Track the Journey from Source to Customer

01

Start here after your audit

Use this when you know how many people visit but cannot connect inquiries to outcomes. Start with a single service, a defined period, and the system where the team records real jobs. Write down what each count means before adding tags. Website analytics and the lead record will often have different coverage.

02

The problem

Traffic and ad-platform conversions cannot show which inquiries become useful leads or customers. At the same time, copying contact details into analytics creates an unnecessary privacy risk.

03

Why it matters

A small, consistent measurement chain is more useful than a large event catalog nobody trusts. Connect stages inside your protected business systems using a stable internal reference; keep contact details out of website analytics.

04

Common mistakes

Reporting breaks when definitions change between tools.

• Treating button clicks as completed inquiries
• Using different names for the same action
• Losing source data when a lead enters follow-up
• Sending email addresses, phone numbers, or form text to analytics
• Changing several campaigns before enough outcome data is available

05

Instrument the minimum journey

Write the definitions before implementing tags.

• Choose one primary website action
• Record a completion event only after confirmed success
• Preserve the source label and internal lead reference in your protected lead system
• Record qualified and customer outcomes in the system of record
• Review counts and sample records before calculating rates

06

Example scenario

Illustrative example · Not a client case study

A repair business records a confirmed booking event after the booking succeeds, preserves a non-personal source code with the lead record, and marks the eventual job outcome in its system of record. Its monthly review can compare sources without placing names, phone numbers, or form text in analytics.

07

Use the report to decide

Review the journey by source over a consistent period. Keep volume, qualified leads, customers, cost, and data quality visible together. Mark missing or delayed data instead of silently treating it as zero.

08

Define five columns before adding a dashboard

A protected working table can start with source, received date, internal lead reference, qualification status, and customer outcome. Keep the person and contact details in your lead system. A random identifier linked to a customer record is pseudonymous, not automatically anonymous or suitable for every analytics platform.

• Visits: define which pages and session definition you count, using the same analytics setup over the comparison period.
• Confirmed inquiry: the form or booking system accepted a real request. A contact-button click is an earlier event.
• Qualified lead: use your own documented rule, such as an offered service in your area with a reachable contact.
• Customer: a job booked or paid, according to one definition used consistently by the team.
• Source: retain a known campaign/source label when available; otherwise record unknown. A 'How did you hear about us?' answer is self-reported evidence, not equivalent to browser attribution.

09

Read an illustrative source report

Suppose one source has 200 measured visits, 10 confirmed inquiries, 6 qualified leads, and 2 customers after the same inquiry group has had enough time to decide. Inquiry rate is 10 / 200 = 5%; qualification rate is 6 / 10 = 60%; qualified-lead-to-customer rate is 2 / 6, about 33%. These fictional numbers explain the calculation, not what your business should achieve.

• Keep the raw counts next to each rate. Two customers are too few to justify a broad claim that this source always performs well.
• Do not divide customers from last month's leads by this month's new inquiries. Follow the same inquiry group through its outcomes.
• If a denominator is zero or missing, leave the rate unavailable. Unknown is different from zero.
• Add channel cost only if it covers the same scope and period; distinguish advertising spend from total acquisition cost before comparing channels.

10

Reconcile records and protect the boundary

Trace a permitted, clearly marked test from entry page to confirmed inquiry and the internal outcome record. Then inspect a few real records inside the authorized system. Expect gaps from consent choices, blocked tags, phone calls, cross-device visits, or delayed outcomes; document them rather than forcing the totals to agree.

• Check that a click, validation error, or page refresh does not create an extra confirmed inquiry event.
• Compare the website's acceptance event with the stored lead. Check whether duplicates and test records are excluded consistently.
• Inspect event names, parameters, page URLs and query strings for accidental contact details or form text before enabling analytics.
• Do not send email addresses, phone numbers, customer names, or free-text requests to Google Analytics. Keep record joins in your protected business systems unless the destination's rules and your data controls explicitly support them.
• Review missing source and missing outcome counts each month. Repair the largest gap before making decisions from small differences between channels.