Pageviews and follower counts feel like progress but tell you nothing about whether anyone will pay. Replace them with five revenue signals you can set up in an afternoon on a free or cheap stack, then run them through a weekly keep-kill-fix loop. That loop, not a mood board, gets you to your first 100 paying users.
Why vanity metrics fail before your first paying user
Pageviews and follower counts feel like progress because they move every day. But they measure attention, not intent, and attention does not pay for your hosting bill. A founder can hit 5,000 pageviews and zero signups, or 200 pageviews and three trials, and only one of those numbers tells you anything about revenue.
Watch signals that connect to money instead: signup conversion rate, activation (did they reach the moment your product delivers value), trial-to-paid rate, churn, and where paying users came from. We break down five of these in our pre-100-users signal guide.
You do not need a paid stack to see them. PostHog's free tier includes 1 million events per month, which covers every revenue event a pre-traction app generates (PostHog pricing).
Which vanity metrics are lying to you right now?
The three most common fake-progress numbers
Pageviews, follower counts, and upvotes feel like progress because they move every day. But none of them touch your Stripe dashboard. A visitor who bounces in four seconds and a follower who never clicks your link both inflate the number without adding a single paying user.
The reason they fail as predictors is simple: they measure exposure, not intent. Revenue signals measure whether someone took an action that costs them something, like signing up, activating, or pulling out a card. Exposure metrics can't tell you which channel produces buyers and which produces lurkers.
If you're checking five numbers each morning and none of them is revenue, you're running a mood board, not a dashboard. The fix is swapping them for signals that map to your funnel, which we cover next, and ranking your channels by buyers instead of eyeballs (see 7 signals that reveal your best channel).
What are the five revenue signals worth tracking instead?
Signal 1: activated users, not signups
Pick one action inside your app that predicts someone getting value: first project created, first file imported, first automation run. Activated users = signups who did that thing. Signups alone flatter you; activation tells you if the product works.
Signal 2: time-to-first-value
How many minutes between signup and that activation action? If it's over a day, your onboarding is leaking money. Track it as a simple median in a spreadsheet or your analytics tool.
Signal 3: paywall conversion rate
Of users who hit your upgrade screen, what percentage pay? This is the purest product-market signal you own. A 3% rate on 200 paywall views beats a 0.5% rate on 5,000.
Signal 4: revenue per channel, not traffic per channel
Tag every signup with its source (Reddit post, X thread, SEO page). Then divide MRR by signups per channel. Pull MRR directly from your payment provider's reporting endpoint, like Stripe's subscription and revenue reporting API, so the number is cash, not clicks. We break down the per-channel version of this in our guide to signals that reveal which channels to keep.
Signal 5: returning usage in week two
What share of activated users come back 7-14 days later? Retention in week two predicts whether revenue compounds or leaks.
| Signal | Definition | Healthy early sign |
|---|---|---|
| Activated users | % of signups completing your core action | Rising week over week |
| Time-to-first-value | Median minutes to activation | Under 30 minutes |
| Paywall conversion | Payers ÷ paywall viewers | 2-5% |
| Revenue per channel | MRR ÷ signups, per source | One channel clearly above the rest |
| Week-two return rate | Returning activated users ÷ activated | 30%+ |
Illustrative benchmarks, not gospel. Your baseline matters more than anyone else's number.
How do you set up revenue tracking in one afternoon (for under $30)?
The minimum stack: analytics, events, payments
Three tools cover it. PostHog free tier for analytics and event tracking, Stripe (free, you pay per transaction) for payments, and a Google Sheet for the decision log. Total fixed cost: $0. Every event you need maps to a Stripe webhook or a PostHog call you can paste into a vibe-coded app in minutes.
The mistake is installing a dashboard before you define the five events that matter. Write them down first: signup, activation (the action that predicts retention in your app), first payment, repeat payment, and churn. Then wire each one. In PostHog, that is one posthog.capture() call per event.
Here is why the definition step matters more than the tooling step. Most founders open their analytics tool first because it feels like progress, then spend the afternoon staring at an empty event stream with no idea what to log. The order is backwards. When you define the five events on paper before touching any tool, every integration decision becomes mechanical: does this button press map to signup, activation, or payment?
If yes, capture it. If no, skip it. That single constraint is what turns a vague "set up analytics" task into something you can actually finish between lunch and dinner, because you are no longer deciding what to measure while you measure it.
In Stripe, webhooks fire on checkout.session.completed and customer.subscription.deleted whether you ask or not. If you shipped with Lovable or Cursor, paste the PostHog snippet into your layout file and ask the AI to fire events on the two or three buttons that count. An afternoon is enough because the stack is small, not because you are cutting corners. If you want the full event list mapped to each signal, we broke it down in 7 performance tracking signals before 100 users.
Skip the data warehouse, skip the BI tool, skip anything with a seat price. Five events, one sheet, one loop.
How do you turn signals into a weekly keep-kill-fix loop?
Numbers only matter if they force a decision. Eric Ries draws this line in The Lean Startup: vanity metrics look good and change nothing, actionable metrics tie directly to a decision you can act on.
So run a 30-minute Friday review. For each of your five signals, pick one verdict: keep (growing, do more), fix (flat, one experiment next week), or kill (dead for four straight weeks, stop touching it). One experiment per fix. One channel cut per kill. Write the verdicts down before you look at any dashboard, then check the numbers against what you expected.
The trap is tracking without deciding. If a signal has not changed a single action in three weeks, delete it and pick a sharper one. The loop is the product; the dashboard is just fuel. For a deeper version of this discipline, see 7 signals your B2B growth systems are actually working.
Worked example: from eyeballs to revenue on a weekend-built app
The before dashboard
Say you shipped a habit tracker in Replit over a weekend. Your dashboard shows 1,400 pageviews, 320 signups, and 87 Twitter followers. You check it every morning. None of those numbers can tell you whether anyone will pay you next month. You feel busy but you're flying blind.
The after dashboard
Swap in five rows on the same afternoon: 11 activations (users who logged a habit 3+ times), 2 paying users at $5/mo, 1 of 6 onboarding emails opened, 14 clicks from a single Reddit comment, and a week-over-week retention of 9%. Now the picture is brutal and useful: signups are fine, activation is broken.
That points your week's one move: fix the onboarding email, not post more. The keep-kill-fix loop runs on numbers like these, not vibes.
Can a tool like HeyCatch run this loop for you?
HeyCatch fits the loop's execution half. It audits your product, runs daily organic moves across Reddit, X, LinkedIn, TikTok, SEO, and email, and adapts a weekly roadmap, so the keep-kill-fix decisions have fresh signals waiting every Monday. It starts at $29/mo (HeyCatch pricing), built for solo founders, not funded teams.
Running HeyCatch through the same criteria honestly
Same test, same honesty. HeyCatch does not manage paid channels like Metaflow AI does, and its funnel analytics are thinner than a proper stack. If you need dashboards beyond signup-to-payment tracking, you will still want PostHog next to it. The beta-tester first-30-days claim is early data, not proof at scale. And it will not decide your ICP for you; the weekly roadmap adapts to the revenue signals you feed it, so the five metrics from earlier sections remain yours to own.
What should you do today?
Open your analytics and answer one question: how many visitors this week reached the step where they'd pay you? If you can't find that number, that's your afternoon task from the tracking setup section.
Then run one keep-kill-fix decision on whatever the number says. Keep what moved someone toward paying, kill what only moved pageviews, fix the one step where people drop off before signup.
Do it again next Monday. Five weeks of that beats three months of watching follower counts.
Frequently Asked Questions
What are revenue signals vs vanity metrics for a solo founder?
Vanity metrics measure exposure: pageviews, followers, upvotes. Revenue signals measure intent: activation, paywall conversion, churn, revenue per channel. The difference is whether the number touches your Stripe dashboard. A follower who never clicks inflates nothing that matters; a user who pays $5 tells you your product works and where they came from.
How do I track MRR on a micro SaaS with no budget?
Pull it straight from your payment provider. Stripe's reporting gives you subscription revenue without any paid tooling, and PostHog's free tier covers event tracking with 1 million events per month. Log the number weekly in a Google Sheet alongside signups per channel. Total fixed cost: zero dollars.
What metrics matter before I have 100 users?
Five: activated users, time-to-first-value, paywall conversion rate, revenue per channel, and week-two return rate. Skip aggregate traffic entirely at this stage. With under 100 users, every signup is individually inspectable, so the question is not how many came but whether each one reached the moment your product delivers value.
Do pageviews matter at all for a SaaS?
Only as a denominator. Pageviews tell you what it costs in attention to produce a payer, which is useful when comparing channels. On its own, a pageview count predicts nothing: 5,000 views can convert to zero trials while 200 views produce three. Track it, but never let it be the number you act on.
How do I know which marketing channel is actually making me money?
Tag every signup with its source, then divide MRR by signups for each channel. The winner is rarely the loudest one: a single Reddit comment can out-earn a month of posting on X. Rank channels by buyers, not clicks, and kill any channel that has produced zero payers for four straight weeks.