Think about the last big thing you did to upgrade your own house. The pool. The fence. The fifteen thousand dollar whatever. 

Now count the touches. From the first time you encountered that company to the day you hired them: how many times did you come across them? A card in the mail. Their truck two streets over. You Googled them. An ad followed you around for a week. You read the reviews. You asked a neighbor. Weeks passed. Sometimes months. Then one day the problem got loud enough, and you reached out. 

That journey has a shape, and the shape is a funnel with three layers. 

Demand is created up top. The card, the truck, the ad, the neighbor. None of it made you a lead. It made you aware. It planted a name in your head before you had a problem, so that when the problem showed up, that name was already on the short list. 

Demand is captured in the middle. This is the layer almost nobody maps, because you experience it as “checking them out.” You Google the company’s name. You find their Business Profile. You read their reviews. You open their website on your phone, standing in your kitchen. Some of this layer is bought, and that is where your PPC lives, catching people who are already searching. Most of it is not bought. It is maintained. This is where interest survives, or doesn’t. 

Demand converts at the bottom. Not just via phone. A call is one way in. An online booking is another. A form is another. Every owner tells me they want the phone to ring. With respect: you don’t want rings. You want appointments. 

There is also a fast lane running down the side of that funnel we call direct response. Some people convert straight off the touch. They call the tracked number on the card. They scan the QR code. They click a digital ad and book right there. That lane is real, it is trackable, and it exists for every channel, not just mail. It is also thin. On a fifteen thousand dollar decision, and increasingly on every decision, the fast lane is the exception, not the rule.

Every sale you almost made died somewhere on that map. After auditing call logs, CRMs, and matchback data across Atlanta home service companies, I can tell you it was killed by one of three enemies. 

Starvation. Leakage. Blindness. 

Each one attacks a different part of the funnel. Each one produces the same symptom, fewer jobs, which is exactly why owners keep prescribing the wrong medicine. More ad spend cannot fix a broken website. A better website cannot fix an empty funnel. Two budgets, one funnel, and the first job is diagnosing which enemy you are actually fighting. 

Enemy #1: Starvation 

Starvation is what happens when nothing new enters the funnel. 

The top band goes quiet. No cards landing. No truck sightings. No ads interrupting anyone’s scroll. No neighbors mentioning your name, because you haven’t been in enough driveways lately to be mentioned. 

Here is the cruel part: a starved funnel can look healthy for months. Your capture layer keeps working. Your Business Profile is dialed in, your site is fast, your PPC is humming. All of it keeps harvesting the demand that already exists in your market. Then the pool thins, and the leads quietly dry up, and the owner stares at the dashboard saying the sentence I hear more than any other: “What used to work stopped working.” 

Nothing stopped working. The harvest ran out because nobody kept planting. 

This is where owners most often reach for the wrong budget. Leads are down, so they push more money into capture: a bigger PPC bid, a new landing page, a fourth lead-gen service. But capture cannot create. It can only convert interest that already exists. Feeding the cash register does not put customers in the store. 

The evidence for funding the slow half of the funnel is not just intuition. Les Binet and Peter Field analyzed nearly 1,000 case studies and found that brands weighting spend toward long-term brand building over short-term activation, roughly 60/40 as a benchmark, produced stronger long-run results: higher market share and stronger pricing power than brands that leaned harder into activation. Brand building is demand creation on a slower clock. It doesn’t show up in a weekly PPC report, and it is the only thing standing between you and a thinning pool. 

We watched this play out with Atlanta Painting Company, a client that has kept demand creation running continuously for more than five years. Not a burst before the busy season. Continuous presence, in the home, before the need exists. The result is a program returning 390% ROI on

tracked calls alone, with average job sizes running about $2,000 higher than their baseline. That is what an always-on creation engine looks like when it compounds. 

The antidote to starvation: an always-on creation engine. Presence before need. 

Enemy #2: Leakage 

Leakage is what happens when demand is created and then dies on the way to the sale. 

The campaign up top is beautiful. The card lands, the homeowner is interested, and within minutes they do exactly what you would do: they Google the company. And they walk into this. 

They search your name and find a competitor’s ad sitting on top of your own listing. That has a name, by the way: conquesting. Your competitor pays Google to appear when someone searches for you. You created the demand. They are standing at the end of the driveway collecting it. 

They look for your Business Profile and it doesn’t exist, or the last photo is three years old. They read your reviews and the most recent reply from the owner is from two summers ago. 

They open your website on their phone and it takes eight seconds to load, the text overlaps the pictures, and there is no way to book anything. A contact form wants eleven fields. Or they call on a Saturday morning, mail in hand, and it rolls to voicemail. 

Here is what makes leakage expensive: the fast lane still works while all of this is happening. People still call the tracked number. The direct response numbers look fine, or at least explainable. But the wide path, the majority who check you out before they commit, walks up to a broken capture layer and leaks out before anything gets counted. The owner looks at a trickle of tracked calls and concludes the marketing doesn’t work, when what actually broke was everything between the marketing and the sale. 

This is the mirror image of the starvation mistake. Leads are leaking, so the owner buys more creation: another mailing, a bigger ad. But creation cannot capture. Pouring more water into a leaky bucket is not a plumbing strategy. 

The maintenance items are not glamorous, which is exactly why they get skipped. Claim and update the Business Profile. Reply to reviews, including the old ones. Test your own site on your own phone the way a stranger on a couch would. Cut the form to the fields you actually need. Answer the phone on weekends, or put a service in place that does. Run a brand campaign so your own name isn’t rented out from under you. None of it is exciting. All of it is cheaper than the demand you are currently losing.

The antidote to leakage: audit the funnel tile by tile. The list above is the checklist. 

Enemy #3: Blindness 

The third enemy is the sneakiest, because it doesn’t kill the sale. The customer buys. Blindness kills your ability to see where the sale came from. 

Go back to the funnel. What can your reports actually see? The fast lane: tracked calls, QR scans, direct ad clicks. And the middle layer: the Google search, the website session, the PPC click. What your reports cannot see is everything up top. The card that started it. The truck sighting. The neighbor’s recommendation. Nothing logs those. 

So when the lead arrives, your CRM does the only thing it can do. It assigns credit to the last touch it can see. Logged: “Google.” Logged: “Online.” Logged: “Website.” The capture layer collects credit for demand it did not create. The cash register signs for the whole sale. 

Even asking the customer doesn’t fix it. In one client’s call log, on the tracked direct-mail line itself, roughly one caller in four credited a different source when asked how they heard about the company. People tell you the last touch they remember, not the journey they took. 

Now, a fair question: if the sale still happened, why does any of this matter? 

Because credit decides budget. Every quarter, you sit down with the scoreboard and decide where next year’s money goes. If the credit is sitting in the wrong column, the money follows it. You feed the cash register and starve the thing that was actually creating your customers. Blindness doesn’t kill today’s sale. It kills next quarter’s, in a budget meeting, months early. And when those sales finally die, the death certificate never says “attribution error.” It says “the market slowed down.” Or “mail stopped working.” 

Blindness is how enemy number three hands you to enemy number one. 

Here is what it looks like when you remove the blindfold. One of our window restoration clients believed, based on self-reporting and CRM source fields, that direct mail had produced a modest amount of revenue. Call it X. Then his team matched every caller against the CRM. No attribution debate, just fixing the matching: the provable direct response alone came in at 2.6 times what the scoreboard had said. Then we ran the full matchback, matching every closed sale against the mailed addresses. The ceiling came in at 10 times the original number. 

We do not claim the ceiling. That is the point of measuring both ends. The floor is what you can prove: the ceiling is everyone who bought and had the card in their home, and some of them never looked at it. The truth lives in between. But even the floor, the pure provable number, was two and a half times what the owner’s own systems believed. And chance cannot explain the

ceiling either: we mail roughly 5% of Atlanta homeowners, and 35% of this client’s leads matched the mail file. Seven times what randomness would predict. 

The marketing was working. He just couldn’t see it. And a scoreboard stuck at X had nearly convinced a smart owner that the top of his funnel wasn’t worth funding. 

If you want the full argument for why matchback deserves the same acceptance as the view-through and multi-touch credit that digital channels grant themselves by default, I wrote a separate piece on exactly that: The Attribution Double Standard. The short version is that every channel in your stack already takes delayed, indirect credit for the wide path. Mail is simply the channel that gets graded on the fast lane alone. 

The antidote to blindness: measure the floor and the ceiling, and make budget decisions from both. 

Which enemy is yours? 

Here is a diagnostic you can run this week, no consultant required. 

Google your business name, on your phone, the way a stranger on a couch would. What is sitting on top of your own listing? How do the reviews look? How fast does your site load, and could someone book a job at nine o’clock on a Tuesday night, or only make your phone ring? 

Then pull your last twenty customers and look at the source your CRM assigned to each one. Ask yourself honestly whether you believe it. 

If leads have been drying up slowly while your dashboards look fine, you are probably starving. If response is strong but jobs aren’t following, you are probably leaking. And if you have ever cut a channel because “the numbers weren’t there,” without ever matching your sales file against your marketing file, there is a real chance you were blind, and the numbers were there the whole time. 

All three enemies are fixable. But they take different medicine, and the most expensive mistake in local marketing is prescribing for the wrong one. More capture spend cannot cure starvation. More creation spend cannot cure leakage. And no spend at all can cure blindness. Only measurement can. 

Two budgets. One funnel. Three enemies. Diagnose first.

Frequently asked questions 

Why did my direct mail (or any advertising) suddenly stop working? It usually didn’t. The two most common culprits are leakage, where the demand your advertising creates dies on a broken Business Profile, an unanswered phone, or a website that fails on mobile, and blindness, where the sales are happening but your CRM credits them to the last click instead of the channel that created the demand. Audit the capture layer and run a matchback before cutting the budget. 

Should I spend more on PPC when leads slow down? Only if the diagnosis says your problem is capture. PPC harvests demand that already exists. If the funnel is starving, meaning nothing new is entering the top, more PPC spend just competes harder for a shrinking pool. Check whether demand creation has been running consistently before adding capture spend. 

What is conquesting? Conquesting is when a competitor pays search engines to show their ad when someone searches your business name. It intercepts demand you created. The defense is running an inexpensive brand campaign on your own name so your listing, not a competitor’s ad, appears first. 

What is matchback attribution? Matchback compares your closed sales file against your marketing file, for direct mail, the list of mailed addresses, to find customers your tracking systems missed. It defines a ceiling on what the channel may have influenced, while tracked direct response defines the floor. The honest answer lives between the two, and both numbers belong in your budget decisions. 

Sources: Les Binet and Peter Field, “The Long and the Short of It,” IPA effectiveness database analysis. Client figures drawn from City Publications Atlanta call tracking, CRM audits, and matchback analyses; client-specific results shared with permission.