R&D Monthly Spend
Total monthly cash outflow on research and development — fully-loaded engineering, product, and design payroll plus tooling, infrastructure dedicated to product development, contractors, and direct R&D vendor spend. The "input" side of R&D efficiency. Common pitfall: companies report base-payroll R&D and exclude the loaded cost (benefits, stock comp at cash-cost basis, allocated rent, dev tooling), under-reporting true R&D burn by 25–40%. Boards should always ask whether the number is base-payroll, fully-loaded, or GAAP R&D expense — they tell different stories. The KBCM/Sapphire SaaS Survey reports R&D as a percentage of revenue for its company panel — use that as the benchmarking lens. — Product KPI anchored to KBCM/Sapphire SaaS Survey 2024 (15th Annual).
Rogue ID: product.rd_monthly_spend
Type: Currency (/month)
Domain: Product
Definition
Total monthly cash outflow on research and development — fully-loaded engineering, product, and design payroll plus tooling, infrastructure dedicated to product development, contractors, and direct R&D vendor spend. The "input" side of R&D efficiency. Common pitfall: companies report base-payroll R&D and exclude the loaded cost (benefits, stock comp at cash-cost basis, allocated rent, dev tooling), under-reporting true R&D burn by 25–40%. Boards should always ask whether the number is base-payroll, fully-loaded, or GAAP R&D expense — they tell different stories. The KBCM/Sapphire SaaS Survey reports R&D as a percentage of revenue for its company panel — use that as the benchmarking lens.
Formula
Sum of fully-loaded R&D-team payroll + benefits + allocated stock-comp + R&D-dedicated infrastructure + R&D tooling + R&D vendor spend, expressed as a monthly figure. Different from GAAP R&D expense (which capitalizes some software development costs); footnote the convention.Why it matters
Largest single line of operating spend at most growth-stage SaaS companies — the input that rd_efficiency converts into revenue. The board reads this to gauge whether the company is over- or under-investing in product velocity relative to revenue ramp.
How to interpret
Compare R&D spend to revenue (or ARR run-rate) to derive R&D-as-% of revenue. Per the KBCM/Sapphire SaaS Survey (latest annual edition — see capital-allocation section), median R&D-as-% of revenue runs ~25–35% at early-growth SaaS and compresses with scale. Out-of-band (e.g. 60%+ at a $20M ARR company) usually signals either heavy platform-investment cycles or under-monetization — flag for context. Always pull the current KBCM/Sapphire edition rather than relying on a memorized range.
Calculation policy
How an AI agent should compute this KPI from messy company data. Free-text rules consumed at reasoning time — not a deterministic DSL. The most common ways to get this wrong are listed under Common miscomputations.
Inclusion rules
- Fully-loaded R&D-team compensation (salary + employer taxes + benefits + cash-basis stock-comp) for engineering, data, QA, DevOps, and the product/design staff building the product, expressed as a monthly figure.
- R&D-dedicated infrastructure and development / CI tooling, plus direct R&D vendor spend and R&D contractors / dev shops.
- Express as a per-month cash outflow: for multi-month source data, divide by the months in the period and state the smoothing window (a trailing-3-month average is common to dampen lumpiness).
Exclusion rules
- Production cloud / hosting that serves customers — that is COGS (
finance.cloud_hosting), not R&D infrastructure. - Sales, marketing, customer-success, and G&A payroll and tooling — only product-building functions count toward R&D.
- GAAP-capitalized software-development labor when the company capitalizes — footnote the convention; capitalized cost is not the cash R&D burn this KPI tracks.
- Non-cash equity expense beyond the cash-cost basis already included — this is a cash view.
Required inputs
- Fully-loaded R&D payroll by person / department.
- R&D-dedicated infrastructure + tooling + vendor + contractor spend for the period.
- Period length, to normalize to a monthly figure.
- The reporting convention flag: base-payroll vs fully-loaded vs GAAP-R&D expense.
Data-source priority
- Finance / FP&A general-ledger R&D cost-center roll-up — the authoritative spend source.
- Payroll system grouped by R&D department, plus AP detail for tooling / vendor / contractor spend.
- Engineering-leadership estimate only when GL cost-centers do not cleanly isolate R&D — flag the lower confidence.
Edge cases
- An engineer split between R&D and customer-support / COGS work: allocate by time, not by whole headcount.
- Capitalized vs expensed development cost: pick one convention (cash R&D burn here) and footnote it; do not mix a capitalized figure into a cash-burn series.
- Lumpy annual tooling / vendor prepayments: amortize across the months they cover rather than spiking a single month.
Validation checks
- This is the SAME underlying R&D spend as the finance roll-up
finance.total_rnd(which aggregatesfinance.rd_payroll+finance.product_design_payroll+finance.rd_tools_software+ the R&D share offinance.contractors_outsourcing) —product.rd_monthly_spendis the product-view,finance.total_rndthe finance-view, of one number. Reconcile them to within a small variance and NEVER add the two (or addrd_monthly_spendto anyfinance.rd_*component), or R&D spend silently double-counts. - R&D-as-% of revenue (rd_monthly_spend × period ÷ revenue) should sit in the KBCM/Sapphire band for the stage (~25–35% at early-growth SaaS, compressing with scale); out-of-band is a context flag, not a story.
- Step-changes should map to hiring, comp true-ups, or a tooling purchase — an unexplained jump usually signals a cost-center reclassification.
Common miscomputations
- Reporting base payroll instead of fully-loaded cost — under-reports true R&D burn by 25–40%.
- Summing
product.rd_monthly_spendandfinance.total_rnd(or its components) as if additive — they are two views of the same spend; this double-counts R&D. - Folding production cloud-hosting COGS into R&D infrastructure — inflates R&D and understates gross margin.
- Mixing a GAAP-capitalized R&D figure into a cash-burn series — the two diverge by the capitalized-labor amount.
Related KPIs
product.rd_efficiencyproduct.total_engineersproduct.innovation_capacity_pctsales.arrfinance.net_burn_rate
Source
KBCM/Sapphire SaaS Survey 2024 (15th Annual) · section: R&D as % of Revenue (capital-allocation section) — published 2024-09-01.
Why does this cite KBCM/Sapphire SaaS Survey 2024 (15th Annual)? Read the ontology methodology for the published vs editorial tier system, attribution rules, and dispute process.
Stage relevance
| Company stage | Priority |
|---|---|
| Series A | Core |
| Series B | Core |
| Series C+ | Core |
| Public | Core |
Suggested for stages: Series A, Series B, Series C+, Public.
Default owning functions
- R&D
- Finance
Machine-readable
- This KPI as JSON:
/api/ontology/product/rd_monthly_spend.json - All Product KPIs:
/api/ontology/product.json - Full catalog:
/api/ontology/index.json
R&D Efficiency
Ratio of net-new ARR generated in a period to R&D spend in the same period — answers "how much revenue does each R&D dollar produce?" Distinct from sales-efficiency metrics (Magic Number, CAC payback) which measure sales/marketing productivity. Common pitfall: R&D-driven ARR (new capabilities, expansion features) shows up on a 2–4 quarter lag after the spend — single-period ratios mis-state the relationship. Boards should look at trailing-twelve-month R&D efficiency, not month-over-month, and pair with `innovation_capacity_pct` to understand whether the spend is on growth bets or maintenance. — Product KPI, I'mBoard-authored (editorial tier).
Time to Capacity Limit
Months of system capacity remaining at the current growth rate before the platform requires major (not incremental) infrastructure investment — typically driven by the binding bottleneck (database, message bus, single-tenant compute ceiling, regional capacity, or compliance-driven re-architecture). Surfaces the "scale runway" alongside the financial runway. Common pitfall: a single number hides which bottleneck binds. Boards should require the bottleneck to be named ("database shard hot-spot binds at ~150K accounts at current growth, ~4 months out"), not just the headline months — a named bottleneck makes the investment decision concrete. — Product KPI, I'mBoard-authored (editorial tier).