I ask every new client the same question in the first meeting.

"Where do your best customers come from?"

Everybody has an answer. It comes out fast and it comes out confident. Referrals. Google. That one networking group. Instagram.

Then I ask the follow up. "How much revenue did that source produce last quarter, and what did it cost you to get it?"

And the room gets quiet.

Not because they are bad operators. Some of the sharpest people I work with cannot answer that question. They know their revenue. They know their ad spend. They know their close rate. But the line connecting a specific dollar in to a specific dollar out does not exist anywhere in the business, so they run on a feeling.

The feeling is almost always wrong. And it is wrong in a very specific, very expensive direction.

Here is the pattern I see over and over.

The source that generates the most leads gets the credit, the budget, and the attention. It is the loudest thing in the CRM. Every morning there are twelve new names in the inbox from it, so it feels like the engine.

Meanwhile the source that generates four leads a month closes three of them at double the average ticket, and nobody is doing anything to get more of it because it never shows up in a report.

Real numbers from a client, lightly rounded. Paid social produced sixty one leads in a quarter and eight customers. Referrals produced nine leads and six customers. Paid social revenue was thirty two thousand at a cost of nine thousand in ad spend. Referral revenue was thirty eight thousand at a cost of essentially nothing except a few hundred dollars of gift cards and lunches.

He was about to increase his ad budget. He had never once sat down and built a system to generate referrals, because referrals "just happen."

They do not just happen. They just happen to be invisible.

That is the whole problem. Volume is visible. Profitability is not. So owners optimize the thing they can see and quietly starve the thing that actually pays them.

Most owners have tried to fix this. It usually fails for three reasons.

First, the tools are built for somebody else. Attribution software was built for ecommerce, where somebody clicks an ad and buys forty seconds later. Your business has a lead who calls on a Tuesday, ghosts for three weeks, comes back after talking to their spouse, and closes six weeks later after two site visits. There is no pixel for that.

Second, last touch lies to you constantly. Someone hears about you from their neighbor, googles your name, clicks your ad, and fills out the form. Your ad platform reports a conversion. Your neighbor gets nothing. You then increase ad spend based on a referral you accidentally paid for.

Third, and this is the big one, nobody assigns the answer to the right moment. Businesses that do track source capture it when the lead comes in and never look at it again. So you know where leads come from. You still have no idea where customers come from.

The fix is not software. I want to be clear about that, because the instinct is always to go buy something. The fix is one question, one field, and one spreadsheet you look at every month.

Ask this, in these words, at intake:

"Before today, how did you first hear about us?"

Three parts of that sentence are doing work.

"Before today" pushes past the last click. It stops someone from saying "your website" when the real answer is "my sister in law told me and then I found your website."

"First hear" gets you the origin, not the final step. Origin is the thing you can buy more of.

And it is open ended. For the first ninety days, do not give them a dropdown. Dropdowns train people to pick the closest option, and the closest option is usually wrong. Let them talk. You will learn things about your own business that a menu would have hidden, like the fact that eleven percent of your customers came from one Facebook group you have never posted in.

After ninety days of open text, build your dropdown from what people actually said. Now the menu reflects reality instead of your assumptions.

Who asks it matters too. If a human is talking to the lead, the human asks it and writes down the actual words. If it is a web form, put it as the last field, optional, open text. Optional fields at the end get answered at a surprisingly high rate when everything before them was easy.

Now the part that separates this from every failed attempt you have made before.

You record the source at the point of sale, not just the point of lead.

In your CRM, create one required custom field on the opportunity or deal, not the contact. Name it Origin Source. It cannot be skipped when a deal moves to won. If your CRM lets you make a field required at a stage change, do it. If it does not, make it a checklist item nobody can close a deal without.

Why the deal and not the contact? Because a repeat customer who originally came from a referral and now buys from an email campaign has two different stories, and you want both. Contact level tracking overwrites. Deal level tracking accumulates.

In Go High Level this is a fifteen minute build. Custom field on the opportunity, required at the won stage, then a smart list filtered by it. Whatever you use, the principle holds. One field. Required. At the deal.

If leads come in from six different places and you are copying them by hand, wire it up. Make.com will take your form submissions, your call tracking data, and your inbox and drop them into the CRM with the source already populated, which removes the single biggest failure point in the whole system. Which is you, at nine at night, forgetting.

And if a meaningful chunk of your leads call instead of filling out forms, get unique tracking numbers per source. One number on the truck, one on the Google listing, one in the ads, one on the direct mail piece. It is maybe forty bucks a month and it eliminates the largest blind spot most service businesses have.

Five columns you enter. Four the sheet calculates for you.

You enter, per source, per month: leads, consults or estimates booked, customers closed, revenue collected, and cost.

It calculates: close rate, cost per customer, revenue per lead, and return on investment.

Revenue per lead is the one to watch. It is the most useful number almost nobody tracks. It collapses volume and quality and ticket size into a single figure you can compare across wildly different channels. A source producing forty leads at eighty dollars of revenue per lead is worth less than a source producing eight leads at nine hundred dollars of revenue per lead, and revenue per lead is the only number that makes that obvious at a glance.

Two rules that make the sheet honest.

Rule one: cohort by entry month, not close month. A lead that arrived in July and closed in September belongs to July. If you credit it to September, every month with a long sales cycle looks like a disaster and every month after a good month looks like genius. Track the lead by when it walked in the door.

Rule two: put a cost on the free stuff. Referrals are not free. They cost you gift cards, lunches, the time you spend maintaining relationships, and whatever you pay out in referral fees. Networking is not free. It costs you four hours a week and a membership. Estimate it, write it down, be roughly right. A source with no cost line will always look like a miracle, and you will never make a good decision about it.

Give it ninety days minimum before you make big calls. And do not kill a source on six leads. Small samples lie even louder than no data.

Once you have a real read, everything sorts into three buckets.

Feed it. Return above three to one and close rate at or above your average. Put more money, more time, and more systems here. If it is referrals, this is where you finally build the ask into your process instead of hoping.

Fix it. Decent volume, bad close rate. The source is not the problem. Something between the lead and the sale is. Look at response time, at who is handling those leads, at what the ad promised versus what the call delivers. A source with volume and a broken close rate is usually the cheapest win in the business, because the traffic is already paid for.

Cut it. Below one to one after ninety days with enough volume to judge. Directory listings and paid lead aggregators land here constantly. Cut it and move the money.

One caution on the reallocation. Do not move more than about twenty five percent of a channel's budget in a single month. Shift too much too fast and you blow up the data you just spent a quarter collecting, and you will not know whether the drop came from your decision or from noise.

I have run this exercise dozens of times and the findings rhyme.

Referrals and repeat business almost always win on return and lose on volume, which means the highest leverage move in the business is usually building a repeatable referral ask, not buying more traffic. Paid channels usually generate the most leads and the worst close rate, which means the fix is your follow up speed, not your targeting. Paid directories are usually a straight loss. And there is almost always one small, weird source doing outsized damage in the good way that you never noticed and never fed.

You are looking for that last one. It is in there.

"My business is too complicated for this." No it is not. Complicated businesses are exactly the ones where guessing costs the most. If you have multiple service lines, add a second column for service line and run the same sheet. That is the entire adaptation.

"People do not remember how they heard about us." Some do not. You are not going for a hundred percent. If seventy percent of your closed deals have a source on them, you have enough to make good decisions, and that is a massive upgrade over the zero percent you have now. Log the unknowns as Unknown and move on. If Unknown climbs above thirty percent, that is a process problem with whoever is intaking, not a customer memory problem.

"I already have this in my ad platform." Your ad platform tells you what your ad platform did. It has no idea about the referral, the truck wrap, the podcast, or the guy who has been watching you for a year. It also has a strong financial interest in taking credit. This scoreboard is the only view that sees everything on equal terms.

Systems die from ceremony. Keep this one small enough that it survives a busy month.

Daily, thirty seconds: whoever intakes the lead writes down the actual words. Not a category. The words.

At close, ten seconds: the person who won the deal fills the Origin Source field. It is required, so this takes care of itself.

Monthly, twenty minutes: enter the five columns for last month. Look at revenue per lead by source. Ask one question, which is whether anything moved enough to change what you do next month.

Quarterly, one hour: run the feed, fix, cut sort. Make the budget calls. Update your dropdown from whatever new sources showed up in the open text.

That is under three hours a year of actual work to permanently stop guessing where your money comes from. I have watched owners spend more time than that choosing a logo font.

I put the whole system into one document so you are not building this from scratch on a Saturday.

The Lead Source Scoreboard. It has the tracker itself with the exact columns and formulas, the intake scripts for phone and web, the CRM field setup, the cohort rules, the decision thresholds for feed, fix, and cut, and a thirty day install plan that gets you from nothing to a working scoreboard without stopping the rest of your business.

It is free. Reply to this email with the word SOURCE and I will send it over.

Then go ask the next person who calls you where they heard about you. Write down exactly what they say.

You are one afternoon away from knowing something about your business that most of your competitors will never know about theirs.

P.S. Reply with the word SOURCE and I will send you The Lead Source Scoreboard. Tracker, intake scripts, CRM setup, decision thresholds, and a thirty day install plan. No charge.

Talk Soon,
Dan

Dan Kaufman
Founder, Dead Simple Growth and Pinnacle Masters