Four Free Tools That Show Where Your Sales Come From

You know something worked last month. Money came in — orders, leads, both — and you have a guess about where it came from. Meta says it drove those sales. Google says the same thing about the same people. Your gut says it was the one reel you almost didn't post.

The first thing worth knowing is that a dashboard can look healthy while measuring nothing. I run a self-hosted first-touch attribution tool. It loaded on every page of my site, recorded pageviews correctly, charts filling in — and nothing in my code had ever called its conversion function. Pageviews record themselves once the script is in. A conversion only records if somebody wired that call into your checkout or your form, and if nobody did, the tool never learns a sale happened.

Below are the four free tools I'd set up first, in the order I'd do them. This is part of the "Where Sales Come From" mini-series: your site and your checkout already produce these numbers, and reading them is the missing part.

Play 1: Find Out Which Ad Made the Sale (UTM + Google Analytics)

Every ad platform takes credit for every sale. Run Meta, Google, and TikTok the same week and all three dashboards report the same buyer. Ruler Analytics measured platforms collectively claiming 30 to 50 percent more conversions than happened: 175 reported against 95 real sales.

The fix:

1. Go to Google's Campaign URL Builder. Free, about 30 seconds a link.
2. Build one tracking link per ad. Set the source (meta, google, tiktok), the medium (paid-social, cpc), and the campaign name.
3. Put that link in the ad instead of your plain homepage URL.
4. Connect Google Analytics, also free, so the clicks land somewhere you can read.

In GA4, read first-touch (where the buyer first heard of you) and last-touch (what pushed them to buy); they're often different channels.

A short redirect keeps those links readable. Point /go/story at the real page with the UTM parameters attached, and the link in your bio stays clean while your numbers still know which channel sent the person.

Then verify the part most people skip. Open your own site in a private window, click through your tracking link, and complete the action you care about — a test order, a form submit. If that conversion doesn't show up in GA4 within a day, the tag is loading but the event isn't firing, and every report built on it is fiction.

If what you're counting is leads rather than sales, set the recorded revenue to $0 instead of guessing at what a lead is worth. A made-up dollar value poisons every ROAS number downstream, and months later you won't remember which figures were invented.

Safety note: Build the UTM links before you launch. Swapping them mid-campaign splits your history.

Play 2: Watch Where People Give Up on Your Site (Microsoft Clarity)

You can see that people left the page. A 70% bounce rate won't say whether the button failed on their phone or the form got cut off.

Session recordings and heatmaps show you the part the number leaves out. You watch a cursor stall over something that looks like a button and isn't one. You see the scroll stop two screens above your call to action. You see rage clicks, the same spot hit four times fast, which almost always means something looked clickable and did nothing. And you see all of it at the phone screen sizes where most of it breaks.

The fix:

Microsoft Clarity is free — "Free forever" right on its own homepage. It records real, anonymous visitor sessions and builds heatmaps of clicks and scroll depth.

1. Create a free Clarity account.
2. Add the tracking code to your site. There's a WordPress plugin if that's where you are.
3. Take your weakest page — lowest conversion rate, highest exit rate — and watch five sessions back to back.

Clarity anonymizes the data and is GDPR and CCPA ready.

Two things to know before you lean on it. A recording is worth nothing until somebody watches it, so put twenty minutes on the calendar the same day you install the tag. And choose your five sessions on purpose — filter to visitors who reached your product or contact page and left without acting. Five sessions of people who were never going to buy teach you nothing.

Play 3: Turn Your Charts Into One Clear Action (Plausible or PostHog + AI)

You open the dashboard, see fifty numbers, and close it again without deciding anything. Next week, same fifty numbers.

The fix:

PostHog has a free plan, no credit card, 1 million events a month included. Plausible gives you a 30-day free trial, also no card, built simple on purpose.

Then screenshot the dashboard, open ChatGPT or Claude, paste the image, and ask:

"What's the one thing I should fix on my site this week based on this data?"

You get back one thing to do, which beats fifty metrics to interpret.

Give the model more than the picture: what the page is supposed to do, what you changed in the last month, and what a decent week looks like for you. Then hold it to one answer. If it comes back describing your numbers instead of naming a change, ask again — "name the single change, and tell me which number should move if it works." That second half is what makes next week's check possible.

Once a week is enough.

Play 4: Find Out Which Product Makes You Money (Spreadsheet + AI)

Ask most store owners which product makes them the most money and you'll hear the name of the one that sells the most. Volume is the number already on the dashboard. Profit per product takes ten minutes of arithmetic, and it usually changes what you promote.

The fix:

Open a spreadsheet. One row per product. Columns for price, product cost, shipping, platform and transaction fees, and the ad spend it takes to sell one. Pull the cost numbers from your supplier's API or your payout reports rather than from memory.

Estimates are the trap here. A pricing doc agrees with whatever you typed into it, so take the numbers from a source that can disagree with you — the Printify or Printful API, your Shopify and Etsy payout reports, the fee lines on your Stripe statements. Two things usually surface the first time somebody does this properly: the same product listed at two different prices in two places, and live variants sitting under the margin floor you thought you'd set. Both are invisible in a document and obvious in a spreadsheet full of verified costs.

Then paste the data into ChatGPT or Claude and ask:

"Which product makes me the most profit? Which is breaking even? Which should I cut or reprice?"

Your best seller can turn out to be the one paying you least once fees and ad spend sit in the same row. When you reprice, check every live variant afterward, not only the ones you edited. Size and finish options carry their own costs, and one of those is usually the reason a product looked fine at the top level.

If you set up one of these this week, make it the UTM links. Clarity, PostHog, and the spreadsheet all work on data you can gather later. Untagged clicks are gone — you can't go back and label last month's traffic.

Quick Recap

• Your dashboard can look healthy while measuring nothing. Mine recorded pageviews for months and never one conversion.
• UTM links plus Google Analytics tell you which source sent the buyer. Build them before launch.
• Microsoft Clarity records real sessions for free. Five sessions on your weakest page is enough to find one fix.
• PostHog and Plausible give you a dashboard you'll open, and a screenshot plus one AI question turns it into one action.
• Pull real product costs before you trust a margin. The identical product is often listed at two different prices in two places.

Start Here

If you can name the four tools but not where to wire them into what you already have, that's the gap.

At daisyguti.ai/work-with-me, there's an AI intake assessment that maps where your business stands before any call. It takes a few minutes and you get a clear read on what to measure first. Daisy is a 20+ year engineer who builds these systems for small business owners.

Sources

1. Google Campaign URL Builder - https://ga-dev-tools.google/campaign-url-builder/
2. Microsoft Clarity (free heatmaps and session recordings) - https://clarity.microsoft.com
3. PostHog free pricing - https://posthog.com/pricing
4. Plausible subscription plans - https://plausible.io/docs/subscription-plans
5. Ruler Analytics: How Double-Counting Conversions in Ad Platforms Skews Your Reporting - https://www.ruleranalytics.com/blog/reporting/conversion-duplication/