SaaS Metrics School

Ben Murray
  • 4.6 (10)
  • MANAGEMENT
  • UPDATED WEEKLY

Ben Murray brings you actionable SaaS metrics lessons that he has learned through years of being in the SaaS CFO trenches. Whether you are new to SaaS or a SaaS veteran, learn the latest SaaS and AI metrics, finance, and accounting tactics that drive financial transparency and improved decision-making. Ben’s SaaS metrics blog consistently rates a 70+ NPS, and his templates have been downloaded over 100,000 times. There is always something to learn about SaaS and AI metrics.

  1. 7H AGO

    2 AI Metrics Every SaaS CFO Should Track Today

    If you're shipping AI product lines, are you measuring the two metrics that actually tell you whether your AI is making money — or burning it? In episode #371, Ben Murray covers two AI unit economics metrics every SaaS CFO and founder should be tracking today: the Inference Expense Ratio and the Work-to-Inference Ratio. Traditional SaaS metrics aren't enough anymore — and a year from now, when your board, investors, and potential acquirers start asking for AI margin and efficiency data, the companies that built the chart-of-accounts structure now will have clean answers. Everyone else will be scrambling. The Inference Expense Ratio (AI revenue ÷ inference cost) — and why you can start calculating this from your GL today if your chart of accounts is set up properly The healthy benchmarks: 10:1 for AI-infused products, 5:1 for AI-native, and why 3:1 is the warning zone where inference is silently eating your gross margin Why this metric only works if your chart of accounts cleanly separates AI revenue from non-AI revenue — and the SKU tagging that makes it possible The Work-to-Inference Ratio — how Salesforce's "agentic work units" concept lets you measure whether your AI is getting more efficient over time Why every AI product needs its own definition of a "work unit" — record updated, report generated, MCP called — and how the wrong definition will distort your margin trends The chart-of-accounts evolution every SaaS company needs right now: from SaaS-only structure to SaaS + AI, with new GL accounts for inference cost in DevOps COGS How the Inference Expense Ratio connects to Ben's ROSE metric — measuring revenue produced per dollar of employee, contractor, and agentic AI spend Tune in to get the AI unit economics framework in place — before your board and investors start asking the questions you can't answer. Resources Mentioned Ben's new AI course: https://www.thesaasacademy.com/ai-finance-metrics-saas ROSE metric: https://www.thesaascfo.com/saas-rose-metric/

    4 min

  2. 1D AGO

    What Belongs in AI COGS? The Financial Framework SaaS Companies Are Scrambling to Build

    Are AI inference costs already eating into your gross margin — and you can't even see them on your P&L? In episode #370, Ben Murray breaks down exactly what belongs in AI COGS for SaaS companies offering an AI-first or AI-infused product line. Inference bills are stacking up fast, infrastructure-layer spend is the surprise line item nobody priced in, and most finance teams haven't built the GL account structure to capture any of it cleanly. If you don't get the framework in place now, you'll be reporting AI gross margin you can't actually defend by next quarter — and your board will notice. The 5 cost categories every AI COGS framework needs — inference, model hosting/GPU infrastructure, the AI infrastructure layer, monitoring and observability, and AI-specific support Why AI inference costs deserve their own GL account — and shouldn't be buried inside your cloud hosting bill where they disappear The surprise cost line one industry report flagged as the #1 unexpected AI expense — hiding in data platform usage, networking, and egress How to structure your COGS cost centers so you can deliver clean margins by AI product line, not just lumped together at the company level Why token tracking by customer cohort (heavy / medium / light users) is now table stakes for any AI product sold as a subscription The deployed-engineer question: should AI support tickets sit with tech support or a specialized team — and how that decision rewires your margin model Tune in to get the AI COGS framework in place before your gross margin lands on a board slide you can't defend. Resources Mentioned Ben's new AI course: https://www.thesaasacademy.com/ai-finance-metrics-saas Ben's blog post: What Should Be Included in AI COGS: https://www.thesaascfo.com/what-should-be-included-in-ai-cogs/ SaaS Metrics Foundation course: https://www.thesaasacademy.com/the-saas-metrics-foundation

    4 min

  3. 2D AGO

    How Claude Opus 4.7's New Tokenizer Quietly Raised Your AI Bill by Up to 35%

    Did your AI bill just jump overnight — even though no one announced a price increase? In episode #369, Ben Murray breaks down the hidden AI price hike that's quietly hitting SaaS P&Ls this month. Anthropic shipped a new tokenizer underneath Claude Opus 4.7 — same menu pricing as 4.6, but real enterprise workloads are showing 12-27% higher effective cost, with some prompts consuming up to 35% more tokens for identical output. Most finance teams won't catch this variance until the invoice lands. If you're running AI in production, paying for Claude Code, or modeling AI COGS into next year's plan, this is the cost dynamic you need on your radar before the next board meeting. Why "same per-token pricing" doesn't mean same cost — and how a new tokenizer can quietly inflate your token consumption by 35% The real-world math: how a $50K/month API spend can balloon to $67K with zero changes to the pricing page What Anthropic's doubled Claude Code per-developer estimate ($6 → $13/day) signals about the end of subsidized AI pricing Why the era of "AI is just going to keep getting cheaper" assumptions is breaking down — and what that means for forecasting and runway The exact metrics to monitor in your Anthropic console today to catch token volume spikes before they hit your GL How to use the Inference Efficiency Ratio (revenue ÷ token costs in COGS) to measure AI margin if you're embedding AI into your product Why finance teams now need to document internal-use AI models the same way they document internal-use software Tune in before your next Anthropic invoice lands — and learn what to track now so AI variance doesn't become a board question. Resources Mentioned Dev.to article: https://dev.to/dev_tips/the-ai-price-hike-that-never-showed-up-on-the-pricing-page-your-bill-went-up-27-anyway-3mn5 Put your AI framework in place: https://www.thesaasacademy.com/ai-finance-metrics-saas

    4 min

  4. MAY 1

    Why Token Usage Tells You Almost Nothing About Your AI Product's Real Value

    Can you actually prove what your AI product is doing for customers — or are you still pointing at token counts and hoping the board nods along? In episode #368, Ben Murray breaks down the four layers of AI measurement that every SaaS company needs to communicate internally and externally. Token usage is table stakes. The real question is whether you can move up the stack from consumption to work performed to verified outcomes to quantifiable P&L impact. Get this wrong, and your AI story falls apart in front of investors, customers, and your own finance team. Get it right, and you finally have ROI math a CFO will actually approve. Why AI inference belongs in COGS / DevOps — and what that means for the gross margin story behind your AI features and product lines How Salesforce's "agentic work units" framing on its latest earnings call signals where AI reporting is heading for the rest of SaaS Where true outcome-based pricing actually lives on the pricing page (HubSpot, Zendesk, and others) — and where Agentforce was really still usage-based in disguise How Layer 4 business impact replaces fuzzy ROI calculators with objective math What to show your board and investors at each layer so your AI value story holds up under scrutiny Tune in before your next board meeting — your AI story needs more than token counts. Resources Mentioned Ben's blog post on AI measurement and AI work units: https://www.thesaascfo.com/the-four-layers-of-ai-measurement-a-cfos-framework/ Ben's academy: https://www.thesaasacademy.com/

    5 min

  5. APR 26

    Salesforce Invented a New KPI on an Earnings Call — Here's Why You Should Too

    Salesforce just invented a new metric on their latest earnings call — not because they needed one, but because Wall Street didn't have the vocabulary to value what they built. In episode #366, Ben Murray breaks down Salesforce's Q4 FY2026 earnings call — not the financials, but the narrative architecture: a new unit of measurement for AI value (the AWU), a framing strategy designed to neutralize the biggest fear enterprise buyers have about AI, and three customer testimonials brought live onto the call. This is the communication playbook every SaaS operator can steal when explaining AI to boards, investors, and customers — at a time when the old metrics (tokens, MAUs, queries) no longer tell the value story. Why Salesforce introduced the Agentic Work Unit (AWU) — and what 2.4 billion AWUs against 19 trillion tokens reveals about the limits of token-based AI metrics The AWU-to-token ratio as a customer health signal — and why this is the metric your AI-enabled SaaS dashboard is missing The "humans and agents working together" framing that lets you sell AI capabilities without triggering the "we're going to lay people off" deal-killer How Wyndham's 8,300-hotel deployment, SharkNinja's 250,000 holiday-season engagements, and Lemkin's SaaStr transformation prove ROI when slides can't How to expand your SaaS metrics dashboard from 5 pillars to 6 — and the AI-era KPIs (AWUs, AI-attributed ARR, input-to-output ratios, customer outcome metrics) that belong in the new pillar Tune in before your next board meeting or AI sales pitch — and steal the vocabulary that's about to define the category. Resources Mentioned Salesforce Q4 FY2026 earnings call transcript Ben's 5-pillar SaaS metrics dashboard — and the upcoming 6-pillar AI-era expansion: https://www.thesaascfo.com/downloads/five-pillar-metrics-framework/

    7 min

  6. APR 25

    Should You Price on Outcomes? What HubSpot's $0.50 Bet Means for Your SaaS Revenue Model

    HubSpot's 50-cent bet may have just forced every SaaS founder to ask whether their current revenue model is still defensible. In episode #365, Ben Murray breaks down HubSpot's April 2nd announcement — slashing its Breeze customer agent from $1 to 50¢ per resolved conversation, plus a shift on its prospecting agent to $1 per qualified lead — and what this risk transfer means for SaaS revenue, forecasting, and the metrics CFOs need to start tracking. With Salesforce Agent Force hitting $800M in Q4 run rate and over 60% of bookings coming from existing-customer expansion, the question is no longer whether AI is reshaping SaaS pricing, but how fast and how unevenly. Ben pulls in his SEC filings research and a sharp counterpoint from Salesforce's own earnings call to show why the "SaaS is dead" narrative is overplayed. The two HubSpot pricing changes that signal a true risk transfer — and the 65% resolution rate (90% for top performers) that makes the bet credible Why "75% of AI agent vendors have no systematic approach to pricing" should put your pricing committee on notice this quarter The forecasting and metrics shift CFOs need to make as outcome-based pricing erodes predictable usage-based revenue — and the new KPIs that replace the old ones How Salesforce Agent Force's $800M Q4 run rate and 60%+ expansion bookings prove the AI revenue thesis — while Robin Washington's earnings call comment complicates the seat-erosion story The pricing reality check Ben pulled from analyzing 100+ SEC filings — and what it means for whether your ICP actually fits outcome-based pricing Listen before your next pricing committee meeting — and bring your CFO. The forecasting implications alone are worth the six minutes. Resources Mentioned Article from: https://thesaaslibrary.com/per-seat-pricing-dead-saas-2026/ SaaStr post by Jason Lemkin: https://www.saastr.com/salesforce-now-has-3-pricing-models-for-agentforce-and-maybe-right-now-thats-the-way-to-do-it/ Salesforce Q4 earnings call Ben's blog post: https://www.thesaascfo.com/your-ai-feature-is-quietly-destroying-your-gross-margin/

    6 min

  7. APR 21

    AI Inference Costs Are Crushing SaaS Gross Margins — Here's What to Do About It

    Is your AI SaaS company skating on thin ice because of exploding compute costs you're not tracking? In episode #365, Ben Murray tackles one of the most pressing financial challenges facing AI-first SaaS companies: the structural margin compression caused by LLM inference costs. Traditional SaaS was built on near-zero marginal cost per customer — that era is over. If you're building on top of AI, every prompt, query, and agentic workflow is a hard COGS line that scales with revenue, and if you're not managing it, it will quietly destroy your unit economics. Why AI-first SaaS companies are running 50–60% gross margins (vs. 70–80% for legacy SaaS) — and what Bessemer data shows about AI supernovas with margins as low as 25%. How inference and compute costs differ fundamentally from traditional SaaS COGS — and why they won't scale down the way hosting costs did Why token costs vary wildly (from $1–2 per million to $30–180+ for frontier models) and how that variability makes feature-level economics a CFO priority 5 tactical ways to reduce LLM spend: model routing, prompt caching, context compaction, semantic caching, and batch processing How to set up your GL accounts and COGS tracking to allocate inference costs by feature — so you actually understand the economics of what you've built Tune in before your next board meeting — because if you're not tracking AI inference costs at the feature level, you're flying blind on your most important unit economics. Resources Mentioned The SaaS CFO: https://www.thesaascfo.com/ Ray Rike — AI to ROI Newsletter: https://ai2roi.substack.com/ Tomas Tunguz: https://tomtunguz.com/ Fungies.io — 5 Ways to Save on LLM Costs: https://fungies.io

    6 min

  8. APR 9

    How to Track Digital Labor in Your SaaS P&L

    In episode #364, Ben Murray breaks down how SaaS finance teams should structure their chart of accounts to properly track inference costs, productivity AI, and agentic AI spend. As organizations shift from W-2 headcount to token costs and agentic software, your current expense coding may be out-of-date. If you can't see where the AI spend is going, you can't tie it to ROI — and you definitely can't make the case for going fully agentic. Why COGS is the right home for product inference costs (Claude, OpenAI, Gemini) — and why lumping them in with hosting is a mistake The three distinct AI spend buckets every SaaS CFO needs to track: direct COGS delivery costs, general productivity tools, and explicit labor substitution (agentic AI) Why agentic AI spend deserves its own GL account — and how that ties directly into your ROSE metric Where the tracking gets fuzzy: productivity tools vs. true labor displacement, and how to think about cause-and-effect as a CFO How AI spend reshapes the ROSE metric as orgs push toward $5M–$10M ARR per FTE targetsTune in to get the chart of accounts framework SaaS CFOs need before AI spend becomes too big to ignore — and too messy to measure. Resources Mentioned ROSE Metric: https://www.thesaascfo.com/saas-rose-metric/

    5 min
See All (371)

4.6
out of 5
10 Ratings

About

Ben Murray brings you actionable SaaS metrics lessons that he has learned through years of being in the SaaS CFO trenches. Whether you are new to SaaS or a SaaS veteran, learn the latest SaaS and AI metrics, finance, and accounting tactics that drive financial transparency and improved decision-making. Ben’s SaaS metrics blog consistently rates a 70+ NPS, and his templates have been downloaded over 100,000 times. There is always something to learn about SaaS and AI metrics.

Information

You Might Also Like