Start with what you actually know.
You don't need to know what ARPA means to begin. Answer a few plain questions — what you make, how many customers, what you spend — and the tool works out the rest, explains every term, and shows how the numbers pull on each other. Every formula runs in your browser; nothing leaves the page.
Growth = Acquisition × Activation × Retention × Monetization × Referral
Begin with what you actually know
No jargon required. Answer these six plain questions — the tool derives the rest and seeds every calculator below. Not sure of a number? Tap a starting point, then edit.
Most explainers hand you a wall of definitions and leave you to work out what to do with them. This one runs the other way. Tell it a few things you already know about your business and it fills in the jargon, draws the relationships, and — in plain English under each section — tells you what the numbers mean for you. Nothing you type leaves your browser.
So what does this mean for you?
Those six answers already say a lot. Each customer pays you £60 a month, costs £250 to win, and stays about 25 months — so they are worth roughly £1,230 in gross profit before they leave. Every calculator below starts from exactly that; change one answer and they all move.
Your path from here
Work top to bottom — the sections are ordered by leverage, highest first. Measure, understand the engine, fix what leaks most, then buy growth and ship the lifecycle that holds it together.
Why start with these five and not the fancy metrics?
Everything else on this page is built from these answers, not looked up somewhere separately. ARPA, CAC, lifetime value and payback aren't independent facts you have to go find — they're just arithmetic on the handful of numbers you already track. Get the basics in and the rest falls out.
ARPA = revenue ÷ customers, CAC = spend ÷ new customers, and churn is simply who leaves each month. From those three the tool derives your lifetime value, your payback period, and the most you can afford to spend winning a customer.
If a number is a guess, that's fine — start with a preset and edit. Drag the sliders below and you'll see that churn and margin move the needle far more than ARPA does, which is exactly the point: it tells you where the leverage actually is.
Growth is multiplicative, not additive
Every term below is a factor in one equation. Because they multiply, the binding constraint is whichever one is worst — a 0.4 anywhere caps the whole product.
Step back from the individual calculators for a moment. Growth is one chain of five stages — a visitor has to get acquired, then activate, then stick around, then pay, and maybe refer a friend. Because each stage passes a fraction of people to the next, the outcomes multiply rather than add — which is why a single weak stage drags the whole result down, no matter how strong the others are. Move the sliders and watch which one is holding you back.
Growth = Acquisition × Activation × Retention × Monetization × ReferralRaising your worst term beats raising your best — fixing the leakiest stage multiplies everything downstream.
Funnels leak, loops compound
A funnel is a one-way drop-off; a loop feeds its output back into its input — referrals make new users who refer; revenue buys ad spend that makes revenue. Channels that don't close a loop have a ceiling.
Why multiply, not add?
The five stages are a chain, not a checklist. A visitor has to get acquired and activate and stick and pay and (maybe) refer — each stage only passes through what the one before it left behind. Because the outcomes stack on top of each other, they multiply; one weak stage caps the whole product no matter how strong the others are.
That's why the maths rewards fixing your worst stage. Lifting a 0.2 to 0.4 doubles the product, while polishing a 0.8 to 0.9 barely moves it. Find the floor, not the ceiling.
You can't move a number you can't see
Every lever below depends on knowing what your users actually do. Before any tactics, get the plumbing in: capture the events, name them well, and decide what listens.
The numbers on this page are estimates until your product is instrumented. Tracking is unglamorous and it is step one for a reason — you cannot improve activation, churn or LTV if you cannot see them move. The good news: it is a one-time setup, and the same event stream powers both your analytics and the messages you send.
Instrument your product
Fire an event for every moment that matters — signup, the activation 'aha', the key recurring action, upgrades, cancellations. PostHog (or whatever analytics you already run) is where they live. No events, no numbers.
Name events consistently
One shared convention — context.object_action, past tense — keeps events composable across product, marketing and lifecycle. Sloppy names are the reason most dashboards rot.
The naming convention →Decide how you'll act on them
Events are only useful if something listens. That something is lifecycle journeys: Hogsend turns your PostHog events into durable TypeScript journeys you review in a PR — so a churn signal or an activation milestone actually triggers a message.
See a journey →Why does tracking come before everything else?
Because every other step is a feedback loop, and a loop needs a signal. You decide whether to spend more on acquisition by watching CAC and payback; you fix churn by seeing which cohorts drop and when; you find the activation aha by comparing the users who stayed against the ones who left. None of that is possible from a billing dashboard alone — it needs product events.
It is also why analytics and lifecycle messaging belong on the same event stream. If your "user activated" event triggers both a chart and an email, the thing you measure and the thing you act on can never drift apart. That shared stream is exactly what the rest of this playbook assumes you have.
Churn sets your LTV, and LTV sets your acquisition budget
LTV = ARPA × gross margin ÷ churn. Churn sits in the denominator, so it is the highest-leverage input you have. Drag a slider; watch the chain and the flywheel react.
This is the engine room. Four numbers — what a customer pays, your margin, how fast customers leave, and what each one costs to win — decide whether growth pays for itself. The piece most people miss is that churn sits in the denominator of lifetime value, so shaving it does far more than nudging price ever could.
Drag any slider and watch the change ripple all the way along the chain to the single most useful number here: the most you can afford to pay for a customer.
So what does this mean for you?
A customer paying £60 a month at 82% margin, churning at 4.0%, stays about 25 months and is worth £1,230 in gross profit. You pay £250 to win them — a 4.9:1 return. That is the healthy band. You could pay up to £410 per customer and still clear 3:1 — about £160 of headroom to spend harder on growth, recouped in 5.1 months.
What to do about it
- Lift ARPA without new customers — annual plans, a higher tier, usage add-ons, or dropping a discount. The fastest LTV win there is.
- Cut the cost to deliver. Margin is the lever most teams never touch: cheaper infra, fewer manual support hours, leaner third-party APIs.
- Lower CAC with conversion work before cutting spend — a better landing page or onboarding drops CAC with zero extra budget.
Have you thought about the costs gross margin hides — card fees (~3%), refunds, and the support hours a cheap plan quietly eats? They belong in margin too.
Halve churn from 4% to 2% and LTV roughly doubles — and your affordable-CAC ceiling with it. Retention is an acquisition budget.
What is this chain telling me?
Read it left to right — each box is computed from the one before it. Your churn sets how long a customer stays (1 ÷ churn = lifetime). Lifetime sets LTV, the total gross profit one customer brings. Then LTV ÷ CAC gives the ratio: how many times over a customer repays what you spent to win them.
The benchmark is 3:1. Below 1:1 you lose money on every customer you acquire; above 5:1 you are being too cautious and leaving growth on the table. Your affordable CAC — the most you can pay per customer and still hit that 3:1 return — is just LTV ÷ 3.
That is why the flywheel spins. Healthy economics mean each customer funds more than one replacement, so you can afford to win more customers, who fund still more growth. Below 1:1 it stalls: the machine takes out more than it puts in, so every turn leaves you with less.
What counts as 'cost' here — and what doesn't
Gross margin here means revenue minus the direct cost to deliver what you sold: hosting and infrastructure, third-party APIs, payment-processing fees (often ~3%), and the support tied to serving a customer. Refunds and chargebacks belong here too.
What it does not include is fixed overhead — salaries, rent, tools, the team building the product. Those are real but they are not per-customer, so they live in your , not your unit economics. Keep them out of margin or every customer looks unprofitable.
Want to be stricter? Use — gross margin minus the variable cost of selling (the ad-to-sale fee, the onboarding hour). It is the most honest input to LTV: it counts everything that scales with one more customer and nothing that does not.
A loop compounds; a funnel just leaks
A funnel runs once and drops people at every step. A loop feeds its output back into its input — users invite users — so it turns over and over. How fast it turns is the viral cycle time, and that clock often matters more than the K-factor itself.
There are four classic loops: viral (users bring users), content (pages bring search traffic that makes more pages), paid (revenue buys ads that make revenue), and sales (revenue funds reps who close more revenue). All share one shape — a where the output becomes the next input — and one dial almost nobody tunes: , how long one turn takes. Halving it doubles how many turns you get in a year, which is why it is the exponent's clock speed.
K is read live from the virality panel above — 2.0 invites × 20%. Adjust it there; the loop re-times here.
So what does this mean for you?
A 100-user cohort, each bringing K = 0.40 over a 14-day cycle, snowballs to 167 people in 180 days — then flattens at its 167 amplification ceiling. Halve the cycle to 7 days and the same loop reaches 167: the ceiling arrives roughly twice as fast. Cycle time is the clock the whole thing runs on — shortening it usually beats squeezing out more K.
What to do about it
- Run a referral program — reward both sides, and trigger the ask at the peak of delivered value, not on day one.
- Open an affiliate / partner channel — pay creators or agencies per signup. It is a paid loop that scales on other people's content, and you only pay on conversion.
- Make outputs shareable — every export, public link or invite is a free impression that feeds the top of the loop.
- Shorten the loop, don't just widen it — cut the time from signup to first invite. Cycle time compounds faster than K.
Have you thought about an affiliate or creator channel? It is the most underrated B2B loop — other people produce the content and audience, and you pay only when it converts.
Funnel vs loop — and why cycle time wins
A funnel is a one-way drop-off: 1,000 visitors → 100 signups → 10 customers, and then it is done. To grow it you have to keep pouring new visitors in the top. A loop closes back on itself — those 10 customers refer the next batch of visitors — so the same effort keeps paying out, turn after turn. Channels that never close a loop have a hard ceiling; loops compound.
The decides whether a loop sustains (above 1) or merely your paid acquisition (below 1). But cycle time decides how fast either plays out. Two products with the same K grow at wildly different speeds if one loops weekly and the other quarterly — the weekly one gets thirteen times as many turns in a year. That is why mature growth teams obsess over shortening the invite → activation → re-invite loop, not just widening it.
Blended CAC flatters; marginal CAC decides
folds free organic into the denominator, so it understates the cost of the customer your next budget actually buys. Scale paid spend: organic stays fixed, paid grows, and blended drifts up to meet .
Every business mixes two kinds of customer: the ones you pay to acquire, and the ones who arrive free — word of mouth, search, referrals. Blended CAC averages the cost across both; paid CAC counts only the bought ones. The averaged number is the one people quote on stage, and it is the one that quietly lies to you the day you decide to scale spend.
So what does this mean for you?
Right now you spend £10,000 a month to win 24 paid customers, with 16 more arriving free. So each bought customer actually costs £417 — but your free organic customers pull the blended average down to £250, a £167 gap. Budget against the £417: the moment you scale paid, the free customers stop hiding it and your real cost surfaces.
What to do about it
- Diversify channels so no single ad auction can spike your blended cost overnight.
- Build a real organic loop (content / SEO, community, referrals) so blended sits below paid for a reason, not an accident.
- Judge every new pound on incrementality — a holdout or geo-test — not last-click attribution.
Have you thought about running a holdout before you trust an attribution dashboard? Most 'attributed' conversions would have happened anyway — incrementality is the only honest read.
Three rules
- Report blended to the board — it is the true cost-to-grow and dodges attribution fights.
- Budget on marginal / paid CAC — adding spend is judged by what THAT spend brings, not the average your free traffic drags down.
- Watch the gap close — if blended is creeping toward paid, your organic engine has stopped scaling and you are a paid-acquisition business now.
The classic blow-up: "our CAC is £40!" (blended) → triple paid spend → CAC "mysteriously" climbs to £90. It did not climb; you stopped hiding paid behind organic and hit saturation at the same time.
What am I looking at in this chart?
The x-axis is how much you scale paid spend — from 1× (today) out to 10×. The gold line is your blended CAC and the coral line is your paid CAC. The gap between them where they start is the discount your free organic customers quietly give you: they cost nothing to acquire, so they pull the average down.
As you scale, your organic numbers stay fixed while paid balloons, so the average gets dragged toward paid and the gold line climbs to meet the coral one. That is the whole lesson: report blended to the board, but budget on paid. Turning up saturation tilts the coral line upward too, because once the easy audience is spent each extra pound buys a pricier customer.
This is exactly how founders get blindsided. "our CAC is £40!" (that is blended) → they triple the budget → "why is CAC suddenly £90?". It did not jump. You simply stopped hiding paid behind free organic, and the average had nowhere left to hide.
The roll-ups that decide whether to step on the gas
Three numbers a board reads in one glance: is sales-and-marketing paying off, is the growth-versus-profit trade healthy, and how much cash each dollar of new ARR costs.
Zoom out from any single metric and a board asks three blunter questions. Is the money you put into sales and marketing coming back? Are you growing fast enough to justify the losses — or profitable enough to forgive slow growth? And how much cash does each new pound of recurring revenue actually cost? These three roll-ups answer all three at a glance.
The inputs are seeded from your Start-here numbers; the growth and margin figures are an example to swap for your own.
Net new this quarter over the prior quarter's spend.
Year-on-year growth rate plus profit margin. The sum should clear 40 — the .
Net cash burned over net new ARR — the , how much you spend to buy a dollar of recurring revenue.
So what does this mean for you?
A magic number of 1.10 says every £1 of sales and marketing brings £1.10 of new ARR — step on the gas. Your growth and margin sum to 45, clearing the Rule of 40. And at a burn multiple of 1.90 you spend £1.90 to add £1 of recurring revenue — fine.
What to do about it
- If the magic number is below 0.75, fix conversion and retention before adding sales-and-marketing spend — otherwise you just burn faster.
- Hold the Rule of 40 deliberately: choose to trade growth for margin (or back), don't drift into a bad mix.
- Track the burn multiple monthly. A rising multiple means each new pound of ARR is getting more expensive — catch it early.
Have you thought about whether your growth is efficient or just expensive? These roll-ups are exactly what catch the difference vanity metrics hide.
ARPA is the monetization term underneath all three — lift it and every roll-up improves at once.
When would I actually use these three?
The magic number answers one question: is more sales spend worth it? It is the new annual revenue you earn per pound of . Above 0.75 you can pour money in with confidence; below 0.5 you should fix your efficiency before spending more.
The is the trade-off referee: your growth percentage plus your profit-margin percentage should clear 40. You can spend hard on growth or bank the margin instead — either is fine as long as the two added together still hold the line.
The is the bluntest of the three: the cash you torch per pound of new recurring revenue. Under 1 is excellent; over 3 means you are leaking. Think of these as the dashboard lights — your , and are the parts under the bonnet.
No metric lives alone
Every number here pulls on the others — lower churn lifts LTV, which lifts your affordable CAC, which lifts growth. Improve one upstream metric and the effect cascades.
By now a pattern should be clear: every number you have moved pulls on the others. Churn changed your LTV; LTV changed your affordable CAC; activation quietly sat upstream of nearly everything. The table below is the cheat sheet — read each row as “improve the left, and the right moves with it” — and the four cards under it are the relationships worth committing to memory.
If this moves, watch that
| If this improves… | …these move with it |
|---|---|
| Churn ↓ | LTV ↑, LTV:CAC ↑, NRR ↑, affordable-CAC ceiling ↑ |
| Activation ↑ | retention ↑, referral ↑, LTV ↑ (all at once) |
| ARPA ↑ | LTV ↑, payback ↓, CAC ceiling ↑ |
| K-factor ↑ | effective CAC ↓, blended CAC ↓ |
| Paid spend ↑ (no other change) | blended CAC ↑, marginal CAC ↑, payback ↑ |
| Conversion rate ↑ | CPL ↓, CAC ↓ (free efficiency, no extra spend) |
Retention sets your acquisition budget
Churn is the denominator of LTV, and LTV ÷ 3 is roughly what you can afford per customer. A retention win is an acquisition win one step removed.
Blended flatters, marginal decides
Blended CAC hides paid behind organic. Scale spend and blended drifts up to meet paid while marginal CAC rises from saturation — they fire together.
Activation cascades
Activated users retain; retained users refer; referrals lower CAC. Activation sits upstream of all three, so one fix moves everything.
LTV:CAC and payback are different questions
LTV:CAC asks does the model work; payback asks how fast the cash comes back. Annual prepay fixes payback without touching LTV:CAC.
The whole chain in one breath
Activation (users reaching value) lifts retention, referral and LTV at once — fix it first. Lower churn and higher ARPA both push LTV up. Higher LTV raises your LTV:CAC, which raises your affordable CAC, which lets you buy more growth. Referral feeds growth directly and for free. Pull any one lever and the whole chain shifts — which is the entire point of the calculators above.
Every metric here has a lifecycle lever
The numbers move when you send the right message at the right moment. In Hogsend each lever is one durable TypeScript journey in your repo, triggered by your PostHog events — reviewed in a PR, not clicked together in a dashboard.
Every lever on this page is, in the end, a message sent at the right moment — a nudge when someone stalls, a check-in before they churn, an ask when they are delighted. That is what a lifecycle program is, and it is exactly what Hogsend runs. Below, each lever maps to a journey that ships in the scaffold; then a calculator puts a number on what the whole program is worth, using your figures from the top.
Onboarding that waits for the aha
A durable journey waits for the activation event, up to N days, then branches: nudge the stalled, advance the active.
See the onboarding use case →Win-back that knows when someone left
Spot the drop in product events and reach out before the renewal, not after.
See win-back →Failed-payment dunning
Reminders that sound human and stop the moment payment clears.
See dunning recipes →Expansion & usage-limit nudges
Catch accounts hitting a limit and offer the upgrade in-moment.
See conversion recipes →Referral & share prompts
Ask at the peak of delivered value, where the K-factor actually lives.
Browse recipes →Digests, anniversaries, NPS
Recurring touches that keep cohorts warm and surface who is slipping.
See retention recipes →A lifecycle program is worth the activations it adds plus the it removes. Plug in your funnel; the value flows straight through the same maths as above.
These touches are emails. Hogsend sends them from durable journeys with no per-contact billing, so widening lifecycle coverage does not widen the bill — your database just gets more rows.
So what does this mean for you?
With 40 new users a month, lifting activation +8 points turns roughly 3 more of them into customers, each worth £1,640. Trimming churn −1.0 points also lifts the value of everyone you keep, from £1,230 to £1,640. Together that is £12,300 a month — about £147,600 a year — from sending the right emails at the right moment.
How is this number worth that much?
A lifecycle program is worth two things. First, the extra customers it activates — the incremental activations it adds, each worth their . Second, the it removes — and because churn sits in the denominator of LTV = ARPA × margin ÷ churn, cutting it lifts the lifetime value of everyone you keep, not just the new ones. The calculator adds both pieces together and annualises the result.
These touches are just emails, sent at the right moment. In Hogsend each one is a durable TypeScript journey triggered by your product events — reviewed in a PR, with no per-contact billing. So widening your coverage (more touches, more cohorts, better ) doesn't widen the bill; your database just gets more rows.
Definitions are the easy part
The value is in the second column — what actually moves when this moves. Hover or tap any term for the formula.
- Fully-loaded means salaries, tools, overhead — not just media. The headline number people quote is media-only and flatters you.
- Always ≤ paid CAC because free/organic customers dilute the denominator. Good for a board headline, dangerous for budgeting.
- The real cost of a bought customer — the number that matters when deciding whether to scale a channel.
- Cost of the next customer. Rises with saturation. This, not average CAC, governs the should-I-add-budget decision.
- The multiplier that turns CPC into CPL into CAC. Improving it lowers CAC with zero extra spend.
- Churn sits in the denominator, so LTV is hypersensitive to it. Always use gross-margin LTV, not revenue LTV, or you overpay for customers.
- 5% monthly churn → a 20-month lifetime.
- ~3:1 healthy · <1:1 lighting money on fire · >5:1 you are under-investing in growth.
- Months to recoup CAC — the cash-velocity question, separate from LTV:CAC. <12mo strong for SMB, 18–24mo tolerable for enterprise.
- The GM% feeding LTV. If this is thin, good LTV:CAC ratios are an illusion.
- Counts heads, ignores their size.
- Caps at 100%. The true leakage rate, no upsell masking. Best-in-class >90%.
- The single most important SaaS number. >100% = you grow even if you acquire nobody. Elite is 120%+.
- Growth quality. >4 = efficient; ~1 = treading water.
- A flattening curve = product-market fit; a curve that decays to zero means no amount of CAC will fix it.
- Should sit where value-to-user meets revenue. Bad NSMs (signups, pageviews) drive the org off a cliff.
- Add 7 friends in 10 days; send 1st invoice. Find it by comparing retained vs churned cohorts.
- Upstream of everything — fixing it lifts retention, referral and LTV at once.
- >20% decent, >50% exceptional for consumer/social.
- Real only when tied to activation/retention. Treat raw totals as noise.
- K>1 = self-sustaining exponential growth. K<1 still amplifies paid but decays without a feed.
- K=0.5 → every paid cohort effectively doubles.
- Halving cycle time matters more than raising K — it is the exponent's clock speed.
- Paid, viral, content/SEO, sales. The compounding engine versus the leaky funnel.
Ship the lever, not just the metric
The scaffold ships 10 journeys and 13 templates — onboarding, win-back, dunning, digests — every one a durable TypeScript file you can review in a PR.
Free to self-host · One scaffold command · No per-contact billing
pnpm dlx create-hogsend@latest my-app