I Spent $15,500 on Marketing. Four Customers. Net Loss. Here's What I Learned.

I paid $15,500 over thirteen months to a marketing agency for my permanent-lighting business. The magazine reached 100,000 homes. The Facebook campaign ran for months. At the end of it, I had four customers and a net loss after install costs. I'm going to tell you exactly what went wrong, because most of it had nothing to do with the agency.

The ads themselves weren't the main problem

The magazine placements brought in real customers. Four signed proposals totaling $22,794, including one customer whose proposal had an explicit "HL Magazine Discount" line item. The channel worked at a basic level. The problem was that nobody, including me, knew it was working while it was happening.

The agency never tracked attribution. No UTM parameters, no QR scan data in a dashboard, no CRM tagging by source. I found those four customers months later by digging through my own GoHighLevel records and cross-referencing proposal files. The agency's monthly reports talked about "reach" and "brand visibility." Not one mentioned a customer by name or a dollar figure.

That's the first failure: measuring what is easy to count instead of what matters.

What was wrong with the Facebook campaign?

The Facebook ads ran as Traffic campaigns, which means Meta optimized for clicks, not leads. In paid advertising, campaign objective determines who the algorithm targets: Traffic campaigns find the cheapest clickers, while Leads campaigns find people most likely to complete a form or make a call. For a $2,500 to $5,600 service ticket, you need the second version.

In one of the final months of the contract, the campaign produced 191 link clicks at a cost per click between $4.97 and $5.74. Estimated spend for the month: around $1,000. Leads reported: zero. Not because the creative was bad or the audience was wrong. Because Meta was never told to find people who would call or fill out a quote form. It was told to find people who would click a link. Those are different populations.

There's a more direct problem too. The Creative Director running the campaign did not have access to my Facebook business account until month nine of the thirteen-month contract. The first eight months were print magazine placements only. The Facebook work was an informal add-on, outside the original contract, that started late and ran for four months total before the contract ended.

The part the agency could not fix

My old WordPress site had a broken contact form. Submissions were not triggering email notifications. When someone filled out the form, I never found out.

I know this because six months into the campaign, the agency's sales rep sent a follow-up email after a review meeting that included this bullet: "Fix website alerts: contact your site provider to resolve the issue preventing form-submission notifications from sending." One bullet. Mentioned once. Nobody followed up on it. I didn't fix it for weeks.

This is the operational gap a lot of marketing contracts never cover. Their job, as they define it, is to drive traffic. What happens after the click lands on your site is your problem. If your contact form is broken, your CTA button is buried three scrolls down, your CRM isn't logging new contacts automatically, or your response time is measured in days, the agency will continue sending traffic until the contract expires.

Research published in the Harvard Business Review found that companies responding to leads within an hour are seven times more likely to qualify the prospect compared to those who wait longer. A broken notification system means no response at all. The lead goes cold before anyone knows it arrived. For more on how fast follow-up affects conversion rates for service businesses, see The 5-Minute Rule.

A working form isn't enough if the page doesn't convert

Even after I fixed the notification problem, there was a deeper question I had not asked: is this website actually built to convert, or does it just look decent? Those are two different things, and the gap between them costs home-services businesses more leads than a broken form ever will.

A pretty website and a high-converting website share almost nothing in common. Conversion is a function of structure, not aesthetics. When we build a home-services site at Front Door Digital, we model the layout and hierarchy after the sites with the highest documented conversion rates in the category, not after the ones that win design awards. The patterns are consistent enough that they're almost a formula.

Here is what a converting home-services page actually has:

• A sticky phone number that follows the visitor down the page. The single most common conversion element missing from home-services sites. A homeowner with a bat in the attic at 10pm wants to tap a number, not scroll back to the top to find it. The phone number should be visible at every scroll depth, every time.
• A clear, single call to action above the fold. Not three options. One. "Call now" or "Get a free estimate." Giving visitors too many choices is the fastest way to get none of them to act.
• Social proof within the first screen. Review count, star rating, years in business, or a recognizable badge. It does not need to be elaborate. It needs to exist before the visitor decides whether to keep reading.
• A headline that names what you do and where. "Austin's Trusted Bat Removal Specialists" converts better than "Welcome to Our Website." The visitor should know within two seconds whether they are in the right place.
• Short forms. Every field you add to a contact form removes a percentage of the people who would have filled it out. Name, phone, and a brief message is usually the ceiling. Adding fields for address, best time to call, how they found you, and service type is how you lose 40 percent of your submissions.
• Speed. A site that takes four seconds to load on mobile loses visitors before they ever see your form. Google's own research found that as page load time goes from one to three seconds, the probability of a mobile visitor bouncing increases by 32 percent. A slow site is a leaky bucket regardless of how much traffic you pour into it.

None of these require a big budget or a redesign. Most of them are structural decisions made at the start of a build that are very hard to retrofit later. That is why we treat conversion architecture as the first step, not a nice-to-have after the site is already live and you're wondering why the phone isn't ringing.

Why a working ad campaign can still produce zero customers

Lead generation and lead conversion are two different systems. An ad campaign lives in the first. Your website, your contact form, your CRM, and your response time live in the second. When the first system works and the second doesn't, you get clicks and no customers.

Here is what that sequence looks like in practice:

• A homeowner sees the magazine ad, visits the site, fills out the contact form.
• The form submits successfully on their end.
• No email notification arrives. No CRM record is created.
• The homeowner waits. Nobody calls back within the day.
• They search again, find a competitor, and book an estimate.

That sequence could have repeated the entire length of the campaign without anyone noticing. The agency was doing its job. The front door was jammed shut.

This is not a dig at the agency. They sold magazine placements and built Facebook creative. They delivered that. What they did not sell was the operational plumbing that turns a click into a customer record into a booked appointment. That's a different product. Most businesses don't realize they need it until after they've paid for traffic that went nowhere. The front-door framework exists exactly because of this gap.

What I would do differently

Get your website and CRM working before you spend anything on ads. That sounds obvious. It is not. The failure mode is invisible: you can run an entire ad campaign, receive zero leads, and assume the ads failed when the real problem was downstream.

Seven things to confirm before any paid campaign:

1. Audit your page for conversion basics first. Does the site have a sticky phone number, a single clear CTA above the fold, visible social proof, and a short form? If not, fix those before spending a dollar on traffic. Sending paid visitors to a page that isn't built to convert is the same as filling a bucket with a hole in it.
2. Check your page speed on mobile. Use Google's PageSpeed Insights tool and look at the mobile score. If your largest contentful paint is above 2.5 seconds, you are losing a measurable percentage of paid visitors before they see your offer.
3. Test your contact form yourself. Submit it and confirm you receive a notification within two minutes. If you don't, the system is broken and every lead it generates disappears.
4. Check your CRM intake. Does the submission create a contact record automatically, tagged with the source? If not, you have no data and no follow-up trigger.
5. Measure your response time. If you can't respond within an hour during business hours, build automation to acknowledge the lead immediately and route it to someone who can call back.
6. Require the right campaign objective. Any paid campaign for a high-ticket home service should track form fills, phone calls, or bookings as the primary result. Not clicks. Not impressions.
7. Define attribution before you start. Ask your agency exactly how they plan to connect their campaigns to actual customers. If the answer is "we'll watch traffic go up," find a different agency.

When I rebuilt my permanent-lighting site on Next.js, form submissions route to both email and CRM in real time, with the source tagged on every record. I know within minutes whether a lead came from organic search, a paid campaign, or a specific page. That infrastructure was in place before any paid marketing ran. It is the reason the site works as a lead machine instead of a brochure.

What should you ask a marketing agency before hiring them?

Ask these before you sign. The answers will tell you whether you are buying customer acquisition or traffic delivery.

1. What campaign objective will you use, and how will leads be tracked end to end?
2. What does a successful month look like in leads generated, not clicks or impressions?
3. Will you audit my website and CRM setup before the campaign starts?
4. How will you attribute a closed customer to this campaign?
5. Who will actually manage the ads, and what is their background in performance marketing?

An agency that answers all five clearly before you sign is worth talking to. An agency that responds to those questions with "reach," "brand awareness," or "visibility" for a service business with real cost of goods is asking you to spend money you can't measure. That's a bad trade for a business where every job has material and labor costs baked in.

The fifth question is one most people never think to ask. The person who sold me the contract was the sales rep. The person who built the ads was the Creative Director with a background in magazine publishing. Neither was a performance marketer. I didn't know that until I was eight months in and asking why we had no leads to show for it. You can see what that situation costs in real dollars when you add up the leads that fell through.

---

Frequently asked questions

Why do agencies focus on traffic instead of leads?

Traffic campaigns are easier to manage and easier to report on. Click volume and impression counts go up reliably with more spend. Lead volume is harder to predict because it depends on the full funnel: the landing page, the form, the response time, and the offer. Reporting on clicks looks like progress even when nothing is converting. If the reporting is built around contract renewal instead of booked work, the easy default is the metric that looks best on a monthly slide.

What is the difference between a Traffic and a Leads campaign objective on Meta?

Campaign objective determines who Meta's algorithm targets and what action it tries to drive. A Traffic objective targets users statistically likely to click a link. A Leads objective targets users likely to complete a form, call a number, or take a tracked action. For a home service with a $2,500-plus ticket price, you need Meta to find people likely to request a quote, not people likely to tap a thumbnail and bounce. Running a Traffic objective on a high-ticket service is one of the easiest ways to spend money without learning whether the campaign can produce leads.

How do I know if my website is ready to receive paid traffic?

Run a self-test before spending a dollar. Fill out your own contact form and measure how long it takes to receive a notification. If it takes more than two minutes or never arrives, the system is broken. Then check whether the submission created a record in your CRM with the source tagged. If it didn't, leads are being logged nowhere. Fix both before running any paid campaign. Sending paid traffic to a site that can't capture or route leads is the operational equivalent of handing someone a business card with the wrong phone number.

What makes a home-services website convert well?

The highest-converting home-services sites share a short list of structural features: a phone number that stays visible as the visitor scrolls, a single clear call to action above the fold, social proof within the first screen, a headline that names the service and location, a short contact form with no more than three or four fields, and fast load times on mobile. None of these are design choices. They are conversion architecture decisions made before the first pixel gets placed. A visually polished site that lacks these elements will underperform a plain-looking site that has them.

What should I look for in a home-services marketing agency?

Look for an agency that audits your current lead-capture setup before proposing anything. They should ask about your response time, your CRM, and what happens to a lead after it arrives on your site. If their proposal does not mention conversion readiness, they are selling traffic delivery, not customer acquisition. Ask specifically what campaign objective they plan to use and what cost-per-lead target they are committing to. A specific answer is a good sign. A general answer about "visibility" and "brand building" is a signal to keep looking.