What a Manufactured Home Dealer Lead Actually Costs, and Why Blended Numbers Mislead

The most common question a manufactured home dealer asks before signing with anyone is "what does a lead cost?"

It is a fair question and it does not have an honest single-number answer. Anyone who gives you one is either quoting an average that describes nobody, or quoting their best result as though it were typical.

What we can do is explain what moves the number, why the usual benchmarks are the wrong instrument, and what to look at instead.

Why there is no single number

Cost per lead for a manufactured home dealer varies widely. Not by ten percent between accounts. By multiples, in ways that are largely structural rather than a matter of execution quality.

Here are the drivers that actually do the work.

State and market. Auction prices for attention are not uniform. A dense coastal metro and a rural interior county are different markets for advertising in exactly the way they are different markets for anything else. The same campaign, same creative, same offer, run in two states, will not produce the same cost per lead, and the gap can be large.

Market density for your specific product. This is separate from population. What matters is how many people in the geography could plausibly transact with you. A county with a large concentration of manufactured housing communities is a different proposition from a county with two. Reach is cheap; reach to relevant people is the thing you are buying. We worked through how to measure this properly in how we picked 11 counties over statewide.

The offer and the ask. "See our current inventory" and "get a delivered price on a three bedroom" and "find out what your home is worth" are three different transactions for the person tapping. The lower the commitment, the cheaper the lead and the less it tends to mean.

Buyer side versus seller side. Advertising to people who want to buy a home and to people who want to sell one are different markets with different competition and volume. Blending them into one figure is meaningless.

Form settings. Verification, conditional logic and higher intent mode each reduce submitted volume and raise reported cost per lead, by design. A dealer with a phone verified form and a dealer without one are not producing comparable numbers, even in the same town. See what conditional logic really costs you.

Season and inventory. Manufactured housing has real seasonality and real supply constraints. An account with nothing to sell this month cannot buy its way out of that.

Channel. Search captures people already looking and typically costs more per lead while producing higher intent. Paid social creates the intent and typically costs less. Averaging the two produces a figure that describes neither, which is why we report them separately.

Every one of those is structural. Two competent operators running the same playbook in different markets will land in different places, and neither of them is doing it wrong.

Why a portfolio blended average is a bad benchmark

Agencies love a blended average because it is a real number they can quote. It is still the wrong number for you, for three reasons.

It describes a distribution, not an account. A portfolio average is the middle of a wide spread. Your account is a point in that spread, and its position is determined mostly by the structural factors above, which the average has deliberately flattened out. Being told the average is X tells you almost nothing about whether your X is good.

It is weighted by whoever spends most. A blended figure across accounts is dominated by the largest spenders, whose markets and offers may look nothing like yours. If two large accounts in cheap markets sit in the portfolio, the average bends toward them and the small account in an expensive metro is measured against conditions it can never meet.

It moves for reasons that have nothing to do with you. A portfolio average changes when a client churns, when a new account onboards, when somebody's budget doubles. None of those are information about your campaign, but all of them change the benchmark you are judged against.

The practical damage is real. We have seen dealers conclude their campaign was failing because it sat above an average they were shown, when their market was structurally more expensive and their actual sales results were fine.

Why comparing to your best ever campaign is worse

There is one benchmark that does more harm than the blended average, and it is the one clients reach for naturally: the best month they ever had.

It is worse for a specific reason. A best ever result is, by definition, drawn from the favourable tail of a distribution. Some of it was execution and some of it was circumstance: a seasonal peak, a competitor who happened to be dark, a creative that caught a moment, a market that had not yet been worked.

Treating the top of the range as the expected value means most periods will look like failures, including perfectly good ones. It also drives bad decisions, because the natural response to "we are not hitting the number we hit in April" is to start changing things, and constant change is itself a reliable way to make an account perform worse.

Best case is a demonstration that something is possible. It is not a forecast.

Best case is a demonstration that something is possible. It is not a forecast, and treating the top of the range as the expected value makes most perfectly good periods look like failures.

What to measure instead

Four things, in order of how much they will change your decisions.

1. Your own account's trend, against itself. The only benchmark with the same structural conditions as your campaign is your campaign. Is cost per lead this month better or worse than your own recent months, and is there a reason for the move you can name? That comparison is fair. A comparison to somebody else's account is not.

2. Cost per contacted lead. A lead you cannot reach is not a lead. Divide spend by leads you actually got a human conversation with, and you get a number that is immune to a whole class of quality problems. This metric alone explains most of the gap between a campaign that looks cheap and a campaign that produces business.

3. Cost per appointment or per qualified opportunity. This is the number that should drive budget decisions. It is also the point at which the follow-up process becomes visible in the metric, which is uncomfortable but correct, because speed and persistence of follow-up affect this number more than most ad changes do. Two dealers with identical cost per lead can have very different cost per appointment, and the difference is usually in what happens in the first hour.

4. Cost per sale, against your gross margin. The final arbiter. A high cost per lead with a strong close rate and a healthy margin beats a cheap lead that never closes. Once you know your cost per sale and your margin, the question "is this working" becomes arithmetic instead of opinion. We laid out the framework in how much to spend on paid ads.

Note what happens as you move down that list: the numbers get harder to produce and more useful. That is not a coincidence. Cost per lead is popular because the platform hands it to you for free, not because it is the most informative thing available. It is the same reason we argue against cost per click in the real cost of a lead, and the reason cost per booked appointment is the figure we actually manage to.

The measurement that makes all of this possible

None of items 2 through 4 work unless leads arrive in your CRM tagged with where they came from, and their stage movements are recorded. That is a plumbing job that has to be done before you need the answer, not after.

The most common version of this problem we encounter: a dealer wants to know which campaign produced their closed deals, the leads are all in the CRM, and there is no reliable link back to the ad. The data was never captured, so the question cannot be answered retrospectively at any price.

Results vary, and here is what that means

We are deliberately not publishing cost per lead figures by state or a table of expected numbers, and we would be sceptical of anyone who does.

Advertising results depend on your market, your offer, your inventory, your pricing, your competition, the season, and how fast and how persistently your team follows up. Two dealers running identical campaigns in different states can see materially different results, and the same dealer can see materially different results in different quarters. Past performance in one account is not a prediction for another, and nothing here should be read as a projection of what any specific business will achieve.

What we will commit to is the process: measure your own account against itself, carry the measurement past the form fill to the conversation and the appointment, and make budget decisions against cost per sale and margin rather than against somebody else's average.

Our methodology

The drivers listed above come from running paid lead generation for manufactured housing dealers, investors and adjacent businesses across a number of states through 2026, with Facebook and Google campaigns tracked into GoHighLevel. We report cost per lead by channel separately rather than blended, and we look at each account against its own history rather than against a portfolio figure, for the reasons described above. We have not published account level figures here because a number from one market is not a benchmark for another.

Keep reading

• The real cost of a lead: why cost per click is the wrong metric
• Lead generation for mobile home dealers
• ROAS vs cost per lead: which metric matters