Lead source attribution: linking source to sales outcome

Short answer

To link lead source to the sales outcome, two fields on every lead record must stay fixed: the first source (where the person came from first) and the outcome (won, lost, close reason, amount). The report then groups leads by the month they were created and shows four figures per source: leads, won leads, win rate and the main loss reason. That table replaces "which channel brings the most leads?" with "which channel brings the most sales?"

Why lead counts mislead

Marketing reports usually stop at lead count: "Instagram brought 300 leads this month, the website 80". The sales outcome lives elsewhere — in the CRM, in payments, in contracts. When the two are not joined, budget goes to the channel with the most leads, even though its leads may buy the least.

The reverse happens too: a channel with few leads — referrals or corporate enquiries, say — brings a large share of sales, but it does not show up in the report and nobody gives it time.

First source, latest touch and what "source" means

A customer saw an ad on Instagram, looked at the website, wrote on WhatsApp and finally called. What is this lead's source? There are three answers:

• First source: the channel or campaign of the person's first enquiry. For judging new-customer acquisition.
• Latest touch: the last channel before the sale. For the question of "who closed it".
• Multi-touch models: split credit across all touches. They need precise data and rarely give reliable results in a small business.

The report table

A source report is a single table. Each row is a source, and the columns are:

1. 1LeadsLeads created in the chosen month — deduplicated, counted as unique people.
2. 2WonOf those leads, the ones closed as won so far.
3. 3Win rateWon / leads. Show leads still open separately.
4. 4Average amountIf amounts are recorded — some channels bring fewer but larger sales.
5. 5Main loss reasonThe most frequently chosen reason among lost leads.

Grouping by creation month

The most common calculation mistake is this: "200 leads came in October, there were 30 sales in October, so 15%". Some October sales came from September leads, and some October leads will be sold in November. The right way is to group leads by the month they were created and follow their outcome: "of October's leads, how many have been sold so far?"

For the same reason, the latest month's win rate will always look low — its leads are still being worked. Say so plainly in the report, or every month will conclude that "last month was bad".

Be careful with small numbers

If a channel brings 12 leads a month and 4 are sold, 33% looks great. But next month 1 of 12 may sell and the rate drops to 8%. With small numbers, percentages swing by chance. Decide on at least three months combined, and always show the lead count next to the percentage.

Other factors that separate source from outcome

• The manager: if every lead from a channel goes to one manager, the channel's result is the manager's result.
• Speed: if night-time leads wait until morning, the problem may be response time, not the channel.
• The offer: different campaigns on the same channel sell different products.
• Offline closes: the sale happened in the shop but was never marked "won" in the CRM — the channel looks weak for no reason.

Read it together with loss reasons

The win rate answers "how much"; the loss reason answers "why". If Instagram leads have a low win rate and the main reason is "price", the problem is not the channel but the expectation the ad creates: the ad does not show the price and the customer expects something cheaper. A short, mandatory reason list is decisive here — see CRM lead statuses.

An illustrative example

This is an illustrative example. A furniture company groups three months of leads by creation month. Instagram brings the most leads, but its win rate is 6% and the main loss reason is "price". The website form brings a quarter as many leads with an 18% win rate. Referrals are only 25 people, but their average amount is the highest.

Decisions: price ranges are added to Instagram ads, a dedicated "who recommended us?" question is written for referred customers, and response time on the website form starts being tracked separately.

Common mistakes

• Overwriting the source with every new enquiry.
• Dividing a month's sales by that month's leads.
• Counting duplicate leads separately — the channel's lead count swells.
• Not recording offline sales in the CRM.
• Shutting a channel down on small numbers.

Limitations

A source report shows correlation, not cause: if one channel's leads buy more, that may be due to the channel, the audience or the season. Not every touch can be tracked, and the quality of the source field depends on the data flow at intake. Use the report to make decisions, but confirm them with a small test. How campaign data reaches the record is covered in marketing campaign to CRM data flow.

Source and outcome in Vexvon

In Vexvon every lead is recorded with the channel it came from, and if an open lead exists for the same number, a new enquiry does not create a new lead — which keeps per-channel counts free of duplicates. A lead is closed as won or lost, a lost lead gets a reason (price, not ready, competitor, other), and the customer's history across all channels is kept in one timeline. A company can extract its own field such as "how did you find us" from the conversation. We do not present a ready-made revenue attribution report. More on Vexvon CRM.

Next step

Group the last three months of leads by creation month and first source, and fill in the five columns for each source. If the ranking by leads differs from the ranking by sales, you have found the most important table for your budget discussion. Joining conversation data with the CRM is covered in conversation analytics and the CRM. More articles are in the lead management section, and we can build your report together during a demo.