Three Numbers Every HVAC Company Should Know About Their Service Area Before Spending a Dollar on Nextdoor

The marketing dashboards most HVAC operators look at give them the wrong numbers.

Followers on the Nextdoor business page, page views, "impressions" on Facebook posts, monthly reach. None of these predict whether your name is showing up where it matters — in the recommendation threads that decide jobs. They predict whether your page is being seen. The recommendation surface is somewhere else.

Three different numbers actually predict it. And they're not visible from inside your business account, which is why almost no operator has ever seen them.

Number 1: Referral-thread count in your service area, last 90 days

How many times, in the last ninety days, did a homeowner in your actual service area post a public request for an HVAC recommendation on Nextdoor, a local Facebook group, or a town subreddit?

This is the denominator of everything else. Without it, the other two numbers don't mean anything. With it, you immediately know the size of the opportunity surface you're working with.

Most operators wildly miscalibrate this number in both directions. Some assume their service area produces 5–10 threads a year ("nobody asks about HVAC online"); the actual number is usually 5–10× higher. Others assume it's hundreds a week; the actual number is high but not that high, and obsessing about volume distracts from the signal-quality question that the next two numbers address.

You can't get this number from inside Nextdoor or Facebook. It requires monitoring those surfaces across your service area, in a structured way, for a defined window. Most operators don't have a way to do this on their own, which is why this number is genuinely missing from their picture.

Number 2: Mention rate

Of those threads, in what percentage was your business name mentioned at all — by anyone, at any time?

This is the signal of whether you're in the conversation. Not whether you're winning it — whether the conversation knows you exist.

A useful frame: if you're a $2M+ HVAC operator with hundreds of past customers in the relevant ZIPs and a mention rate below ~20%, something is structurally wrong with how your customers retain your business name (see the happy-customer post). At a mention rate of ~40–60%, you're a normal established operator who's neither dominant nor invisible. Above ~70%, you're winning the cross-thread compounding game described in the why-same-three post.

The single most useful operator move when you first see this number is to look at the threads where you weren't mentioned and ask: did we have a past customer in this neighborhood? If yes, the gap is recall. If no, the gap is service-area density.

Number 3: First-hour mention rate

Of the threads where your name was mentioned, what percentage of mentions came in the first hour?

This is the signal of whether you're in the conversation when it matters. The first-hour post explains why the active decision window of a thread is short — about sixty minutes — and why mentions that arrive later have a fraction of the value.

You can have a high mention rate and a low first-hour mention rate. That usually means a few of your past customers are reliable late-commenters: they see the thread the next morning, drop your name, the OP has already booked someone else. The mention is good for the long-term reputation of the thread (it still appears when a future searcher finds it) but doesn't win the immediate job.

You can also have a moderate mention rate and a very high first-hour mention rate. That pattern usually means you have a small core of highly-engaged advocate-customers who actively watch their local groups and post fast. This is a much more valuable configuration than a long tail of slow mentioners.

Why these three, not others

The temptation, once you start thinking about this, is to add more numbers — sentiment of mentions, length of comments, ratio of "second this" responses. None of these add much over the three above for operator decision-making at the $1M+ scale.

The three numbers above answer the three operationally distinct questions:

• Is the surface even worth thinking about for our market? (Number 1.)
• Are we in the conversation at all? (Number 2.)
• Are we in the conversation when it counts? (Number 3.)

If the first number is zero or near-zero, neighborhood-surface marketing isn't the lever for your market and you should put the budget elsewhere. If the first number is meaningful but the second is near-zero, the problem isn't volume — it's that your past customers aren't naming you, and the fix is operational, not channel-mix. If the first two look healthy but the third is low, the issue is timing and advocate density, not awareness.

Each diagnosis points to a different intervention. Adding more metrics tends to obscure those distinctions rather than sharpen them.

How to get these numbers

You can produce all three manually for one neighborhood: pick a Facebook group, scroll back 90 days, count the HVAC-recommendation threads, count the ones that mention you, note the timestamps. It's tedious but doable for a single group.

Across all the groups in a real service area — five to fifteen Facebook groups, Nextdoor coverage of every neighborhood, the town subreddit, sometimes Reddit's metro sub — the manual approach stops being practical. That's the gap the free neighborhood report closes for an operator's specific market.

The point of this post isn't to sell the report. The point is that these three numbers are the operator-grade replacement for the vanity metrics that the standard marketing dashboards offer. Whatever way you collect them — manually, with a tool, or via FirstThread — knowing them is the precondition for spending the next $1k of marketing budget intelligently.

Most operators don't know any of the three. The ones who start tracking even rough versions of them tend to make very different decisions within a quarter.