End-to-End Marketing Analytics for Small Business Without Expensive Tools: an n8n Setup

End-to-end analytics means seeing the full path of money: from an ad impression to a paid deal in the CRM. Attribution platforms do this for hundreds of dollars a month. For a small business that is often more than the value they return.

The good news: for 90% of small-business needs, you can assemble end-to-end analytics from what you already have - ad accounts, a CRM and Google Sheets - connected through n8n. Here is the architecture.

What end-to-end analytics means in plain words

Regular analytics answers "how many clicks and leads did we get". End-to-end analytics answers "how much money did each channel bring". The difference is fundamental: one channel can deliver cheap leads that never buy, another - expensive leads that pay back five times over. Without attribution you cannot see that, and you cut the wrong budget.

To build it you need to connect three data sources: spend (ad accounts), leads with source tags (site and forms), and money (CRM or sales records).

Why not a paid platform

Attribution suites are great products for mid-size companies with a dozen channels and heavy volume. For a small business they typically bring three problems: a subscription that eats a visible share of the ad budget; long connector setup for your particular zoo of services; and overkill - you pay for 100 reports and look at two.

An n8n setup covers those same two reports with no subscription: self-hosted n8n is free and charges nothing per operation, and Google Sheets is free entirely.

The architecture

1. Traffic tagging. Every ad link gets UTM parameters - that is discipline, not technology. Every lead form stores its UTMs and passes them into the CRM with the contact. This is the only place where the site or forms usually need a small change.

2. Spend collection. Once a day n8n pulls spend, impressions and clicks per campaign from the ad accounts via API and stores them in a data sheet.

3. Revenue collection. The same n8n pulls deals from the CRM: amount, status, source (the UTMs from step 1). No CRM? A sales spreadsheet works too, as long as it records the customer's source.

4. Merge and math. A JavaScript node joins spend and revenue by channel/campaign and computes the key metrics: CPL, CAC, ROMI - per channel, weekly and cumulative.

5. The dashboard. Results land in a simple Google Sheets dashboard: rows are channels, columns are spend, leads, sales, ROMI. Plus a short AI summary in plain language delivered to Telegram or email: "channel X paid back 3.2x, channel Y is negative for the third week - check the creatives".

What to understand before you start

1. Accuracy will not be perfect - and that is fine. Some leads arrive untagged (calls, DMs). For decisions like "which channel to scale", the 85-90% accuracy this setup delivers is enough.
2. The hard part is tagging discipline, not technology. One untagged campaign - one hole in the report. Fixed with link templates and a validation check in the same n8n.
3. Start with two reports, not twenty. Spend-leads-sales per channel weekly + month-over-month dynamics. Add more only when the data raises questions.

When this setup stops being enough

Honestly: when you get call tracking with heavy call volume, a dozen channels with daily budget reallocation, and a team that needs access rights and interfaces. That is the level where a paid platform starts to pay for itself. Below it - which is most small businesses - the n8n setup solves the problem for the price of a server.