A cheap lead is only useful if it has a reasonable chance of becoming a customer. When sales happen by phone, in person or after several follow-ups, the ad platform may see the form submission but miss the outcome.
Sending those outcomes back from your CRM changes the information available for campaign decisions. It's the approach we use to connect CRM sales records with Meta advertising. This guide covers why that matters, how the connection works in principle and how to keep the data trustworthy.
Why Cost Per Lead Can Point You the Wrong Way
Cost per lead is easy to see in the ads manager, so it often drives decisions. It can mislead when lead quality differs between campaigns.
| Campaign | Spend | Valid leads | Paid sales | Cost per lead | Media cost per sale |
|---|---|---|---|---|---|
| A | A$3,000 | 300 | 15 | A$10 | A$200 |
| B | A$3,000 | 150 | 30 | A$20 | A$100 |
Made-up figures to show the arithmetic. They are not Evolve client results.
Campaign A looks twice as efficient on cost per lead. Campaign B produces twice as many sales for the same spend. The comparison only holds if both groups of leads come from the same period and use the same attribution window, and you'd still need revenue per sale and gross margin before calling either campaign profitable.
Connect the Lead Record to the Eventual Sale
Closing the loop starts in your own systems:
From there, use your CRM's currently supported Meta Conversions API integration for eligible CRM and offline events. Check current Meta documentation and your CRM's integration before following an older tracking walkthrough.
Conceptually, each event carries:
Share the minimum needed. Never send sales notes or sensitive personal information as marketing data, and make sure your privacy policy and consent practices cover what you share.
Keep Reporting and Optimisation Separate
Two different questions often get mixed up:
Sending sales data doesn't by itself change what a campaign optimises for. Meta's lead generation guide describes connecting CRM data through the Conversions API alongside a conversion leads performance goal for suitable instant form campaigns. It's a 2023 overview rather than current setup instructions, and what's available for a particular campaign or account changes, so check before planning around it.
Be wary of simple claims in either direction. Lead campaigns don't necessarily chase the cheapest leads, and sending conversion data doesn't automatically teach the platform to find buyers just like yours.
Make the Sales Data Dependable
Campaign decisions are only as good as the events behind them. Check:
Reconcile your CRM and accounting totals separately from the sales the platform attributes. Cancellations and refunds belong in your own net-sales reporting, whatever the platform shows.
Use a Scorecard the Sales Team Recognises
A shared scorecard keeps marketing and sales looking at the same numbers:
| Measure | Notes |
|---|---|
| Spend | Advertising spend for the period |
| Valid leads | After spam, tests and duplicates are removed |
| Qualified leads | Using the sales team's definition |
| Paid or net sales | Net of cancellations and refunds |
| Media cost per sale | Advertising spend divided by the relevant sales |
| Revenue and contribution margin | With the included costs written down |
Media cost per sale isn't the full cost of winning a customer. Full acquisition cost also includes defined sales, agency and system costs. Keep the two separate and label them clearly.
Compare mature cohorts, meaning groups of enquiries that have had time to buy, and use the same windows and attribution definitions each time. Campaigns can claim overlapping credit, and the return on ad spend a platform reports isn't profit, or proof that the ads caused the sales.
Change Decisions, Then Measure the Effect
Once the sales data is reliable, use it to guide budgets, creative, targeting and which leads the sales team should prioritise. Check the volumes before making big changes; a handful of sales is too few to judge a campaign.
Improvements after a change can have other causes: sales training, seasonality, a new offer, a different attribution setting or a change in cancellations. Where it's feasible, a controlled test or holdout gives a better read on what the ads actually added.
Start with one sales event and one reliable data flow. Check it against your own records before expanding. Most of the work is coordinating marketing and the CRM, not buying another tracking tool.