. He led Ag Growth International’s Brazil operation from roughly US$35M to more than US$150M in three years, and subsequently held country and regional P&L responsibility in North America within the same company.