How to Track MRR, Churn, and LTV Without Spreadsheets

Every SaaS founder I have ever talked to has, at some point, built the spreadsheet. It usually starts on a Sunday night, around month four of paying customers. You pull a Stripe export, you build a tab for MRR, you add a tab for churn, you write a VLOOKUP that aggregates by customer. For a quarter, maybe two, it works.

Then the cracks show up. A customer upgrades mid-month and the formula counts them twice. A refund hits and you forget to subtract it. An annual plan comes in and your monthly MRR jumps by $1,200 for one month because somebody divided by zero instead of twelve. The dunning queue you meant to track is now just a list of disappointed Slack reactions. And the question your investor or your co-founder asks — what is our trailing three-month NRR? — takes you 45 minutes to answer with a confidence interval of "probably correct."

This is the moment every recurring-revenue business has. Spreadsheets are fine for a quarter, then they become the bottleneck. This post is the operator's guide to leaving them behind: which metrics actually matter, how to define each one precisely enough that two co-founders cannot argue about the number, the five dashboard views you actually need, and how to wire it all up without writing your own analytics platform from scratch.

The pitch for a spreadsheet is that it is flexible. The reality is that flexibility is the bug, not the feature. Five things go wrong, almost always in this order.

First, the reconciliation work piles up. Every month-end becomes a manual job — pull the Stripe export, dedupe customers who upgraded, normalize annual plans to monthly, subtract refunds, add back proration. By month six this is a two-hour task. By month twelve it is a half-day task you keep pushing to Sunday night.

Second, there is no real-time view. You know what your MRR was at the end of last month. You do not know what it is right now, today, including the cancellation that came in at 9:14 this morning. For a founder making pricing or hiring decisions, that lag is the difference between a confident call and a guess.

Third, the formulas drift. The person who built the sheet leaves, gets promoted, or just forgets the assumptions baked into row 47. A new pricing plan launches and nobody updates the normalization logic. Six months later you discover that everyone on the $29/mo annual plan has been counted at $348 of monthly MRR instead of $24.17.

Fourth, edge cases are where the truth lives, and spreadsheets handle them poorly. Mid-month upgrades, partial refunds, paused subscriptions, failed payments that recover three days later, customers who downgrade and then re-upgrade in the same period, comped accounts, internal accounts. Each one is its own special case. Each one is a place the formula quietly lies to you.

Fifth, there is no single source of truth. Sales has a number, finance has a different number, the founder has a third number in their head. None of them are wrong, exactly. They are just three different snapshots from three different times with three different assumptions. Nobody knows which is the real one.

Before you can track anything, you have to agree on what you are tracking. The number one mistake I see in seed-stage SaaS metrics conversations is two co-founders arguing about churn when they are using different definitions. Here is the operator's reference, with definitions tight enough to settle that argument.

MRR is the sum of all monthly-normalized subscription value across your active customer base on a given date. Annual plans get divided by twelve. Quarterly plans get divided by three. One-time fees, setup charges, and overages do not count.

MRR is a stock, not a flow. It is what you have right now, not what you earned this month. The distinction matters because the flow version (sometimes called "recognized monthly revenue") includes one-time charges and looks bigger, which is appealing and wrong.

MRR breaks into four components every month:

Net New MRR = New + Expansion − Contraction − Churned. This is the single most useful number you can put on a monthly review. It tells you whether the business is growing or shrinking this month, and what the shape of that movement is.

A company with $20k new MRR and $19k churned MRR is technically still growing by a thousand dollars. Same company a quarter later, with $20k new and $25k churned, is shrinking even though new sales look identical. Net New MRR is the number that catches the second scenario.

ARR = MRR × 12. It is the annualized version of your current monthly run rate. ARR is the number investors want to see on a pitch deck and is the standard unit for SaaS conversations once you cross roughly $1M run rate. Below that, MRR is more honest because monthly variance is large enough that annualizing creates false precision.

GRR measures how much MRR you kept from a starting cohort, excluding any expansion. The formula:

GRR = (Starting MRR − Contraction − Churned) / Starting MRR

If you started the period with $100k MRR from a cohort, lost $5k to downgrades and $5k to cancellations, you retained $90k, which is 90% GRR. GRR can never exceed 100%. A healthy SaaS business runs GRR at 90% or higher annually. SMB SaaS often runs 80-90%; mid-market and enterprise SaaS targets 90-95%+.

NRR is GRR with expansion added back in. The formula:

NRR = (Starting MRR + Expansion − Contraction − Churned) / Starting MRR

NRR above 100% is the single best leading indicator in SaaS. It means your existing customer base is growing in revenue even with churn factored in, before you sell to a single new customer. A company at 110% NRR is compounding on installed base alone. Best-in-class enterprise SaaS routinely reports 120%+ NRR. SMB SaaS typically lives between 90% and 105%, with anything above 100% considered strong.

The pair to watch is GRR and NRR together. High NRR with low GRR means you are masking high churn with aggressive upselling — eventually the upsell ceiling catches up. High GRR with NRR near 100% means you retain well but do not expand, which limits long-term efficiency.

These two are constantly confused. They measure different things and tell different stories.

Logo churn = number of customers who cancelled / number of customers at start of period. It counts customers, not dollars. If you lost 5 customers out of 100, that is 5% logo churn.

Revenue churn = MRR cancelled / starting MRR. It is dollar-weighted. If those 5 lost customers were your smallest accounts, revenue churn might be 2%. If they were your largest, it might be 15%.

For a healthy SaaS business, revenue churn should be lower than logo churn — meaning your bigger customers stick around longer than your smaller ones. If revenue churn exceeds logo churn, you have a top-of-funnel problem disguised as a retention problem: your largest customers are leaving disproportionately.

LTV is the total gross profit you expect to earn from an average customer across their entire relationship with you. The simplest defensible formula:

LTV = ARPU × Gross Margin × Average Customer Lifetime

Where ARPU is average revenue per user per month, gross margin is the percentage of revenue left after direct costs (hosting, payment processing, support labor), and average customer lifetime in months equals 1 / monthly churn rate.

For a SaaS business with $50 ARPU, 80% gross margin, and 4% monthly churn: lifetime = 1 / 0.04 = 25 months. LTV = $50 × 0.8 × 25 = $1,000.

The gross margin step is where most founders cheat. If you skip it and just multiply ARPU by lifetime, you are reporting revenue, not value. The difference matters because your customer-acquisition spend has to come from gross profit, not gross revenue.

CAC = (Sales + Marketing spend in a period) / new customers acquired in that period. The denominator is just new logos, not expansion or renewals. Fully-loaded CAC includes salaries, ad spend, tools, content production, and anything else attributable to acquisition. Most early-stage founders under-count CAC by 30-50% by leaving out their own time or contractor costs.

For a hard truth: if you spent $20k on marketing last month and got 40 new customers, CAC is $500. There is no fancier way to compute it that makes it lower.

LTV:CAC is the ratio of customer value to acquisition cost. A 3:1 ratio is the rough industry benchmark for sustainable SaaS growth — you earn three dollars of customer lifetime value for every dollar spent acquiring them. Below 3:1, you are burning cash to grow. Above 5:1, you are usually under-investing in acquisition.

In the example above: LTV $1,000, CAC $500, ratio 2:1. That business needs to either raise prices, reduce churn, or cut acquisition spend before it scales.

Payback period is the number of months it takes to recover your CAC from a customer's gross profit. The formula:

Payback period (months) = CAC / (ARPU × Gross Margin)

For the same example: $500 / ($50 × 0.8) = 12.5 months.

A healthy SMB SaaS payback period is under 12 months. Enterprise SaaS can tolerate 18-24 months because deal sizes and retention are higher. Anything past 24 months for an SMB business is a structural problem — you are essentially financing your own growth with cash you do not have yet.

You can build a dashboard with 40 charts. You can also drown in it. Here are the five views that actually drive decisions for a SaaS business between $5k and $200k MRR. Everything else is decoration.

One headline number — current MRR — with a stacked breakdown of how you got there this month. New, expansion, contraction, churned, plus the resulting Net New MRR. This is the view your finance lead, your co-founder, and you should look at every Monday morning.

The value of seeing all four components together is that it forces a story. "We grew 4% net this month" tells you nothing. "We grew 4% net this month on $18k new, $4k expansion, and only $2k churned" tells you the engine is working. "We grew 4% net this month on $30k new and $14k churned" tells you the bucket is leaking faster than ever even though the top line looks healthy.

A simple line chart of total MRR by month for the trailing 12 months, with the four components stacked beneath as bars. This is the view you screenshot for investor updates and board meetings. It is also the view that catches slow-bleed problems — quarters where Net New MRR is technically positive but the trend line is flattening.

The most common mistake here is showing MRR without showing churn underneath it. A growth line that looks fine in isolation can hide a churn rate that is doubling beneath the surface.

A grid where rows are signup cohorts (customers who joined in January, February, March, and so on) and columns are months since signup (month 1, month 2, month 3...). Each cell shows what percentage of that cohort's starting MRR you still have.

This is the single most important view for diagnosing whether your product is actually working. Cohorts that retain 70%+ at month 12 are healthy. Cohorts that drop to 50% by month 6 are signaling that customers are churning before they get value. If newer cohorts retain better than older ones, your product is improving. If newer cohorts retain worse, something is breaking — usually in onboarding or pricing.

Cohort tables are also where you spot the difference between a SaaS business and a churn-and-burn business that looks like one.

A line chart showing your LTV:CAC ratio month over month, with a horizontal reference line at 3:1. This is the view that tells you whether your acquisition strategy is sustainable.

A few patterns to watch for. A ratio that climbs over time as you add scale is healthy — you are getting more efficient. A ratio that stays flat while spend grows means you are buying growth at constant unit economics, which is fine but capped. A ratio that drops over time as spend grows means you are scaling past the audience that converts efficiently, which is the warning sign that precedes a cash crunch.

A list of customers with currently-failed payments, how many retry attempts have happened, when the next retry is, and the historical recovery rate. Failed payments are the single largest source of involuntary churn for most SaaS businesses, and the recoverable portion of that churn is usually 30-50% with good dunning logic.

The key metric here is the involuntary churn rate — what percent of your monthly churned MRR comes from failed payments versus customer-requested cancellations. If involuntary churn is over a third of total churn, your dunning system is leaking. Improvements here usually pay back faster than improvements anywhere else in the funnel because the customer was not trying to leave.

For a recurring-revenue business, the source-of-truth data lives in two or three places. Knowing which sources feed which metrics is the difference between a dashboard that updates itself and one that you maintain by hand.

Stripe (or Chargebee, Recurly, or another billing platform) is the primary source for subscription state. Every MRR component, all churn calculations, and dunning data come from your billing system. If you are using Stripe Billing, the subscription object plus the invoice object give you everything you need — when subscriptions started, what they cost, whether they are active, paused, or cancelled, and which invoices failed.

Your internal database (or your CRM) holds the customer-level context — who they are, what plan they signed up under, what attribution source brought them in, what stage of the lifecycle they are in. This is where you join Stripe data to business context.

Product analytics (Mixpanel, PostHog, Amplitude, or internal event logging) is optional for basic SaaS metrics but essential for the next layer — usage-based health scores, leading indicators of churn, and engagement-weighted cohort analysis.

The trap most teams fall into is having these three sources but never joining them. The MRR number lives in Stripe, the customer context lives in the CRM, the usage data lives in the analytics tool, and the spreadsheet that combines them gets out of date the moment somebody updates their job.

There are roughly four paths SaaS operators take to get off spreadsheets. Each has trade-offs that depend on stage and technical capacity.

Stripe's built-in dashboard is free and ships with every Stripe account. It gives you MRR, ARR, basic churn, and growth charts out of the box. It is the right answer for solo founders under $10k MRR who just need a number to look at. The limitations show up once you need cohort analysis, custom segments, or LTV computations — Stripe's dashboard is not built for those.

Dedicated SaaS metrics tools like ChartMogul, Baremetrics, and ProfitWell (now part of Paddle) connect directly to your billing platform and produce purpose-built dashboards for MRR, cohort retention, LTV, CAC, and dunning. They handle most edge cases automatically — proration, annual plan normalization, partial refunds — and they are the default path for most SMB SaaS businesses between $10k and $1M MRR. Pricing is typically based on tracked MRR, with starter plans around $100-200/month and scaling from there.

Internal builds on a data warehouse become the path once you outgrow off-the-shelf tools. A Postgres or BigQuery warehouse, dbt for transformations, a BI layer like Metabase or Looker for visualization. This is overkill below $1M ARR for almost all teams. Above that, the flexibility starts to matter — custom cohort definitions, integration of usage signals, multi-product MRR breakdowns.

Integrated platforms that combine billing, customer data, and analytics in a single workspace cut a different way. Instead of integrating four tools, you operate on one customer database where the billing data, the CRM record, and the usage signals already live together. The trade-off is depth: an integrated platform's analytics will not match a best-in-class analytics tool's depth, but it eliminates the integration tax entirely.

Deelo Analytics pulls directly from Deelo Invoicing, which integrates with Stripe (and other billing connectors) as the system of record for subscription data. That means MRR, ARR, churn, LTV, CAC, and payback period are computed natively against the same customer records that live in Deelo CRM. There is no Zap between your CRM and your analytics dashboard, because they are reading the same database.

The practical version of what this looks like: you can ask the Deelo AI Assistant "what was our trailing three-month NRR, and which segment drove it?" and get an answer with the underlying customer list, not a number you have to validate. You can drill from a churned-MRR chart into the actual list of customers who churned this month, and from there into their support history, their last login, and the invoice that failed. The data is the same data the rest of the platform runs on.

This is not the right setup for everyone. If you are a single-product SaaS business that just wants the best possible cohort retention chart, a dedicated analytics tool will out-spec us on that one chart. The Deelo proposition is different: it is what an integrated workspace looks like when the analytics, the CRM, the invoicing, and the assistant all run off the same customer record. The integration tax goes to zero. The drift between systems goes to zero. The Sunday-night spreadsheet job goes to zero.

Pulling the numbers is half the job. The other half is sitting down with them once a month and asking the right questions. Here is a 30-minute template that catches what the dashboard alone will not.

After watching enough founders pull metrics, the same five errors show up over and over. Each one looks small. Each one will lie to you about the health of the business for months before you catch it.

The reason recurring-revenue businesses build spreadsheet metrics dashboards is that the alternative — picking a tool, integrating it, agreeing on definitions across the team — feels heavier than the spreadsheet. For the first hundred customers, it is. By the thousandth, the spreadsheet has cost you more in reconciliation time, missed signals, and disagreements about what the number actually is than every dedicated tool combined.

The operators who run their SaaS business with confidence are not the ones with the prettiest dashboards. They are the ones whose MRR number, churn rate, LTV, and payback period are computed the same way every month, against the same source of truth, with the same definitions everyone on the team has agreed to. The tool you use to get there matters less than getting there. Pick one this quarter. Stop reconciling on Sunday nights.