Almost every marketing report that lands on our desk for a Manáry diagnostic looks the same: impressions, clicks, cost per click, leads, and cost per lead. Clean, organized, with a trend chart. And that's where it ends. The question the business owner actually asks, "how much of this spend turned into sales?", goes unanswered or gets a guess.
This e-book is about the part of the funnel nobody measures: the path between the lead and revenue. It's written for the people who decide the budget (owners, executives, marketing managers) and for the people who run the work (analysts, agencies, sales teams). There's no magic formula and no miracle tool. Just what we put in place with our clients so the numbers add up.
Chapter 1Where the report stops (and why)
The report stops at the lead because that's where the easy part ends. Up to the form submission, everything happens inside the ad platforms and the website: Google Ads knows who clicked, GA4 knows who filled out the form, the Meta pixel knows who converted. After the form, the lead enters a different world: the salesperson's WhatsApp, the sales team's spreadsheet, the CRM nobody updates, the billing system that speaks another language.
There are three reasons, and they almost always show up together.
1. The data lives in different places
Media lives in the ad platforms. Behavior lives in GA4. Sales live in the CRM or the ERP. Revenue lives in finance. Each system has its own identifier, and none of them was set up to talk to the others. The result: the lead that cost $40 in Google Ads and the customer who paid a $12,000 invoice are, as far as your systems are concerned, two unrelated people.
2. Teams answer for different metrics
Marketing is measured on lead volume and cost per lead. Sales is measured on closed deals. When each team's goal ends at the other team's border, nobody has a reason to look at the whole path. Marketing delivers cheap leads that don't buy; sales complains about lead quality and stops logging what happens to each contact.
3. Measuring up to the lead feels like enough
Cost per lead is an easy, fast number, and it improves with optimization. The problem is that it measures the wrong stage. Two campaigns with the same cost per lead can produce completely different sales, and the platform's algorithm will push budget toward the one that generates more form fills, not the one that generates more customers.
When the math stops at the lead, your media learns to generate leads. When the math reaches revenue, your media learns to generate customers.
The cost of not closing the loop
Not closing the loop isn't just a reporting problem. It shows up in decisions that cost money every month:
- Budget in the wrong channel. The channel with the lowest cost per lead gets more investment, even if it brings in the leads least likely to buy.
- Cuts to the right channel. Channels that generate few leads, but leads that buy (referrals, branded search, LinkedIn in complex sales), look expensive and are the first to go.
- Fights between marketing and sales. Without data, each team defends its own version. The conversation turns into opinion.
- Scaling into losses. If you don't know your customer acquisition cost, raising the budget can multiply your volume and, along with it, your losses.
In the next chapters, you'll see the stages that need a number, the most common leaks, the step-by-step for connecting your systems, and the metrics that actually guide budget decisions.
The eight stages that need a number
Closing the loop starts with mapping the entire funnel, all the way to the repeat purchase. Each stage answers a business question and has a place where its data lives. If a row in this table has no owner and no system in your company, that's where the math breaks.
| Stage | Question it answers | Main metric | Where the data lives |
|---|---|---|---|
| 1. Reach | How many of the right people saw the brand? | Impressions, reach, CPM | Ad platforms |
| 2. Interest | How many were interested enough to click or search? | Clicks, CTR, branded searches | Ad platforms, Search Console |
| 3. Visit | How many reached the site and stayed? | Engaged sessions, engagement rate | GA4 |
| 4. Lead | How many left their contact information? | Leads, conversion rate, cost per lead | GA4, form, CRM |
| 5. Qualified lead | How many have the right profile and are ready to buy? | Qualification rate, cost per qualified lead | CRM |
| 6. Opportunity | How many turned into a real negotiation? | Opportunities, cost per opportunity | CRM |
| 7. Sale | How many bought, and for how much? | Sales, close rate, deal size, CAC | CRM, ERP |
| 8. Revenue and repeat purchase | How much actually came in, and how much came back through repeat purchases? | Revenue received, margin, LTV, repeat purchase | Finance, ERP |
In e-commerce, stages 5 and 6 merge with the cart and checkout, but the logic is the same: the sale recorded on the platform is not the revenue. Canceled orders, returns, subsidized shipping, and coupons all change the number that matters.
Define a qualified lead before you measure
Stage 5 is the most neglected. Without a written definition, every salesperson qualifies leads their own way and the number is useless. A good definition fits in three lines and uses criteria that the form itself or the first conversation can confirm:
- Profile: the industry, company size, job title, or region you serve well.
- Need: the problem the person wants to solve is a problem your product solves.
- Timing: there's a deadline or a budget to make a decision in the coming months.
Marketing and sales sign off on this definition together. It becomes the yardstick for measuring the quality of each campaign, not just its volume.
Chapter 3The four leaks that make revenue disappear
When the numbers don't add up, the problem is almost always in one of these four places. It's worth going through the list with your team and marking where your operation leaks.
Leak 1: the lead arrives with no source
Symptom: the CRM has hundreds of contacts with the source listed as "website," "WhatsApp," or left blank. Nobody can say which campaign brought in which customer.
Why it happens: campaign links don't use UTM parameters, or use them with no standard. The form doesn't capture those parameters. The WhatsApp button takes the person off the site without carrying that information along.
How to fix it: standardize UTMs on every paid and organic link, capture them in hidden fields on every form, and use WhatsApp links with a prefilled message that identifies the source. Turn on Google Ads auto-tagging to store the click identifier.
Leak 2: the CRM has no stages and no loss reasons
Symptom: the CRM is just a contact list. There's no way to tell how many leads became opportunities, how many were lost, and why.
Why it happens: the CRM pipeline was set up once and never used, or it has too many stages. Salespeople don't update it because updating it doesn't help them sell.
How to fix it: keep the stages few, each with a clear entry criterion, and make a loss reason required whenever a lead is discarded. Five to seven stages cover most consultative sales. If your team sells over WhatsApp, integrate WhatsApp with the CRM so the conversation and the stage move together.
Leak 3: the sale happens outside the system
Symptom: the lead goes into the CRM, but the sale closes over the phone, at the counter, in the physical store, or in a side spreadsheet. In the system, the opportunity sits untouched forever.
Why it happens: the sales process doesn't include recording the close, or there are two systems (CRM and ERP) that don't talk to each other.
How to fix it: define a single place where the sale is marked as won, with its value. If deals close in the ERP, set up a weekly routine (manual at first, automated later) that brings the status back to the CRM, using the customer's email, phone number, or tax ID as the key.
Leak 4: revenue never makes it back to your media
Symptom: the CRM may know who bought, but Google Ads and Meta keep optimizing for form submissions.
Why it happens: nobody set up the return path. Platforms learn from the events they receive; if all they get is "lead," all they know how to find is leads.
How to fix it: send qualified lead and sale events, with their value, back to the platforms. In Google Ads, you do this with offline conversion imports (using the click identifier) or with enhanced conversions for leads. In Meta, with the Conversions API. From then on, optimization starts targeting the people who buy.
Recognize any of these leaks?
In a Manáry diagnostic, we walk through your funnel and show you where the revenue disappears. No sales pitch.
How to connect media, website, CRM, and revenue
The path below works with the tools most companies already have: GA4, Google Tag Manager, the ad platforms, a CRM, and the finance system. The order matters: each step depends on the one before it.
Create a UTM standard and use it everywhere
Define a single convention for utm_source, utm_medium, and utm_campaign (all lowercase, no accents, separated by hyphens) and apply it to ads, emails, posts, your Instagram bio, and QR codes. Document the standard on a page the whole team can refer to. A link builder keeps people from inventing their own format.
Capture the source on every form
Add hidden fields for utm_source, utm_medium, utm_campaign, the landing page, and, if possible, the click identifier (gclid on Google, fbclid on Meta). Store these values in a cookie on the first visit so you don't lose the source when people browse around before converting.
Bring everything into the CRM with a unique key
Every lead enters the CRM with the source fields filled in and a key that will follow the person all the way to finance: email, phone number, or tax ID. Without a key, there's no link between the lead and the sale.
Organize the sales pipeline into stages with clear criteria
Lead, qualified, opportunity, proposal, won, and lost. Every stage change gets a date and, for losses, a reason. Those dates are what let you measure the sales cycle and the conversion rate between stages.
Send qualification and sales back to the platforms
Set up qualified lead and sale events (with value) to be sent to Google Ads and Meta. If there's no direct integration yet, start weekly and by hand, with an import spreadsheet. What matters is that the algorithm starts learning from the people who buy.
Reconcile with the revenue you actually received
Once a month, compare the sales in the CRM with what actually came into finance. Cancellations, unpaid bills, and returns correct the number. That reconciled revenue is what goes into your return calculation.
If your sales team sells mainly over WhatsApp, step 2 includes the WhatsApp button: use links with a different prefilled message for each source, or a tool that logs the conversation in the CRM along with the click source.
Chapter 5The metrics that matter at each stage
With the funnel connected, you can calculate metrics your media report never showed. These are the ones we track for every client. Click on each one to see the formula and how to read it.
Cost per qualified leadHow much does a contact worth pursuing cost?
Formula: channel spend ÷ qualified leads from that channel.
It replaces cost per lead as your optimization metric. A channel with more expensive leads can have the cheapest qualified leads.
Qualification rateHow many leads look like buyers?
Formula: qualified leads ÷ total leads.
It measures the quality of your targeting and your message. When it drops, the problem is usually the audience or the ad's promise, not the sales team.
Cost per opportunityHow much does a real negotiation cost?
Formula: spend ÷ opportunities opened.
It's the best leading indicator of sales in long cycles, because it shows up weeks before the deal closes.
Close rateHow many opportunities turn into sales?
Formula: deals won ÷ opportunities closed (won + lost).
Read it alongside your loss reasons. If the rate drops in one specific channel, the leads from that channel are arriving with different expectations.
CACHow much does it cost to win a customer?
Formula: (media + tools + marketing and sales team) ÷ new customers in the period.
Calculate both total CAC and media-only CAC. The first tells you whether the business holds up; the second helps you compare channels.
Real return (reconciled ROAS or ROI)How much revenue came back for every dollar invested?
Formula: attributed reconciled revenue ÷ spend.
The ROAS the platform shows is the platform's own estimate. Real return uses the revenue that actually reached finance. The gap between the two is usually large, and that's what should guide your budget.
Sales cycleHow long does it take to get from lead to cash?
Formula: average (or median) number of days between the lead coming in and the deal being won.
It sets the right window for evaluating a campaign. Judging a campaign after 15 days when the product has a 60-day cycle means switching off a good campaign.
LTV:CAC and paybackDoes the customer pay back what they cost? How fast?
Formulas: LTV ÷ CAC; and CAC ÷ monthly margin per customer.
A common industry benchmark is an LTV at least three times your CAC. Payback shows how much cash the company needs in order to grow. Use Manáry's CAC and LTV calculator to run these numbers.
The dashboard that closes the loop
A good dashboard isn't the one with the most charts. It's the one that lets you make the week's decision in five minutes. We organize ours into four blocks, in this order.
Block 1: executive view
One row per month with spend, qualified leads, sales, reconciled revenue, CAC, and return. This is the block the business owner reads. If it takes more than ten numbers, it isn't executive yet.
Block 2: funnel
The eight stages from chapter 2, with the volume at each one and the rate of progression between them, month over month. This is where you find out whether the problem is generating demand, qualifying it, or closing it.
Block 3: channels
One row per channel with spend, cost per qualified lead, cost per opportunity, sales, and CAC. This is the block that reallocates the budget.
Block 4: cohorts
Leads grouped by the month they came in, showing how many turned into sales within 30, 60, and 90 days. This is the block that keeps you from judging long-cycle campaigns too early.
| Cadence | Who attends | The question for the meeting | Dashboard block |
|---|---|---|---|
| Weekly | Media and sales | Did any channel or campaign fall outside expectations? | Channels and funnel |
| Monthly | Marketing, sales, and leadership | Are CAC and return on target? Where does this month's budget go? | Executive view and channels |
| Quarterly | Leadership | Are the customers who came in paying back what they cost? | Cohorts and LTV:CAC |
The tool matters least. Looker Studio, Power BI, or even a well-kept spreadsheet will do the job, as long as the data arrives on its own or through a short routine, and always in the same format.
Chapter 7Attribution without dogma
Once the loop is closed, the next question comes up: did the sale come from Google, Meta, email, or a referral? Every platform will say it was theirs. The good news is that you don't need a perfect model. You need a consistent one.
- Start with last non-direct click in the CRM. It's simple, everyone understands it, and it's already far better than attributing nothing.
- Compare it with GA4's data-driven model to see the channels that help along the way, like video, content, and remarketing.
- Ask the customer. A "how did you hear about us?" field on the form or in the first conversation reveals referrals, podcasts, events, and everything else no pixel captures.
- Test the incrementality of your biggest channels. Pausing a channel for a few weeks in one region, or cutting its budget in a controlled way, shows how much of your sales really depended on it.
An attribution model is like a ruler: it doesn't have to be perfect, it has to stay the same, so you can compare one month with the next.
Chapter 8A 30-day plan to close the loop
Don't try to fix everything at once. This is the plan we use at the start of a project. In four weeks, the company goes from "I don't know how much our media sells" to its first reliable number.
Week 1: inventory and definitions
List your channels, your tools, and where each piece of data lives. Write down, with marketing and sales, the definition of a qualified lead and the stages of the sales pipeline. Choose the customer's unique key.
Week 2: a source on everything
Standardize your UTMs, add hidden fields to your forms, adjust your WhatsApp links, and check that every new lead reaches the CRM with its source filled in.
Week 3: the CRM in order
Simplify the stages, make the loss reason required, and agree on an update routine with the sales team. Run the first reconciliation against the previous month's sales.
Week 4: the first dashboard and the return path
Build the executive view and the channels block. Run the first import of qualified leads and sales into Google Ads and Meta. Schedule the first monthly meeting with the number on the table.
By the end of the month, you won't have a perfect system yet, but you'll have what most companies don't: the cost of a customer by channel, with date and source. From there, every adjustment improves a number that matters.
Frequently asked questions
Do I need to switch CRMs to close the loop?
Almost never. Most CRMs on the market offer custom fields, stages, and integration via API or spreadsheet. The problem is usually the process, not the tool. Only switch if your current CRM can't record source, stage, and value.
What if my team sells over WhatsApp?
You can still measure it. Use WhatsApp links with a message that identifies the source, log the conversation in the CRM (there are integrations for this), and update the stage as the negotiation moves forward. What matters is that the sale doesn't live only on the salesperson's phone.
How long does it take for the platforms to optimize for sales?
It depends on volume. The platforms need a minimum number of conversions per week to learn. If you have few sales, optimize for qualified leads or opportunities, which happen more often, and use sales to evaluate results.
What if the sales cycle is very long?
Use leading metrics, like cost per opportunity, and track cohorts. A campaign that runs in March may only turn into revenue in June. Judging it by the month the money was spent distorts the decision.
Does this apply to e-commerce?
Yes, with some adjustments: the sale is already recorded in the store, but real revenue changes with cancellations, returns, and shipping. Reconcile with finance and track repeat purchases to get to LTV.
Want your funnel measured all the way to revenue?
It's the first thing Manáry does for every media and data client. A specialist looks at your operation and hands you the plan.
Keep reading
Unlock the full material with a quick sign-up. It works for every Manáry resource.
It takes less than a minute. No email sequence you did not ask for.Want us to apply this to your business?
This material is what Manáry does every day with clients. If it makes sense, a specialist looks at your numbers and hands back the priorities, with no sales pitch.
Unlock the full read
Five quick fields and you read it right now, on the page. One sign-up unlocks every Manáry resource.