AI-Native Finance Catalog: AI Companies ke liye Pricing, Forecasting, aur Financial Architecture
Agar aap is sab mein naye hain - yahan se start karein
Yeh lamba document hai. Isko use karne ke liye poora padhna zaroori nahin. Agar aap finance mein naye hain, ya early-stage AI company chala rahe hain, to "mujhe kya karna chahiye?" ka seedha jawab yeh hai.
Is week. Billing ke liye Stripe ya us jaisa tool set karein. Isay simple bookkeeping tool se connect karein: Pilot, Bench, Puzzle, Mercury Treasury, ya koi aisa tool jo basics automate karta ho. Ab se teen numbers track karein: revenue, gross margin (revenue minus compute aur usage-based vendor costs), aur cash runway months mein.
Is month. Agle 18 months ke liye simple spreadsheet banayein, har month ki ek row ke saath. Wohi teen numbers forward project karein. Har month ke pehle business day par update karein. Actuals ko forecast se compare karein. Jo gap niklega wahi batayega ke business asal mein kaise behave karta hai.
Is quarter. Jab teen months ka revenue data aa jaye to average gross margin dekhein. Agar 50% se neeche hai, unit economics shayad broken hain. Zyada AI-native businesses ko scale par survive karne ke liye 60%+ gross margin chahiye, aur SaaS norms 75-85% expect karte hain. 50% se neeche ka matlab compute costs, vendor pricing, ya pricing model ko investigate karein.
Is year. CFO hire na karein. Accounting team hire na karein. Enterprise FP&A software na khareedein. Audit na chalayein jab tak investor explicitly require na kare. Jo time bachta hai usay revenue grow karne par lagayein, kyun ke finance ka zyada hissa tab matter karta hai jab manage karne ke liye meaningful revenue ho.
Pehle 12 months ki prescription yahi hai: Stripe + bookkeeping tool + teen numbers + simple forecast spreadsheet. Baqi document us waqt ke liye hai jab yeh setup chota par jaye: revenue model complex ho, investors demanding hon, ya team itni bari ho jaye ke simple stack scale na kare.
Agar oopar wali prescription par wapas jane se pehle thora broad overview chahiye ho, to neeche Beginner ka 10-minute version wider map deta hai.
Is document ka beginner path
True beginner hain to is document ko linearly na parhein. Catalog founders, CFOs, controllers, aur investors ke liye bana hai; iska zyada hissa abhi aap ke liye nahin. In five sections ko is order mein parhein aur baqi sab skip karein jab tak actual revenue na ho:
- Agar aap is sab mein naye hain - yahan se start karein (oopar) - literal year-one prescription.
- Beginner ka 10-minute version (neeche) - four families, twelve approaches ek sentence mein.
- Approach 2 - Per-Call / Usage Pricing (Section A) - sab se common AI pricing model aur likely pehla model.
- Approach 7 - Compute COGS Accounting (Section B) - AI businesses mein gross margin ke bare mein har founder ko kya samajhna chahiye.
- Appendix A - Glossary (end par) - jab koi term unfamiliar ho to yahan dekhein.
Beginner reading path itna hi hai: five sections mein roughly 4,000 words. Executive summary, finance diagnostic, strategic fit matrix, baqi ten approaches, cross-cutting concepts, AI-era shifts, failures, aur anti-patterns skip kar sakte hain jab tak specific question na ho.
Jab meaningful revenue ho jaye, usually $1M+ ARR, tab document par wapas aayein aur baqi sections apni need ke hisaab se parhein.
Yeh document kahan fit hota hai
Yeh document The AI-Native Company series ke andar hai. The Agent Factory Thesis architecture define karti hai. The AI Worker Catalog batata hai kya build hota hai. The Sales Catalog aur The Marketing Catalog batate hain company product kaise sell karti hai aur demand kaise create karti hai. Finance Catalog batata hai company books kaise rakhti hai, products ki pricing kaise karti hai, future kaise forecast karti hai, aur funders ko report kaise karti hai.
Yeh operational sawal ka jawab hai: jab cost structure, pricing models, aur forecasting traditional SaaS se meaningfully different hon, to AI-native company ka financial side asal mein kaise chalaya jaye?
Isay standalone bhi parh sakte hain. Sales Catalog ke kuch cross-references skip karne se core argument lose nahin hota.
Is document ko kaise parhein
Yeh document tool hai, story nahin. Different readers isay different tareeqon se use karenge.
Agar aap finance mein naye hain. Oopar diya gaya beginner path follow karein. Pehli dafa poora catalog parhne ki koshish na karein.
Agar aap early-stage AI company chalanay wale founder hain. Neeche Finance Diagnostic aur Strategic Fit Matrix use karein taake pata chale kaunsi pricing architectures buyer aur stage se fit hoti hain. Section A ke relevant approaches parhein. Deeper accounting aur forecasting sections tab tak skip karein jab tak forecast karne layak revenue na ho.
Agar aap AI company mein CFO, controller, ya finance lead hain. Yeh document aap ke liye built hai. Top to bottom parhein. Approaches pricing se start hoti hain, phir accounting mechanics, forecasting, aur external reporting tak jati hain.
Agar aap investor ya board member hain. Section D ka Investor & Board Reporting aur end ke qareeb Common finance failures sab se directly relevant hain.
Jargon par note. Yeh document accounting, FP&A, aur SaaS finance ki vocabulary use karta hai. Specialized term pehli dafa usually nearby plain language mein explain hota hai. Appendix A: Glossary quick reference deta hai. Neeche Finance terms you must know first section fifteen important terms explain karta hai.
Professional advice par note. Yeh document strategic frameworks aur operational reference deta hai, professional accounting, tax, legal, ya financial advice nahin. ASC 606 revenue recognition, training costs capitalization, audit treatment, sales tax, aur corporate-structure decisions ke liye qualified professionals se guideance lein.
Confidence tagging par note. Benchmark claims kabhi kabhi tags ke saath aate hain: [Industry benchmark] broad practitioner consensus; [Emerging pattern] 2024-2026 ki AI-native companies mein dekha gaya pattern; [Author thesis] observed patterns se informed extrapolation. Untagged numbers bhi isi spectrum mein hain.
Beginner ka 10-minute version
Agar sirf ten minutes hain to yeh section parhein. Yeh batata hai ke AI-native companies finance kaise handle karti hain.
"AI-native finance" kya hai aur regular SaaS finance se kaise different hai?
AI-native finance un companies ke pricing, accounting, forecasting, aur reporting ka discipline hai jinke products foundation models, AI agents, ya compute-heavy AI workloads use karte hain. Difference teen jagahon par hai. Pehla, cost structure: traditional SaaS 75-85% gross margins tak jata hai kyun ke hosting cost revenue ke muqablay mein choti hoti hai; AI-native companies aksar 50-70% gross margin par hoti hain kyun ke compute meaningful cost hai. Doosra, pricing models: traditional SaaS per-seat subscription bechta hai; AI-native companies per-call, per-token, per-outcome, ya hybrid pricing use karti hain. Teesra, forecasting complexity: traditional SaaS stable unit costs assume kar sakta hai; AI-native forecasts ko foundation-model prices ki 30-60% annual fall, usage-driven ramp curves, aur different revenue-recognition structures model karne parte hain.
Finance approaches ki four families
- Pricing architectures (1-5). AI companies customers se kaise charge karti hain: per-seat, per-call, per-outcome, value-based, ya hybrid.
- Revenue & cost mechanics (6-8). Company earned aur spent money ko books mein kaise record karti hai: revenue recognition, compute COGS, cohort analysis with model-cost decay.
- Planning & capital allocation (9-11). Company forecast aur budget kaise banati hai: pilot economics, falling compute costs ke under forecasting, aur compute/people/marketing/runway ke beech capital split.
- External reporting (12). Company investors, board, aur auditors se kaise baat karti hai: investor metrics, board dashboards, aur audit-defensible disclosures.
Twelve approaches ek sentence mein
- Per-Seat Pricing. Har user ke liye fixed monthly fee; traditional SaaS se familiar, lekin variable compute-cost products ke liye aksar weak.
- Per-Call / Usage Pricing. API call, token, query, ya audio second ke hisaab se charge; AI infrastructure ka dominant model.
- Per-Outcome Pricing. Sirf defined result deliver hone par charge: resolved ticket, processed claim, booked meeting.
- Value-Based Pricing. Measured customer value ka percentage charge; strategic enterprise deals ke liye.
- Hybrid Pricing. Base subscription plus usage overage, ya subscription plus outcome bonus.
- Revenue Recognition for AI Contracts. ASC 606 rules jo decide karte hain revenue books par kab count hota hai.
- Compute COGS Accounting. Foundation-model calls, GPU rentals, aur infrastructure compute ko P&L par kaise treat karna hai.
- Cohort Analysis with Model-Cost Decay. Customer cohorts ko track karna jab foundation-model costs time ke saath girti hain.
- Pilot Economics & Contract Mechanics. Paid pilots, production contracts, aur enterprise AI deals ke multi-stage commercial structure ko model karna.
- Forecasting Under Falling Compute Costs. 12-24 month forecast banana jo annual compute price reductions explicitly model kare.
- Capital Allocation. Incremental dollars ko compute, people, marketing, aur runway mein split karna.
- Investor & Board Reporting. Metrics, dashboards, aur disclosures design karna jo AI-native investors aur boards expect karte hain.
Beginner difficulty per approach
- Easy: Per-Seat Pricing (1), Per-Call Pricing (2)
- Medium: Per-Outcome Pricing (3), Hybrid Pricing (5), Revenue Recognition (6), Compute COGS (7), Pilot Economics (9), Capital Allocation (11), Investor Reporting (12)
- Advanced: Value-Based Pricing (4), Cohort Analysis (8), Forecasting Under Falling Costs (10)
Finance terms you must know first
Finance unfamiliar ho to yeh fifteen terms sab se pehle samajh lein. Baqi document inhi par build karta hai.
Revenue. Customers se earned money. Income statement ki top line.
Bookings. Kisi period mein signed deals ki total contract value. Revenue se different: $1.2M one-year contract sign hone ke din $1.2M bookings hai, lekin revenue $100K/month recognize hota hai.
Recognized revenue. GAAP rules ke under kisi period mein income statement par aane wala contracted revenue.
ARR (Annual Recurring Revenue). Subscription customers ki annualized contract value. SaaS ka sab se tracked metric.
COGS (Cost of Goods Sold). Product deliver karne ke direct costs. AI-native companies mein foundation-model API costs, hosting, infrastructure, aur variable customer-success time include hota hai.
Gross margin. Revenue minus COGS, revenue ke percentage ke taur par. Traditional SaaS norms 75-85%; AI-native norms 50-70%.
NRR (Net Revenue Retention). Existing customers se retained recurring revenue including upsell. 100% se upar ka matlab same customer base revenue terms mein grow kar raha hai.
CAC (Customer Acquisition Cost). New customer acquire karne ka fully-loaded cost: sales, marketing, aur related functions.
LTV (Lifetime Value). Customer lifetime mein expected total gross-margin contribution.
LTV/CAC ratio. Lifetime value divided by acquisition cost. Healthy SaaS programs 3x se upar target karte hain.
CAC payback period. Customer ki gross-margin contribution se acquisition cost recover hone mein kitne months lagte hain.
Cash runway. Current burn rate par company kitne months operate kar sakti hai before cash khatam ho.
Burn rate. Har month company se net cash outflow.
Burn Multiple. Cash burned divided by net new ARR added. Lower better; AI-native early-stage ke liye under 2x healthy, mature SaaS ke liye under 1.5x.
Compute COGS. AI workloads chalane ka cost: foundation-model API calls, GPU inference, infrastructure compute. AI-native companies mein aksar revenue ka 20-60%.
ASC 606. US accounting standard jo revenue recognition govern karta hai. International equivalent IFRS 15.
Minimum financial metrics for AI-native companies
Agar sirf ten metrics track karne hain to yeh karein. Yeh minimum scorecard hai; Section E aur F full metric set dete hain.
| # | Metric | Formula | Kyun matter karta hai | Target |
|---|---|---|---|---|
| 1 | Revenue (recognized) | Period mein GAAP ke under earned revenue ka sum | Income statement ki top line | Month-over-month grow |
| 2 | ARR | Subscription contracts se annualized recurring revenue | Standard SaaS scale metric | Stage-dependent |
| 3 | Gross margin | (Revenue - COGS) / Revenue | Unit economics ka health signal | 50-70% AI-native, 75-85% mature SaaS |
| 4 | Compute as % of revenue | Compute COGS / Revenue | AI-specific cost ratio | Scaling stage par 20-35% |
| 5 | Cash on hand | Period end par liquid cash | Survival metric | Kam az kam 18 months runway |
| 6 | Monthly burn | Operating expenses - revenue collected | Cash drain | Stage-dependent |
| 7 | Cash runway | Cash on hand / Monthly burn | Survival funded kitni der hai | 18+ months |
| 8 | NRR | (Starting ARR + Expansion - Churn - Contraction) / Starting ARR | Existing customer health | >110% healthy, >130% strong |
| 9 | CAC payback period | CAC / (Monthly recurring revenue per customer x Gross margin) | Acquisition break-even time | <18 months |
| 10 | Burn Multiple | Net cash burned / Net new ARR added | Growth phase capital efficiency | <2x AI-native, <1.5x mature SaaS |
Cash aur runway weekly track karein; revenue, ARR, gross margin, compute %, NRR, aur burn monthly; CAC payback aur Burn Multiple quarterly. Bookkeeping tool se update karein; books se diverge karne wali spreadsheet maintain na karein.
Executive summary
AI-Native Finance Catalog 2026 aur us ke baad AI-native company ke financial side ke liye recipe book hai. Pricing, accounting, forecasting, aur reporting ke multiple tareeqe hain; right way buyer, stage, contract structure, aur investor expectations par depend karta hai. Yeh document twelve approaches ko four families mein organize karta hai aur batata hai kaunsa approach kis situation se fit hota hai.
Four families. Pricing architectures (Approaches 1-5) define karti hain company customers se kaise charge karti hai. Revenue & cost mechanics (Approaches 6-8) define karte hain company earned aur spent money ko books mein kaise laati hai. Planning & capital allocation (Approaches 9-11) future dekhne ka discipline hai. External reporting (Approach 12) investors, board, aur auditors ke saath communication define karta hai.
Five financial pillars. Har approach margin, cash, predictability, capital efficiency, aur audit defensibility ko optimize karne ki koshish karta hai. Strong financial architectures in pillars mein se teen ya zyada ko saath optimize karti hain; weak architectures ek pillar ko short-term improve kar ke long-term collapse create kar sakti hain.
Scope par note. Catalog primarily B2B AI-native companies ke liye hai, seed se Series C tak. Consumer AI companies aur late-stage public-company finance out of scope hain, lekin kuch approaches dono contexts mein apply hoti hain.
Maturity spectrum. Har approach Proven, Emerging, ya Speculative tag hota hai:
- Proven: at-scale companies isay chala rahi hain; playbooks aur benchmarks established hain.
- Emerging: AI-native companies 2026 mein isay chala rahi hain, lekin playbook abhi evolve ho raha hai.
- Speculative: buyer behavior ya practices abhi scale par exist nahin karti.
What this page is for
Yeh document teen kaam karta hai.
Chooser. Founder ya finance leader Strategic Fit Matrix, Finance Diagnostic, aur Approach Summary Table se apni stage, buyer, aur contract structure ke liye right architecture choose kar sakta hai.
Reference. Existing finance team deep sections se apni operation audit kar sakti hai: gross margin, cohort behavior, forecast accuracy, aur reporting mechanics compare kar ke.
Sequencing guide. Successful AI-native companies scale ke saath financial architecture evolve karti hain. Common Hybrid Models section common evolution paths map karta hai.
How to choose a financial architecture
Sab se clean predictor pricing complexity aur company stage ka intersection hai.
| Stage -> / Pricing complexity ↓ | Pre-revenue (Seed) | Early revenue ($1M-$10M ARR) | Scaling ($10M+ ARR) |
|---|---|---|---|
| Simple (per-seat ya single architecture) | Per-Seat (1) | Per-Seat (1), Per-Call (2) | - |
| Moderate (usage-based, single architecture) | Per-Call (2) | Per-Call (2), Per-Outcome (3) | Per-Call (2), Per-Outcome (3) |
| Complex (hybrid ya value-based) | - | Hybrid (5) | Hybrid (5), Value-Based (4) |
Sab se important cell complex x scaling hai: Hybrid Pricing aur Value-Based Pricing. Yeh highest revenue per customer aur defensible pricing power de sakti hain, lekin sophisticated finance, sales, aur customer-success operations mangti hain.
Finance diagnostic: eight questions
Financial architecture pick karne se pehle in eight dimensions par honestly score karein:
- Buyer type. Developer / API consumer -> Per-Call. SaaS operator -> Per-Seat ya Hybrid. Enterprise outcome buyer -> Per-Outcome ya Value-Based.
- Average deal size. <$10K/year -> Per-Seat ya Per-Call. $10K-$100K -> Per-Call ya Hybrid. $100K+ -> Per-Outcome, Value-Based, ya Hybrid.
- Cost structure variability. Compute cost small/stable -> Per-Seat. Usage ke saath cost vary ho -> Per-Call. High value-per-outcome ho -> Per-Outcome possible.
- Sales motion. Self-serve PLG -> Per-Call ya Per-Seat. Vendor-led mid-market -> Per-Seat, Per-Call, ya Hybrid. Enterprise field -> Per-Outcome, Value-Based, ya Hybrid.
- Customer technical sophistication. High -> Per-Call works. Low -> Per-Seat ya Hybrid, kyun ke predictable bills chahiye.
- Contract length. Monthly self-serve -> Per-Call ya Per-Seat. Annual SaaS -> any architecture. Multi-year enterprise -> Hybrid ya Value-Based.
- Forecast accuracy required. Tight -> Per-Seat ya Hybrid. Loose -> Per-Call ya Per-Outcome.
- Internal finance maturity. Founder spreadsheet mein books rakhta hai -> Per-Seat ya Per-Call. Controller in place -> Per-Outcome possible. Full finance team -> Value-Based aur complex Hybrid feasible.
Diagnostic correct architecture nahin batata; yeh batata hai kaunsi architectures aap ki starting position se available hain.
Approach summary table
| # | Approach | Maturity | Best for | Main strength | Main risk |
|---|---|---|---|---|---|
| 1 | Per-Seat Pricing | Proven | Predictable-usage SaaS | Forecast simple | Price cost se disconnect |
| 2 | Per-Call / Usage Pricing | Proven | Developer-buyer infrastructure | Price cost ke saath align | Customer bill anxiety |
| 3 | Per-Outcome Pricing | Emerging | Defined-result use cases | Maximum value capture | Outcome attribution complex |
| 4 | Value-Based Pricing | Emerging | Strategic enterprise deals | Premium pricing | Contracting maturity chahiye |
| 5 | Hybrid Pricing | Proven | Mid-market aur enterprise scale | Predictability + capture balance | Communicate karna complex |
| 6 | Revenue Recognition | Proven | Revenue wali any company | Audit defensibility | ASC 606 complexity |
| 7 | Compute COGS Accounting | Proven | Any AI-native company | Margin clarity | Misclassification risk |
| 8 | Cohort Analysis with Model-Cost Decay | Emerging | $5M+ ARR companies | Unit economics ki truth | Data discipline chahiye |
| 9 | Pilot Economics & Contract Mechanics | Proven | Enterprise sales motions | Pilot-to-production conversion | Premature production accounting |
| 10 | Forecasting Under Falling Compute Costs | Emerging | Usage models | Realistic margin trajectory | Decay over-optimism |
| 11 | Capital Allocation | Proven | Any post-Series A | Strategic spend discipline | Compute over-investment |
| 12 | Investor & Board Reporting | Proven | Any post-Series A | Stakeholder alignment | Vanity metrics |
Which approach should I run?
Decision flowchart architecture choice narrow karta hai.
Four key questions: buyer developer hai? average deal size $100K se upar hai? predictable revenue chahiye? finance team kitni mature hai? Answers architecture ko simple se complex tak guide karte hain.
The financial maturity curve
Har AI-native company financial maturity ke teen stages se guzarti hai. Stage 3 architecture stage 1 par chalana founders ke liye money waste karne ka common tareeqa hai.
Stage 1 - Pre-revenue (Seed-stage). Product hai lekin revenue limited. Finance work minimal: burn track, runway manage, basic taxes, Series A diligence ke liye Quality of Earnings review ki readiness. Usually Per-Seat (1) ya Per-Call (2). Finance team: founder + Pilot/Bench/Puzzle.
Stage 2 - Early revenue ($1M-$10M ARR). Product-market fit signals aur meaningful customer count. Monthly close, board reporting, basic forecasting, first cohort analyses start. Finance team: controller (full-time ya fractional), bookkeeper, founder major decisions mein involved.
Stage 3 - Scaling ($10M+ ARR). Series B ke liye preparing ya completed. Full FP&A, audit preparation, complex contract accounting, investor/board reporting. Hybrid Pricing (5), Value-Based Pricing (4), Cohort Analysis (8), aur Capital Allocation (11) board-level topics ban jate hain.
Founders ke liye implication: financial architecture one-time decision nahin. Scale tak pohanchne se pehle yeh usually at least do dafa evolve hoti hai.
Maturity legend
- Proven. Approach scale par operate ho raha hai, playbooks aur benchmarks ke saath.
- Emerging. 2026 mein AI-native companies use kar rahi hain, lekin canonical playbook still evolving hai.
- Speculative. Buyer behavior ya practice abhi scale par exist nahin karti.
A. Pricing architectures
Company customers se kaise charge karti hai. Pricing architecture single most consequential financial decision hai: revenue recognition, sales compensation, customer success, forecast complexity, aur gross margin sab is se affect hote hain.
Approach 1 - Per-Seat Pricing
Maturity: Proven. Beginner difficulty: Easy.
In Plain English. Per-Seat Pricing woh SaaS model hai jo 2010s mein sab ne seekha: customer per user per month fixed fee deta hai. Buyer ka bill predictable, seller ka revenue predictable, accounting straightforward. AI-heavy products ke liye problem yeh hai ke compute cost seat count se nahin, usage se scale hoti hai.
Core idea. Har user ke liye predictable fee charge karein, accept karte hue ke heavy users negative unit economics create kar sakte hain.
When to use it. Jab product AI-augmented ho, AI-defined nahin. Jab buyer executive ho jise predictable budget chahiye. Jab compute cost per seat subscription revenue ke 10-15% se kam ho.
Mechanism. Per-seat buyer aur seller dono ko predictability deta hai. Structural problem price aur cost ka disconnect hai: foundation-model APIs per token, audio second, ya image charge karti hain, jab ke customer per seat pay kar raha hota hai. Heavy users most expensive hote hain lekin light users ke barabar pay karte hain.
Fictional walk-through. MeetingMind $30/seat/month par sold hai. 100 seats = $36K/year. Agar 20 users heavy hain to compute around $7K/year ho sakta hai, margin comfortable. Agar heavy users 50% ho jayein to compute $15K+ ho jata hai aur margin 60% tak gir sakta hai.
Example. AI-augmented productivity tools aksar per-seat pricing ke saath usage limits rakhte hain.
Primary risk. Heavy users par negative unit economics. Mitigation: compute-per-seat cohort monitor karein, caps ya overages introduce karein, aur Hybrid Pricing natural evolution samjhein.
First move. Current customer base mein average compute cost per seat calculate karein. Agar seat revenue ka 15% se upar hai, Hybrid Pricing transition plan karein.
Approach 2 - Per-Call / Usage Pricing
Maturity: Proven. Beginner difficulty: Easy.
In Plain English. Per-Call Pricing AI infrastructure ka standard hai. Customer API call, token, audio second, generated image, ya query ke hisaab se pay karta hai. Revenue usage ke saath scale hota hai; cost bhi usage ke saath scale hoti hai.
Core idea. Price ko directly usage aur cost ke saath align karein.
When to use it. Developer ya technical buyer ho, product genuinely usage-variable ho, aur team usage instrumentation, billing, dashboards, alerts, aur customer-success bill management mein invest kar sakti ho.
Mechanism. Har call cost se upar priced hoti hai, is liye gross margin structurally protected hota hai. Downside bill shock hai: usage spike se bill spike hota hai, aur customers budget se upar ja kar churn kar sakte hain.
Fictional walk-through. TextAI token pricing karta hai. Customer experiments par $200/month se start karta hai, production mein ramp kar ke $150K/month tak jata hai. CFO bills par complain karta hai, lekin TextAI ka gross margin steady 65% rehta hai.
Example. OpenAI, Anthropic, Cohere, Mistral, ElevenLabs, Replicate, Together AI, Fireworks AI.
Primary risk. Bill shock aur churn. Mitigation: usage dashboards, budget alerts, capacity-planning conversations, aur hard caps.
Secondary risk. Forecast unpredictability. Mitigation: cohort-based usage forecast aur lead indicators track karein.
First move. Agar product usage-variable hai aur buyer technical hai, start se Per-Call Pricing ship karein. 60%+ gross margin target karein aur first customer se pehle usage dashboard banayein.
Approach 3 - Per-Outcome Pricing
Maturity: Emerging. Beginner difficulty: Medium.
In Plain English. Per-Outcome Pricing mein customer sirf defined result par pay karta hai: resolved support ticket, processed claim, booked meeting, ya successfully completed agent task. AI fail ho to customer pay nahin karta.
Core idea. Delivered outcome par charge karein, price ko seller ke software cost ke bajaye customer ke labor cost se anchor karein.
When to use it. Jab use case clear, measurable, attributable outcome rakhta ho; customer ka alternative humans hire karna ho; aur company outcome-attribution infrastructure mein invest karne ko ready ho.
Mechanism. Seller customer ke labor budget ka fraction capture karta hai. Customer-support ticket human all-in $5 cost karta hai to AI outcome price ceiling $1-$3 per resolved ticket ho sakti hai. Seller ka compute cost usually $0.20-$0.80 per outcome floor set karta hai. Beech mein price live karta hai.
Technical foundation outcome attribution hai: har billable outcome ke liye verifiable record chahiye ke AI ne kya kiya, kya process kiya, aur result kaise confirm hua. Accounting mein revenue outcomes deliver hone par recognize hota hai, contract sign hone par nahin.
Fictional walk-through. TicketBot har self-resolved support ticket par $0.50 charge karta hai. 50,000 tickets/month ka customer $25K monthly bill deta hai sirf jab TicketBot tickets resolve kare. 30% resolution par bill $7.5K hota hai.
Example. Support automation, claims processing, SDR meeting booking, aur document-processing workflows.
Primary risk. Outcome disputes. Mitigation: audit-grade telemetry, clear success definition, aur customer-visible logs.
Secondary risk. Revenue recognition lag. Mitigation: bookings, deferred revenue, aur recognized revenue separately report karein.
First move. One workflow choose karein jahan outcome clear ho. Outcome definition, attribution logs, aur customer dispute process document karein before pricing launch.
Approach 4 - Value-Based Pricing
Maturity: Emerging. Beginner difficulty: Advanced.
In Plain English. Value-Based Pricing customer ko create hone wali measured value ka percentage charge karta hai. Yeh premium enterprise architecture hai: price seat, call, ya outcome se nahin, customer ke business impact se tied hota hai.
Core idea. Customer value measure karein, us value ka defensible share capture karein.
When to use it. Jab buyer sophisticated enterprise ho, value measurement possible ho, baseline agreed ho, aur executive sponsor contract mein value-share accept kare.
Mechanism. Pehle baseline period measure hota hai. Phir AI deployment ke baad value gap calculate hota hai: cost avoided, revenue generated, risk reduced, ya productivity improved. Vendor 5-25% value share capture kar sakta hai. Contract legal, finance, aur audit complexity high hoti hai.
Fictional walk-through. FinOpsAI enterprise ke cloud spend ko $20M/year se $16M/year kar deta hai. Contract 15% savings share rakhta hai, to vendor $600K/year charge karta hai.
Example. High-value enterprise workflows jahan value measurement credible ho: procurement savings, fraud reduction, claims leakage, revenue recovery.
Primary risk. Value measurement dispute. Mitigation: baseline, formula, audit rights, aur data sources contract mein lock karein.
Secondary risk. Recognition conservatism. Mitigation: revenue accountant se policy pre-review karwayein.
First move. First architecture ke taur par Value-Based pursue na karein. Pehle Per-Call, Per-Outcome, ya Hybrid se operational maturity build karein.
Approach 5 - Hybrid Pricing
Maturity: Proven. Beginner difficulty: Medium.
In Plain English. Hybrid Pricing do ya zyada architectures ko ek contract mein combine karta hai. Common pattern: base subscription plus usage overage above included quota. Customer normal usage ke liye predictable budget leta hai aur heavy usage ke liye incremental pay karta hai.
Core idea. Predictability, cost alignment, aur value capture ko combine karein.
When to use it. Jab pure per-seat ya per-call break down kar raha ho: heavy users margin compress kar rahe hon, light users churn risk create kar rahe hon, ya enterprise buyers sophisticated contract demand kar rahe hon.
Mechanism. Common structure "Per-Seat plus Usage Overage" hai. Alternatives: Platform Fee plus Usage, Subscription plus Outcome Bonus, ya Tiered Subscription. Accounting mein subscription revenue ratably recognize hota hai; usage revenue usage ke saath. ASC 606 multi-component allocation demand karta hai.
Fictional walk-through. AgentPlatform $5,000/month platform fee leta hai including 1M agent calls, plus $0.005 per extra call. Customer $60K base annual contract sign karta hai, usage ramp kar ke year one revenue $240K tak le jata hai.
Example. GitHub Copilot Business/Enterprise tiers, Cursor enterprise plans, aur mature enterprise AI vendors.
Primary risk. Customers pricing samajh nahin pate. Mitigation: projection dashboards, quarterly true-up, aur customer-success education.
Secondary risk. Revenue recognition complexity. Mitigation: AI contracts samajhne wala revenue accountant involve karein.
First move. Current pricing plus one missing component design karein: overage ya subscription floor. Day one par six-component contract design na karein.
B. Revenue & cost mechanics
Finance ka technical kaam: customer activity ko auditable books mein lana, compute costs classify karna, aur cohort discipline maintain karna. Pricing imperfect ho to company kuch saal chal sakti hai; revenue recognition ya COGS classification first audit par broken ho to trust damage hota hai.
Accounting aur tax advice par note. Yeh section ASC 606, COGS classification, capitalization, deferred revenue, aur audit defensibility discuss karta hai. Yeh professional advice nahin. Non-subscription contract sign karne se pehle AI-native experience wale CPA se baat karein.
Approach 6 - Revenue Recognition for AI Contracts
Maturity: Proven. Beginner difficulty: Medium.
In Plain English. Revenue recognition ka sawal hai ke revenue books par kab count hota hai. $1.2M one-year contract sign ho to day one par $1.2M revenue book hota hai, har month $100K, ya kuch aur? US mein ASC 606 aur international mein IFRS 15 isko govern karte hain.
Core idea. ASC 606 ke five steps AI contracts par apply karein: contract identify, performance obligations identify, transaction price determine, price allocate, aur obligations satisfy hone par revenue recognize.
When to use it. Jab company ke paas koi bhi contracted revenue ho. Per-Seat simple hai; Value-Based complex hai; framework universal hai.
Mechanism. AI contracts teen cheezon ki wajah se complex hote hain: variable consideration, multiple performance obligations, aur outcome dependency. Is wajah se bookings aur recognized revenue diverge kar sakte hain.
Fictional walk-through. OutcomeAI Q1 mein $4M annual outcome-based contracts sign karta hai. End of Q1 tak sirf 200K tickets resolve hote hain, $2/ticket par $400K recognized revenue. Bookings $4M, recognized revenue $400K, deferred revenue $3.6M.
Example. Non-subscription AI companies ko yeh complexity face hoti hai. Pure subscription companies simple recognition rakhti hain lekin ASC 606 compliance still chahiye.
Primary risk. Aggressive recognition jo auditors restate kar dein. Mitigation: first non-subscription contract se pehle experienced revenue accountant involve karein.
Secondary risk. Too-conservative recognition jo growth hide kare. Mitigation: bookings, deferred revenue, aur recognized revenue consistent separate report karein.
First move. ASC 606 brief lein, one-page revenue-recognition policy memo banayein, aur first audit cycle se pehle accountant se review karwayein.
Approach 7 - Compute COGS Accounting
Maturity: Proven. Beginner difficulty: Medium.
In Plain English. Compute COGS Accounting decide karta hai ke foundation-model API calls, GPU rentals, inference infrastructure, fine-tuning compute, aur embeddings P&L par kaise flow karte hain. AI-native companies mein compute often revenue ka 30-60% hota hai, is liye COGS line critical hai.
Core idea. Compute costs ko COGS aur operating expenses ke beech correctly classify karein, consistent policy ke saath.
When to use it. Always, compute cost start hone ke moment se.
Mechanism. Compute costs teen categories mein fall karte hain:
- Direct production compute: customer requests fulfill karne ka cost. Yeh COGS hai.
- Product-development compute: training, fine-tuning, evaluation, experiments. Usually R&D expense, kabhi capitalization possible.
- Internal-use compute: employees ke AI tools. Operating expense.
Prepaid compute commitments balance sheet par asset bante hain aur consumption ke saath COGS mein expense hote hain.
Fictional walk-through. AgentCo $5M ARR par $2M compute spend karta hai: $1.5M production inference, $300K training/evaluation, $200K internal tooling. Correct classification: $1.5M COGS, $300K R&D, $200K opex. Wrong classification gross margin distort karti hai.
Example. Har AI-native company ko compute-COGS policy develop karni padti hai.
Primary risk. Inconsistent classification jo margin trends mask kare. Mitigation: formal policy, consistent application, auditor review.
Secondary risk. Development compute aggressively capitalize kar ke near-term earnings inflate karna. Mitigation: capitalization conservative rakhein.
First move. Har compute cost list karein, production/product-development/internal-use mein classify karein, aur one-page policy memo banayein.
Approach 8 - Cohort Analysis with Model-Cost Decay
Maturity: Emerging. Beginner difficulty: Advanced.
In Plain English. Cohort analysis same period mein acquired customers ko time ke saath track karta hai: retention, revenue, gross margin. Traditional SaaS stable unit costs assume karta hai. AI-native companies mein foundation-model prices 30-60% per year gir sakti hain, is liye same cohort time ke saath more profitable ho sakta hai without behavior change.
Core idea. Cohort behavior aur falling model costs ka contribution separate karein taake true unit economics samajh aaye.
When to use it. Jab 12-24 months customer data ho, compute revenue ka 20%+ ho, aur finance team per-cohort gross margin track kar sake.
Mechanism. Do effects separate hote hain: cohort behavior effect (retain, expand, churn) aur model-cost decay effect (serve karne ka cost acquisition ke baad kitna gira). Synthetic cost baseline maintain karein: original acquisition-period prices par aaj ka cost kya hota.
Fictional walk-through. Sigma ka 2024 cohort 55% gross margin par acquired tha; 2026 mein 72% par operate kar raha hai. Analysis batata hai 7 points customer behavior se aur 10 points model-cost decay se aaye.
Example. Larger AI-native vendors internally yeh analysis chala rahe hain; published case studies abhi limited hain.
Primary risk. Margin improvement ko galat tor par cohort behavior samajhna jab main driver model-cost decay ho. Mitigation: synthetic-cost baseline maintain karein.
First move. Ek large cohort pick karein. Acquisition aur today gross margin calculate karein. Today gross margin ko acquisition-period compute prices par recalculate karein.
C. Planning & capital allocation
AI-native company future ko kaise model karti hai, capital kaise allocate karti hai, aur contracts kaise structure karti hai. Yeh approaches fundraising, hiring sprints, infrastructure commitments, aur pricing changes ke waqt sab se consequential hoti hain.
Approach 9 - Pilot Economics & Contract Mechanics
Maturity: Proven. Beginner difficulty: Medium.
In Plain English. Enterprise AI deals aksar full production contract se start nahin hotay. Pehle paid pilot hota hai: 3-6 months, production size ka fraction, proof ke liye. Pilot economics production economics se different hote hain: delivery cost higher, contract smaller, recognition timing different.
Core idea. Paid pilots ko production contracts se separate revenue category treat karein.
When to use it. Enterprise sales motion mein, especially $50K+ average deal size aur 60+ day sales cycle.
Mechanism. Pilot usually 10-25% projected production contract, success criteria, high customer-success engagement, aur end par conversion decision rakhta hai. Pilot revenue fixed-fee deliverables ke according recognize hota hai. Pilot delivery direct cost high ho sakta hai. Conversion 50-75% mature enterprise AI companies mein common pattern hai.
Fictional walk-through. MedAI 90-day $50K pilots sign karta hai, successful hone par $400K/year production contract. 12 pilots = $600K pilot revenue; 8 convert = $3.2M production ARR.
Example. Glean, Harvey, Sierra, Cresta, Writer jaise vendors pilot-first motions use karte hain.
Primary risk. Pilot revenue ko ARR mein include kar ke investor trust lose karna. Mitigation: pilot revenue ARR se separate report karein.
First move. Company structure mein pilot define karein: size, duration, conversion criteria. Books mein separate revenue category banayein.
Approach 10 - Forecasting Under Falling Compute Costs
Maturity: Emerging. Beginner difficulty: Advanced.
In Plain English. AI-native 12-24 month forecast mein foundation-model prices ki fall explicitly model karni hoti hai. Constant compute prices assume karna out-quarters ki margin understate karta hai aur runway projections distort kar sakta hai.
Core idea. Forecast ko two explicit layers mein banayein: customer-revenue model aur compute-price model. Dono combine kar ke margin projection banayein.
When to use it. Jab compute spend revenue ka 20%+ ho, forecast period 12 months se longer ho, ya fundraising / large hiring / infrastructure commitments imminent hon.
Mechanism. Traditional forecast revenue layer aur cost layer rakhta hai. AI-native forecast compute-price layer add karta hai. Observed decay rates 30-60% per year use ho sakte hain, lekin scenarios zaroor hon: aggressive, base, conservative.
Fictional walk-through. GenStudio $8M ARR aur $3M compute spend ke saath Series B forecast bana raha hai. Flat compute assumption margin 62.5% rakhti hai. Compute-price-decay layer add karne se future margin materially different dikhti hai, lekin model ko usage growth ke saath refine karna parta hai.
Example. Sophisticated AI-native companies Series B aur beyond ke liye compute-price decay explicitly model karti hain.
Primary risk. Decay rate par over-optimism. Mitigation: multiple scenarios; runway planning conservative case se.
First move. Last six quarters ka compute spend as % of revenue calculate karein. Provider price changes document karein. 30%/year base decay se forecast karein aur sensitivity run karein.
Approach 11 - Capital Allocation
Maturity: Proven. Beginner difficulty: Medium.
In Plain English. Capital Allocation decide karta hai incremental dollars kahan jayen: compute, engineers, salespeople, marketing, ya runway. AI-native companies mein compute spend curve is decision ko traditional SaaS se zyada intense banata hai.
Core idea. Har incremental dollar ko compute, people, customer acquisition, aur runway ke beech explicit strategic choice samjhein.
When to use it. Series A onward, especially fundraises, large customer payments, ya M&A ke waqt.
Mechanism. Four demands compete karti hain: compute, people, customer acquisition, runway. Burn Multiple core metric hai: cash burned / net new ARR. AI-specific question yeh hai ke compute efficiency mein engineering invest karein ya product scaling mein.
Fictional walk-through. FlexAI Series B par $50M fresh capital raise karta hai. Standard SaaS allocation $20M people, $15M acquisition, $10M runway, $5M compute ho sakta hai. AI-native-aware plan $8M compute-efficiency engineering ko allocate karta hai kyun ke future $100M revenue base par 30% margin improvement huge payoff hai.
Example. Series B ke baad AI-native companies compute-efficiency engineering ko capital plan mein explicit weigh karti hain.
Primary risk. Compute over-investment. Mitigation: demonstrated demand ke saath capacity allocate karein.
Secondary risk. Compute-efficiency under-investment. Mitigation: quarterly efficiency review aur explicit engineering allocation.
First move. One-page capital-allocation framework banayein aur quarterly review karein.
D. External reporting
Company investors, board, aur auditors se kaise baat karti hai: metrics, dashboards, disclosures. AI-native companies traditional SaaS norms se meaningfully different cheezen report karti hain.
Approach 12 - Investor & Board Reporting
Maturity: Proven. Beginner difficulty: Medium.
In Plain English. Investor & Board Reporting company ki financial state ko metrics, dashboards, aur narrative mein distill karta hai. Traditional SaaS metrics still apply: ARR, NRR, gross margin, CAC payback, Burn Multiple, Magic Number. AI-native companies ko in ke saath AI-specific metrics bhi add karne hain.
Core idea. Canonical SaaS metrics report karein, plus AI-specific metrics jo model-cost decay, compute concentration, outcome attribution, pilot conversion, aur recognized revenue gaps capture karte hain.
When to use it. Series A onward. Pre-revenue companies burn aur runway basics se start kar sakti hain.
Mechanism. Complete report teen tiers rakhta hai:
- Tier 1: ARR, NRR, GRR, gross margin, contribution margin, CAC payback, Burn Multiple, cash runway.
- Tier 2: compute as % of revenue, cohort gross margin trend, pilot-to-production conversion, outcome attribution accuracy, bookings vs recognized revenue, model-cost-decay benefit.
- Tier 3: compute concentration risk, forecast accuracy, capital allocation breakdown.
Fictional walk-through. GrowthAI Series B board report mein ARR $25M, NRR 130%, gross margin 65%, Burn Multiple 1.4x, runway 24 months, compute 28% of revenue, pilot-to-production 70%, aur forecast accuracy +/-8% report karta hai.
Example. Series B aur beyond ki sophisticated AI-native companies Tier 2 aur Tier 3 metrics include karti hain.
Primary risk. Vanity metrics over substance. Mitigation: cash, recognized revenue, aur gross margin first; bookings aur pipeline context ke saath.
First move. Last board report ke metrics list karein. Tier 1/2/3 se compare karein aur two-three additions identify karein.
E. Metrics & KPI framework
Previous sections batate hain AI-native finance kya karta hai: price, account, plan, report. Yeh section batata hai finance kya measure karta hai: specific metrics aur KPIs jo AI-native company ki health determine karte hain.
The metrics hierarchy
Har AI-native company ki financial reality four-layer hierarchy se nikalti hai:
Layer 1 - AI Worker operational metrics. AI khud kaise perform kar raha hai: outcomes, accuracy, escalation, throughput.
Layer 2 - Unit economics. Per-customer ya per-outcome profitability: contribution margin, gross margin per call, LTV, CAC, LTV/CAC.
Layer 3 - Company-level financial metrics. ARR, NRR, gross margin, contribution margin, burn, runway.
Layer 4 - Investor and capital-efficiency metrics. Burn Multiple, Magic Number, Rule of 40, ARR per employee, capital efficiency.
Finance teams jo sirf Layer 4 report karti hain asal drivers se blind rehti hain. Diagnostic information Layer 1 aur 2 mein hoti hai.
AI Worker operational KPIs
Six core AI Worker operational metrics:
1. Outcome rate. Attempts ka percentage jo successful outcome produce karta hai.
Outcome rate = Successful outcomes / Total attempts
Healthy ranges worker type ke hisaab se change hoti hain: support 60-85%, sales outreach 2-15%, code generation 30-70%.
2. Quality. Human ya auditor-rated outcome quality.
Quality = Average rated score (1–5 or 1–10 scale) across audited outcomes
Outcome rate aur quality dono saath dekhna zaroori hai.
3. Throughput. Outcomes per unit time.
Throughput = Outcomes / Time period
Automation leverage = AI throughput / Human throughput
Structured tasks mein AI Workers 5-20x human throughput dikha sakte hain.
4. Reliability. Uptime, error rate, aur similar inputs par behavior consistency.
Reliability = (Uptime %) × (1 − Error rate) × (Behavioral consistency score)
5. Cost per outcome. Ek outcome produce karne ka fully-loaded cost.
Cost per outcome = (Compute cost + Infrastructure cost + Allocated overhead) / Total outcomes produced
6. Cost-per-outcome trend. Time ke saath cost per outcome ka change.
Cost-per-outcome trend = (Cost per outcome this period − Cost per outcome prior period) / Cost per outcome prior period
Healthy AI Worker cost-per-outcome 20-40% per year decay dikha sakta hai.
Per-architecture financial KPIs
Har pricing architecture ke apne KPIs hain.
Per-Seat Pricing KPIs.
- Seats sold, seats churned, net seats added
- Seat utilization rate
- ARPU aur ARPA
- Compute cost per seat
- Compute-cost-per-seat distribution
Seat utilization rate = Active users / Paid seats
ARPU = Total revenue / Active users
Compute cost per seat = Total compute cost / Paid seats
Per-Call / Usage Pricing KPIs.
- Active customers
- Calls per active customer
- Revenue per call
- Gross margin per call
- Customer concentration
- Usage growth rate
- Bill-shock churn rate
Calls per active customer = Total billable calls / Active customers
Gross margin per call = (Revenue per call − Cost per call) / Revenue per call
Customer concentration (top 5) = Revenue from top 5 customers / Total revenue
Per-Outcome Pricing KPIs.
- Outcomes delivered per period
- Outcome attribution accuracy
- Outcome dispute rate
- Average revenue per outcome
- Cost per outcome
- Contribution margin per outcome
- Customer outcome consumption growth rate
Contribution margin per outcome = (Revenue per outcome − Variable costs per outcome) / Revenue per outcome
Outcome attribution accuracy = Outcomes with audit-grade telemetry / Total outcomes billed
Value-Based Pricing KPIs.
- Baseline measurement period results
- Measured value vs baseline
- Value-share capture rate, usually 5-25%
- Audit completion rate
- Variable consideration recognition rate
- Customer renewal rate at contract end
Hybrid Pricing KPIs.
- Subscription-vs-usage revenue split
- Overage rate
- Average overage revenue per overage customer
- Conversion to higher tier
- Bill predictability score
Stage-by-stage metric priorities
Pre-revenue (Seed). Top 3: cash runway, monthly burn, lead indicators. ARR, NRR, gross margin, CAC abhi meaningful nahin.
Early revenue ($1M-$5M ARR). Top 5: ARR, gross margin with compute-cost line, cash runway, NRR, CAC payback.
Mid stage ($5M-$25M ARR). Above plus Burn Multiple, contribution margin, pilot-to-production conversion, compute as % of revenue, cohort analysis.
Scaling ($25M+ ARR). Full Tier 1, Tier 2, Tier 3. Cadence decide karein: weekly, monthly, quarterly, annually.
Common mistake: Series A scale par Series B metrics report karna. Early company ke liye runway, burn, aur customer count enough hotay hain.
AI-specific operational efficiency KPIs
Engineering aur finance ko yeh metrics saath track karne chahiye:
Cost per token (input vs output). Foundation-model API ka unit cost, input aur output alag track karein.
Inference cost per query.
Inference cost per query = (Foundation-model API cost + Supporting compute cost) / Total queries served
Cache hit rate. Requests ka percentage jo cache se serve hota hai.
Batch processing efficiency. Batched vs real-time cost per outcome.
Model utilization rate. Self-hosted infrastructure ke liye GPU utilization.
Prompt token efficiency. Input token se generated value.
Time-to-first-token / time-to-completion. Customer experience aur human alternative se competition ka signal.
Capital efficiency metrics beyond Burn Multiple
ARR per employee.
ARR per employee = Total ARR / Total FTEs
Gross profit per employee.
Gross profit per employee = (Total ARR × Gross margin) / Total FTEs
R&D as percentage of revenue. Growth phases mein AI-native norms 35-55%.
S&M as percentage of new ARR. Sales and marketing spend / net new ARR.
G&A as percentage of revenue. Mature SaaS norms 10-15%; 20% se upar bloat signal ho sakta hai.
Rule of 40.
Rule of 40 = Annual revenue growth % + EBITDA margin %
Rule of 50/60 for fast-growing AI-native companies. Kuch investors hypergrowth AI-native companies ke liye higher threshold accept karte hain.
Capital efficiency ratio.
Capital efficiency ratio = Total ARR / Total capital raised
Worked example: AgentCo at $10M ARR
Company profile. AgentCo AI customer-support automation company hai. Pricing hybrid hai: $5,000/month subscription per customer including 50,000 resolved tickets, plus $0.50 per ticket above quota. 100 customers, average $100K ACV, 50 employees, Series B ki preparation.
Annual P&L.
| Line item | Amount | % of revenue |
|---|---|---|
| Bookings | $14M | 140% |
| Revenue (recognized GAAP) | $10M | 100% |
| Compute | $2.5M | 25% |
| Hosting & infrastructure | $400K | 4% |
| Customer-success allocation | $600K | 6% |
| Total COGS | $3.5M | 35% |
| Gross profit | $6.5M | 65% |
| R&D | $4M | 40% |
| Sales & Marketing | $3.5M | 35% |
| G&A | $2M | 20% |
| Operating loss | ($3M) | (30%) |
| Cash burn | ($2.5M) | (25%) |
| Cash on hand | $25M | - |
| Runway | 10 years at current burn | - |
Layer 1 - AI Worker operational metrics.
| Metric | Value | Healthy? |
|---|---|---|
| Outcome rate | 78% | Yes |
| Quality | 4.4 / 5 | Yes |
| Throughput | 120/hr | Yes |
| Reliability | 95.5% | Yes |
| Cost per outcome | $0.42 | Yes |
| Cost-per-outcome trend | -28% YoY | Yes |
Layer 2 - Unit economics.
| Metric | Value | Healthy? |
|---|---|---|
| ACV | $100K | - |
| CAC | $50K | - |
| LTV | $500K | - |
| LTV/CAC | 10x | Excellent |
| CAC payback | 14 months | Healthy |
| Contribution margin per ticket | 16% | Tight |
| Contribution margin per customer | 71% | Healthy |
Layer 3 - Company-level financial.
| Metric | Value | Healthy? |
|---|---|---|
| ARR | $10M | - |
| Bookings | $14M | Healthy growth |
| NRR | 128% | Strong |
| GRR | 92% | Healthy |
| Gross margin | 65% | Healthy |
| Compute as % of revenue | 25% | Healthy |
| Cohort gross margin trend | +3 points/quarter | Strong |
| Compute concentration | 75% one provider | Risk |
Layer 4 - Capital efficiency & investor metrics.
| Metric | Value | Healthy? |
|---|---|---|
| Burn Multiple | 0.7x | Excellent |
| Magic Number | 1.0 | Healthy |
| ARR per employee | $200K | Acceptable |
| Gross profit per employee | $130K | Acceptable |
| R&D as % revenue | 40% | Appropriate |
| S&M as % new ARR | 100% | Healthy |
| G&A as % revenue | 20% | Review |
| Rule of 40 | 10% | Below target |
| Capital efficiency ratio | 0.33x | Typical early-stage |
Dashboard batata hai company healthy hai lekin teen action items hain: G&A discipline, compute concentration mitigation, aur Rule of 40 improvement.
F. AI Worker reference and benchmarks
Section E framework deta hai; Section F reference layer hai: worker-type KPI cards, consolidated benchmarks, diagnostic playbooks, dashboard templates, compute economics deep-dive, aur operational health metrics.
Per-worker-type KPI cards
Neeche 12 common AI Worker categories ke cards hain. Ranges starting templates hain; material decisions se pehle apne data se validate karein.
1. Customer Support AI Worker
Use cases: inbound support triage, automated responses, common-query deflection, escalation routing. Typical pricing: Per-Outcome ya Hybrid. Operational KPIs: resolution 60-85%, CSAT 4.0-4.5/5, mean time 30 sec-5 min, false-resolution <5%, escalation accuracy >90%, hallucination <1%. Financial KPIs: revenue $0.50-$3 per resolved ticket, cost $0.20-$0.80, contribution margin 50-75%, LTV/CAC 5-25x.
2. Sales Outreach AI Worker (SDR)
Use cases: outbound prospecting, personalized email, follow-up, meeting booking, CRM enrichment. Pricing: Per-Outcome ya Per-Seat with caps. Operational KPIs: reply 2-8%, reply-to-meeting 10-25%, personalization >80%, bounce <5%, compliance violations 0%. Financial KPIs: $50-$300 per booked meeting, cost $5-$50, LTV/CAC 8-20x, CAC payback 8-14 months.
3. Code Generation AI Worker
Use cases: IDE completion, function generation, refactoring, tests, review. Pricing: Per-Seat with usage caps ya Hybrid. Operational KPIs: acceptance 25-45%, pass rate 60-80%, hallucinated APIs <2%, latency <200ms. Financial KPIs: $20-$100 per developer seat/month, compute $5-$30, gross margin 65-80%.
4. Document Analysis AI Worker
Use cases: contract review, invoice processing, due diligence, regulatory filings. Pricing: Per-Outcome, sometimes quality tiers. Operational KPIs: accuracy 92-98%, throughput 100-10,000 docs/hour, hallucination <1%, review flag 5-20%. Financial KPIs: revenue $1-$25/document, cost $0.20-$5, margin 60-80%.
5. Voice Agent
Use cases: inbound calls, outbound campaigns, appointment setting, customer service. Pricing: per-minute, per-call, ya per-outcome. Operational KPIs: containment 30-70%, quality 4.0-4.5/5, latency <800ms, ASR >95%, hang-up <8%. Financial KPIs: $0.25-$2.50/minute ya $1-$15/call, cost $0.10-$0.40/minute, gross margin 50-70%.
6. Search & Retrieval AI Worker
Use cases: enterprise search, semantic Q&A, RAG assistants, document discovery. Pricing: Per-Seat ya Per-Query. Operational KPIs: retrieval precision 70-90%, answer accuracy 75-90%, p95 latency <3 sec, citation accuracy >90%. Financial KPIs: $30-$150/seat/month, compute $8-$40, gross margin 60-75%.
7. Claims Processing AI Worker
Use cases: insurance claims, healthcare prior authorization, expense processing. Pricing: Per-Outcome ya Value-Based. Operational KPIs: auto-adjudication 40-75%, decision accuracy >96%, appeal/reversal <5%, compliance violations 0%. Financial KPIs: $5-$50 per claim, cost $1-$10, contribution 65-85%, NRR 120-150%.
8. Meeting Summarization AI Worker
Use cases: notes, action items, decision docs, CRM updates. Pricing: usually Per-Seat bundled. Operational KPIs: coverage 80-95%, accuracy 90-98%, hallucination <2%, speaker attribution >85%, edit rate <30%. Financial KPIs: $10-$40/seat/month, compute $3-$15, gross margin 65-80%.
9. Marketing Content AI Worker
Use cases: blog posts, ad variants, email campaigns, social media, SEO content. Pricing: Per-Seat ya per-generated-output. Operational KPIs: acceptance 30-60%, quality 3.5-4.5/5, brand voice >85%, originality >90%. Financial KPIs: $50-$500/seat/month, compute $15-$100, gross margin 60-75%, SMB churn high.
10. Legal Research AI Worker
Use cases: case-law research, contract analysis, compliance checking, legal drafting. Pricing: premium Per-Seat. Operational KPIs: citation accuracy >95%, hallucination below 0.5%, completeness 80-95%, time saved 30 min-4 hr. Financial KPIs: $200-$2,000 attorney seat/month, compute $50-$300, gross margin 70-85%.
11. Recruiting AI Worker
Use cases: sourcing, resume screening, outreach, scheduling, engagement. Pricing: Per-Seat ya Per-Hire. Operational KPIs: sourcing precision 60-80%, reply rate 15-35%, bias mitigation tracked, throughput 50-500 candidates/recruiter/week. Financial KPIs: $200-$1,500/seat/month, per-hire 5-25% first-year salary, gross margin 60-75%.
12. Financial Analysis AI Worker
Use cases: earnings analysis, portfolio research, financial modeling, M&A, equity research. Pricing: premium Per-Seat. Operational KPIs: calculation accuracy >99%, source citation >95%, hallucination below 0.5%, latency <60 sec. Financial KPIs: $500-$5,000 analyst seat/month, compute/data $100-$500, gross margin 75-88%.
Consolidated benchmarks table
| Metric | Layer | Pre-revenue (Seed) | Early ($1-5M ARR) | Mid ($5-25M ARR) | Scaling ($25M+ ARR) |
|---|---|---|---|---|---|
| ARR | 3 | <$1M | $1-5M | $5-25M | $25M+ |
| ARR growth (YoY) | 3 | NM | 200%+ | 100-200% | 50-120% |
| Gross margin | 3 | NM | 50-70% | 60-75% | 65-78% |
| Compute as % revenue | 3 | NM | 25-50% | 20-35% | 15-30% |
| NRR | 3 | NM | 105-125% | 115-135% | 120-140% |
| GRR | 3 | NM | 85-95% | 90-95% | 92-96% |
| CAC payback | 2 | NM | <24 months | <18 months | <14 months |
| LTV/CAC | 2 | NM | 3-8x | 5-12x | 5-15x |
| Burn Multiple | 4 | NM | <2.5x | <2.0x | <1.5x |
| Magic Number | 4 | NM | 0.5-1.0 | 0.8-1.5 | 0.7-1.2 |
| ARR per employee | 4 | NM | $100-200K | $150-300K | $200-400K |
| Cash runway | 3 | 18-24 months | 18-24 months | 18-24 months | 18-24 months |
| Compute concentration | 3 | NM | <90% | <80% | <70% |
| Pilot-to-production conversion | 3 | NM | 40-60% | 55-70% | 65-80% |
| Outcome attribution accuracy | 1 | NM | >90% | >95% | >97% |
Diagnostic playbooks
Jab metric off ho, pehle investigation karein:
- Burn Multiple >2.5x rising: cohorts, S&M efficiency, headcount growth investigate karein.
- NRR below 100%: gross retention vs expansion separate karein, churn cohort attributes dekhein.
- Gross margin declining: compute-cost trend, price realization, cost-per-outcome trend dekhein.
- CAC payback >18 months: segment unit economics, sales-cycle trend, win rate dekhein.
- High outcome rate but low gross margin: pricing, compute cost, overhead allocation check karein.
- Bookings much higher than recognized revenue: revenue policy, deferred revenue waterfall, attribution telemetry validate karein.
- Cost-per-outcome flat/rising: workflow drift, cache hit rate, prompt efficiency, model upgrades inspect karein.
- Customer concentration >30% top 5: diversification roadmap aur churn-protection plan banayein.
- Compute concentration >80% one provider: price-change exposure, outage exposure, multi-provider integration cost estimate karein.
- R&D >60% revenue past Series A: engineering output aur capital allocation review karein.
Cohort dashboard template
Cohort analysis with model-cost decay highest-leverage analytical tool hai. Standard structure:
| Cohort | Customers acquired | Q+0 | Q+1 | Q+2 | Q+3 | Q+4 |
|---|---|---|---|---|---|---|
| Q1 2024 | 25 | 100% | 96% | 92% | 88% | 88% |
| Q2 2024 | 30 | 100% | 97% | 93% | 90% | 90% |
| Q3 2024 | 32 | 100% | 97% | 91% | 91% | 88% |
| Q4 2024 | 35 | 100% | 94% | 91% | 91% | 89% |
AI-native finance iske saath revenue retention aur gross margin decomposition add karta hai: behavior contribution vs model-cost-decay contribution. Yeh batata hai margin improvement pricing power se aa rahi hai ya compute tailwind se.
Stage-specific investor diligence checklists
Series A diligence. Monthly revenue, customer count flow, cohort retention, gross margin with compute breakdown, top customers, CAC, burn, capital efficiency, 18-month forecast, compute provider breakdown, org chart.
Series A bar roughly: $1-$3M ARR, 200%+ growth, gross margin above 50%, early NRR above 110%.
Series B diligence. Series A materials plus model-cost-decay decomposition, pilot conversion, segment unit economics, compute concentration, revenue recognition policy, capital allocation framework, Burn Multiple, Magic Number, Rule of 40, 24-month forecast.
Series B bar roughly: $5-$15M ARR, 100%+ growth, Burn Multiple under 2x, NRR above 120%, gross margin above 60%.
M&A diligence. Audited financials, Quality of Earnings, forecast accuracy, customer/vendor contracts, IP assessment, compliance, compute concentration, outcome attribution audit, tax structure, working capital analysis.
Compute economics deep-dive
Compute AI-native companies ka largest variable cost hota hai. Isay per-unit, per-modality, aur per-provider level par samajhna operational AI finance ka core hai.
Per-modality cost ranges (2026).
| Modality | Typical cost range | Cost driver |
|---|---|---|
| Text generation | $0.50-15 per 1M input tokens; $1.50-75 per 1M output tokens | Model size and quality |
| Voice synthesis | $0.05-0.30/minute | Voice quality |
| Voice recognition | $0.02-0.20/minute | Real-time vs batch |
| Image generation | $0.005-0.10/image | Resolution, quality |
| Video generation | $0.10-2.00/sec | Resolution, length |
| Embeddings | $0.02-0.30 per 1M tokens | Dimensionality, quality |
| Fine-tuning | $50-500 per 1M training tokens + host compute | Model size, method |
Provider pricing comparison framework.
- Foundation-model API providers: variable cost, no upfront commitment, easy path, less margin control, concentration risk.
- Hyperscaler offerings: AWS Bedrock, Azure OpenAI, GCP Vertex AI. Procurement and compliance benefits.
- Self-hosted / open-weight models: fixed GPU cost, utilization discipline, operational burden.
Build-vs-buy economics for compute.
API cost per inference = $X (variable, scales linearly)
Self-host cost per inference = (GPU hourly cost / inferences per hour at target latency) + amortized engineering cost
Most companies start on APIs, evaluate self-hosting around $5-$15M ARR, aur $25M+ ARR par hybrid strategy adopt kar sakti hain.
Cost-per-modality benchmarking. Customer-support text agent $0.20-$0.40 per resolved ticket; voice agent $0.30-$0.70/minute; image generation $0.01-$0.05/image. Monthly track karein.
Operational health metrics for AI Workers
Core KPIs ke beyond mature monitoring yeh metrics include karta hai:
Drift detection rate. Inputs ka percentage jo design distribution se bahar hota hai.
Hallucination rate by domain. Fabricated facts ko domain ke hisaab se segment karein.
Latency distribution (p50, p95, p99). Mean latency worst-served users hide kar sakti hai.
Prompt-injection resistance. Adversarial inputs par correct refusal ya containment rate.
Refusal rate appropriateness. Over-refusal aur under-refusal ko separately measure karein.
Evaluation-set performance trend. Curated eval set par time ke saath performance. Yeh regression detection ka canonical mechanism hai.
Additional worked dashboards
AgentCo dashboard $10M ARR mid-stage hybrid company cover karta hai. Neeche three more stages aur architectures:
Worked example: SeedAI at pre-revenue (Seed stage)
Profile. Pre-revenue AI agent company, launch se 4 months pehle, 8 employees, $3M Seed raised, 5 design partners, commercial revenue nahin.
Layer 1 metrics.
| Metric | Value | Notes |
|---|---|---|
| Outcome rate | 65% | Up from 45% three months ago |
| Quality score | 3.8/5 | Improving |
| Cost per outcome | $0.85 | High; should fall |
Layer 2 metrics. Not meaningful.
Layer 3 metrics.
| Metric | Value | Notes |
|---|---|---|
| Monthly burn | $200K | Team + compute |
| Cash on hand | $1.8M | After $1.2M deployed |
| Cash runway | 9 months | Tight |
| Compute spend | $15K/month | Beta usage |
Layer 4 metrics. Not meaningful.
What this dashboard tells the team. SeedAI ke liye runway, burn, beta engagement, quality trend, aur cost-per-outcome trend hi matter karte hain. Complex KPI dashboard energy waste hai.
Worked example: ScaleAI at $50M ARR Series B (value-based pricing component)
Profile. Enterprise AI company, $50M ARR, 180 employees, hybrid pricing with value-based engagements. $18M ARR value-based, $32M Per-Outcome/Hybrid.
| Metric | Value | Healthy? |
|---|---|---|
| Outcome rate | 81% | Yes |
| Outcome attribution accuracy | 96% | Yes |
| Cost per outcome | $0.31 | Fell 30% YoY |
| LTV/CAC subscription | 7x | Healthy |
| LTV/CAC value-based | 12x | Strong |
| ARR | $50M | - |
| Bookings | $68M | Strong |
| NRR | 135% | Strong |
| Gross margin | 70% | Strong |
| Variable consideration recognition | 60% | Watch metric |
| Burn Multiple | 1.2x | Strong |
| Rule of 40 | 45% | Strong |
What this dashboard tells the team. Value-based contracts premium pricing deliver kar rahe hain. Watch metric variable consideration recognition hai; audit cycles mature hone par GAAP revenue unlock ho sakta hai.
Worked example: ScaleCo at $150M ARR Series C+ (mature scaling)
Profile. Late-stage AI-native company, primarily Per-Outcome pricing, $150M ARR, 450 employees, Series C+, 800 customers.
| Metric | Value | Healthy? |
|---|---|---|
| Outcome rate | 84% | Strong |
| Cost per outcome trend | -22% YoY | Healthy |
| Outcome attribution accuracy | 98% | Excellent |
| LTV/CAC | 9x | Strong |
| CAC payback | 13 months | Strong |
| ARR | $150M | - |
| Bookings | $185M | Strong |
| NRR | 138% | Excellent |
| Gross margin | 75% | Strong |
| Compute as % revenue | 18% | Excellent |
| Burn Multiple | 0.4x | Excellent |
| Rule of 40 | 50% | Strong |
| Capital efficiency ratio | 0.94x | Strong |
What this dashboard tells the team. ScaleCo IPO-readiness ke qareeb metrics rakhta hai. Focus areas: model-cost decay slowdown, R&D compression decisions, aur Series D / strategic alternatives ka choice.
Cross-cutting concepts
Compute-as-COGS reality. Traditional SaaS hosting ko small footnote samajhta hai. AI-native finance compute ko primary line treat karta hai, often largest variable cost.
Bookings vs recognized revenue. Usage- ya outcome-based contracts mein signed bookings aur GAAP revenue materially diverge kar sakte hain. Dono numbers report karna zaroori hai.
Model-cost decay as a margin tailwind. Foundation-model prices fall karne se aaj acquire hue cohorts future mein more profitable ho sakte hain.
Pilot-to-production conversion gap. Enterprise AI paid pilots se start hota hai; mature conversion 50-75% ho sakti hai. Pilot revenue aur production ARR separate report karein.
Outcome attribution as an audit risk. Per-outcome pricing audit-grade telemetry demand karta hai; warna billing disputes aur revenue restatements ka risk hota hai.
Compute concentration risk. One or two foundation-model providers par high dependency vendor risk create karti hai.
What AI changes about every finance discipline
1. Gross margin redefined. Traditional SaaS 75-85%; AI-native 50-70% due to compute.
2. Forecasting under continuous price decay. Forecasts ko compute price decay explicitly model karna chahiye.
3. Revenue recognition complexity at smaller scales. AI-native companies $5M ARR par woh complexity face kar sakti hain jo traditional SaaS $50M par karta hai.
4. Pilot-to-production motion as standard. Enterprise AI direct annual contract ke bajaye paid pilot se production tak jata hai.
5. New role: AI Finance Engineer. Finance data infrastructure ke liye engineer/data scientist chahiye: cohort analysis, compute attribution, outcome attribution, forecasting.
Common hybrid models
Per-Call (2) -> Per-Call + Subscription (5). Usage pricing se subscription floor add hota hai, revenue predictability ke liye.
Per-Seat (1) -> Per-Seat + Usage Overage (5). Compute share revenue ka 15% exceed kare to usage overages add hoti hain.
Per-Seat (1) -> Per-Outcome (3). AI labor-replacement work kar raha ho to specific functionality outcome-priced ho sakti hai.
Pilot (Approach 9) -> Production Contract. Enterprise AI commercial sequence ka standard path.
Per-Call (2) -> Per-Outcome (3) for specific workflows. Specific workflows jahan outcome pricing 3-10x higher revenue per call de sakti hai.
Common finance failures
Compute-as-hosting misclassification. Compute ko small hosting line treat karna. Fix: COGS mein distinct compute line report karein.
ARR inflation through pilot inclusion. Paid pilot revenue ko ARR mein include karna. Fix: pilot revenue separate report karein.
Aggressive revenue recognition. Optimistic variable consideration assumptions. Fix: accountant aur auditors se early review.
Compute commitment overcommitment. Demand se zyada prepaid compute khareedna. Fix: demonstrated demand ke against commitments size karein.
Cohort analysis without model-cost decay separation. Margin improvement ki wrong attribution. Fix: synthetic-cost baseline.
Forecasting with constant compute prices. Future margins understate karna. Fix: compute-price-decay layer.
Premature CFO hire. $2M ARR par CFO hire karna aur capital burn karna. Fix: fractional CFO/controller until complexity justifies full-time CFO.
Investor reporting heavy on bookings, light on cash. Cash aur recognized revenue underemphasize karna. Fix: reporting cash and recognized revenue se lead kare.
AI-native finance anti-patterns
Treating model spend as fixed infrastructure. Fixed-fee compute deal ko per-customer attribution ke baghair use karna.
Ignoring compute concentration risk. 90%+ compute ek provider par leave karna.
Pricing based on cost rather than value. Compute markup par price karna jab customer value much higher ho.
Forecasting without model-improvement scenarios. Current AI capability constant assume karna.
Building Tier 2 metrics retroactively. Series B ke waqt cohort/model-cost data backfill karne ki koshish. Data infrastructure pehle banayein.
Minimum viable finance stack and stage recommendations
Minimum viable finance stack (Pre-revenue through Early Traction).
- Stripe ya equivalent billing - month 1. Subscription invoicing, usage metering, payment collection.
- Pilot, Bench, ya Puzzle bookkeeping - month 1. Monthly close, financial statements, tax prep.
- Mercury ya Brex banking/treasury - month 1. Modern banking infrastructure.
- Three numbers weekly - month 1. Revenue, gross margin, runway.
- Quarterly forecast spreadsheet - month 6. Simple 18-month revenue and burn projection.
- External auditor relationship - Series A diligence. AI-native experience wali CPA firm identify karein.
Stage-based recommendations.
| Company stage | Primary finance practices | Avoid for now |
|---|---|---|
| Pre-revenue | Stripe + bookkeeping, three numbers weekly, runway forecast | CFO hire, FP&A software, formal audit |
| Early revenue ($1M-$5M ARR) | Controller, board reporting basics, revenue recognition policy | CFO hire, custom FP&A platform |
| Scaling pre-Series B ($5M-$15M ARR) | VP Finance/senior controller, monthly close, cohort analysis, AI Finance Engineer | CFO unless IPO trajectory |
| Post-Series B ($15M+ ARR) | CFO, FP&A team, cohort analysis, audit-defensible attribution | Premature IPO infrastructure |
Common mistake CFO too early hire karna hai. Right sequence: founder books -> fractional controller -> full-time controller -> VP Finance -> CFO.
How to use this catalog
First, har approach run karna zaroori nahin. Most companies two to four pricing architectures use karti hain, revenue/cost mechanics universally apply karti hain, planning gradually develop karti hain, aur external reporting stage ke metrics ke saath karti hain.
Second, sequencing perfection se important hai. First three years mein basics right rakhein: Per-Call/Per-Seat, Stripe + bookkeeping, three numbers tracked, simple forecast.
Third, AI era finance functions ko apna data infrastructure engineer karna parta hai. Cohort margin with model-cost decay, outcome attribution accuracy, compute concentration, forecast accuracy under price decay out-of-box tools mein nahin milte.
Common beginner questions
"AI-native finance regular SaaS finance se kaise different hai?"
Teen differences: gross margins 50-70% vs 75-85% due to compute; pricing usage/outcome/hybrid ho sakti hai; forecasting compute-price decay explicitly model karti hai. Finance mechanics otherwise same hain.
"Kya mujhe CFO chahiye?"
Usually $10M ARR se pehle nahin, kabhi $25M+ tak nahin. Fractional CFO ya experienced controller use karein jab tak full-time strategic finance leader genuinely required na ho.
"Bookings aur recognized revenue mein difference kya hai?"
Bookings signed deals ki contract value hain. Recognized revenue GAAP revenue hai jo obligations satisfy hone par P&L par aata hai.
"AI company ke gross margin ko kaise sochun?"
Compute ko COGS line include kar ke calculate karein. 60-70% healthy; below 50% warning sign.
"Revenue recognition kab worry karni chahiye?"
First contract sign hone ke moment se. Pure subscription simple hai; outcome/value-based contracts experienced accountant require karte hain.
"Revenue forecast kaise karun jab itna unpredictable hai?"
Two layers: customer revenue and compute costs with explicit decay scenarios. Base aur conservative case board ko present karein.
"Board ko kya metrics report karun?"
Tier 1: ARR, NRR, gross margin, Burn Multiple, runway. Tier 2: compute %, cohort gross margin trend, pilot conversion, bookings vs recognized revenue. Tier 3: compute concentration, forecast accuracy, capital allocation.
"Solo founder with no finance background kya kare?"
Weekly revenue, gross margin, aur runway honest rakhein. Stripe + Pilot/Bench + Mercury use karein. Fundraising ke waqt fractional controller hire karein.
Appendix A: Glossary
ARR (Annual Recurring Revenue). Subscription contracts ki annualized contracted revenue.
ASC 606. US revenue recognition accounting standard.
Audit defensibility. Books ka auditors, investors, aur acquirers ki scrutiny survive karna.
Bookings. Signed deals ki contractual value, revenue recognition timing se independent.
Burn Multiple. Cash burned / net new ARR.
CAC (Customer Acquisition Cost). New customer acquire karne ka fully-loaded cost.
CAC payback period. Acquisition cost recover hone ka time.
Capital allocation. Incremental dollars compute, people, acquisition, aur runway mein split karna.
Capital efficiency. Deployed capital per revenue output.
Cash runway. Current burn par operations fund karne ke months.
Cohort analysis. Same period mein acquired customers ko time ke saath track karna.
Compute COGS. Compute cost jo cost of goods sold mein flow karta hai.
Compute concentration risk. Single provider par compute spend ka high percentage.
Contribution margin. Revenue minus all variable costs.
Deferred revenue. Collected ya contracted revenue jo GAAP ke under abhi recognize nahin hua.
Forecast accuracy. Forecasted vs actual revenue ka historical match.
FP&A. Forecasting, budgeting, aur strategic financial analysis function.
Gross margin. Revenue minus COGS as percentage of revenue.
GRR. Existing customer recurring revenue retained excluding upsell.
Hybrid pricing. Do ya zyada pricing components ka combination.
LTV. Customer lifetime gross-margin contribution.
LTV/CAC ratio. Lifetime value divided by acquisition cost.
Magic Number. New ARR in quarter / prior quarter sales-and-marketing spend.
Model-cost decay. Foundation-model prices ka time ke saath girna.
NRR. Existing customers including upsell se retained recurring revenue.
Outcome attribution. Proof infrastructure ke AI ne kaunsa outcome deliver kiya.
Per-call pricing / Usage pricing. API call, token, second, ya query ke hisaab se pricing.
Per-outcome pricing. Defined result deliver hone par charge.
Per-seat pricing. User ke hisaab se fixed fee.
Pilot. Short paid engagement before production contract.
Pilot-to-production conversion rate. Pilots ka percentage jo production contracts bante hain.
Prepaid compute commitment. Provider ko fixed compute volume ke liye upfront commitment.
Predictability. Forecast accuracy.
Revenue recognition. Revenue books par kab count hota hai.
Runway. Cash runway.
SaaS metrics. ARR, NRR, gross margin, CAC, payback, LTV, Burn Multiple, Magic Number.
Service-as-Software. Outcome-based AI pricing models ka label.
Synthetic cost baseline. Original prices par cohort ka hypothetical cost.
Tier 1 / Tier 2 / Tier 3 metrics. Investor reporting framework.
Variable consideration. Contract price ka uncertain future events par dependent hissa.
Value-based pricing. Customer value created ka percentage charge karna.
Notes
-
Bessemer Cloud Index aur Bessemer Venture Partners public-cloud-software margins aur AI-native economics par useful public references hain.
-
Andreessen Horowitz growth team, especially AI margins aur unit economics par writing, model-cost decay aur cohort margin dynamics ko explain karti hai.
-
David Skok at Matrix Partners SaaS finance framework ke canonical references mein se hain; Burn Multiple, Magic Number, aur CAC payback Tier 1 framework ko inform karte hain.
-
Tomasz Tunguz aur Theory Ventures ki research AI-native finance benchmarks aur trends par ongoing source hai.
-
Christoph Janz ka "5 Ways to Build a $100M Business" SaaS revenue-architecture foundation provide karta hai jise AI-native pricing extend karti hai.
Other influences: David Sacks on Burn Multiple; Patrick Campbell at Profitwell on pricing strategy; FASB ASC 606; AICPA technical advisory committees; Big Four revenue-accounting teams jo outcome-based aur value-based contracts ke audit-defensible practices develop kar rahe hain.