Which Marketing Channel Brought That Chat Lead? A Simple Approach to Attribution

You run ads, post on social media, appear in search results, hand out flyers and get referrals. A lead arrives through the chat on your website. Which of those efforts earned it? Working that out is called attribution, and it matters because it tells you where to spend the next hour or the next pound. Big companies use complicated models and dedicated tools. A small business can get most of the value from three habits: asking people how they found you, tagging links so your analytics can see them and keeping a simple lead sheet that records the source. This guide explains what attribution is, why it is imperfect, how to collect source information from chat leads, how to use UTM tags, how to compare channels fairly, which simple models to use and what to avoid. It is honest about Crescelo's role: it captures leads and the conversations behind them, and lets you choose form fields, so you can ask 'How did you hear about us?', but it does not automatically track advertising campaigns or marketing sources for you.

Why attribution is hard

• Customers use several touchpoints: they see an ad, search your name, read a review and then chat.
• Browsers and privacy tools limit tracking.
• Word of mouth and offline sources leave no digital trail.
• The last click is not necessarily the reason they chose you.

Perfect attribution does not exist. The goal is a good enough picture to decide where to invest.

Method 1: just ask

The simplest and most useful method is to ask the customer. Add a field to your lead form, or have the assistant ask naturally after a helpful answer: 'By the way, how did you hear about us?' Offer common options plus 'other'.

Self-reported sources are subjective, but they capture what analytics cannot, such as recommendations and offline exposure. Crescelo lets you choose lead form fields, so you can add this question and see the answer with each lead.

Method 2: tag your links with UTM parameters

UTM parameters are labels added to a link so analytics tools can see where a visit came from. For links you control, such as ads, newsletters, social posts and your bio link, add them.

Example: yoursite.com/book?utm_source=instagram&utm_medium=social&utm_campaign=spring-offer. Your website analytics then groups visits by these labels. Keep naming consistent: lowercase, no spaces. Note that Crescelo does not read UTM tags into leads itself, so pair this with the 'how did you hear' question, or compare chat volume with visits from each source in your analytics.

Method 3: give channels their own landing pages

A dedicated page for each major channel, such as a hosted page for your Instagram bio link, lets you count visits and chats per channel. Crescelo's hosted pages let you create a page with its own assistant and link; using a separate page per channel makes the source clear from where the visitor started. See Instagram and TikTok bio links that lead straight to booking.

Method 4: a lead sheet

Crescelo can send leads to Google Sheets, which makes this straightforward. Add the outcome and value columns yourself as jobs are won.

Simple attribution models

For most small businesses, the self-reported source combined with tagged links is enough. Do not chase complex models before you have a reliable lead sheet.

Compare channels fairly

• Divide leads by cost: cost per lead for each paid channel.
• Go further and compute cost per won job, using the outcome column.
• Consider lead quality: a channel with fewer, better leads may be more valuable.
• Allow for time: some channels take weeks to show results.
• Value free channels: search visibility and referrals cost time, not ad spend.

The numbers above are illustrative. Note how the lead count alone would hide the fact that referrals convert best.

Privacy and consent

• Analytics and advertising cookies need consent in many places.
• Explain why you ask how someone found you; it is optional.
• Do not collect more than you need.
• See privacy law for lead capture.

Common mistakes

• Trusting last click alone.
• Ignoring offline and referral sources.
• Inconsistent UTM names.
• Never recording outcomes, so you count leads but not customers.
• Assuming the platform tracks campaigns when it does not.
• Over-engineering before you have data.

A worked example

A cleaning company asks each chat lead 'How did you hear about us?' and records the answer in a Google Sheet. After three months, the sheet shows 30 leads from search, 18 from social media, 11 from friends and 6 from flyers. Recording the outcome shows that referrals were won far more often than the others, while social leads mostly asked for prices and never replied. The owner decides to add a small thank-you offer for customers who recommend a friend, cut back on social advertising and improve the services page for search visitors. None of this needed special software, only a form question and a spreadsheet kept up to date.

These are illustrative numbers. Your own table will look different, and that is the point: it shows where your next hour or pound is best spent.

Frequently asked questions

See the lead capture feature, pricing and read chat analytics for service businesses.

Further reading

• Google Search Central: creating helpful, people-first content: What Google looks for in useful pages.