What Neighbors Actually Ask For: A Year of Watching Local Service Requests

After a year of reading neighborhood request posts every single day, across eight trades, the honest answer is this: neighbors almost never ask for a service the tidy, keyword-shaped way software expects. They do not post “pressure washing recommendation.” They post “my driveway is embarrassing, who do I call,” or “is $400 too much for this?” The demand shows up in four recognizable shapes, and it moves on a clock. If there is one thing a year of watching taught me, it is that local services demand is more seasonal, and far more oddly worded, than any keyword tool will ever tell you.

I built NextJet after growing a pressure-washing company almost entirely off neighborhood posts, so now I watch these feeds every day for owners in pressure washing, plumbing, roofing, HVAC, cleaning, lawn care, handyman work, and window and door replacement. I am not going to throw a fake “we analyzed 10,000 posts” number at you. I have no interest in counting them. What I can give you is the shape of the thing, the patterns that repeat so often you start predicting the next post before you finish reading the first line.

The four shapes a request actually takes

Strip away the trade and the town, and nearly every service request I read is one of four moves. Naming them is useful, because each one asks for a different reply, and three of the four are easy to walk right past.

1. The direct ask

The obvious one. “Can anyone recommend a good plumber?” Clean, named, and exactly what people picture when they imagine finding leads this way. It is also the least common of the four, and the most crowded, because it is the only shape the whole town recognizes as a lead. The replies pile up under it fast. If this is the only kind of post you are watching for, you are fighting everyone else for the smallest slice of the demand.

2. The price-sanity check

“Just got quoted $650 to clean my roof. Is that reasonable?” The neighbor is not naming a company. They are asking the street whether a number is fair. Most pros scroll past it because it is not phrased as a request. It is one of the best leads on the feed. This person has an active quote, real intent, and a quiet worry that they are being overcharged. A calm, honest answer about what drives the price wins that job more often than the lowest bid, because you showed up as the helpful one instead of the salesy one.

3. The vent that is really a lead

“Ugh, my gutters are overflowing every time it rains and I am so done with it.” No question mark. No “recommend.” Just frustration. But that is a person with a problem they have decided to fix, they just have not asked out loud yet. A keyword alert set to “gutter cleaning” never sees it, because the neighbor never wrote those words. A human reading the feed catches it instantly, and a warm reply that says “that is a quick fix, here is what is probably going on” gets there before it ever becomes a proper request post.

4. The “do you also do X”

A neighbor asks for a handyman, then someone in the comments adds “does anyone know a good handyman who can also fix a fence gate?” The real job is buried a layer down, inside a thread about something adjacent. These reward breadth. Handyman and pressure-washing requests especially phrase themselves a hundred ways, which is exactly why a single alert keyword cannot cover the trade and why reading the actual post matters.

How the four shapes compare

Here is the same taxonomy in one view. The last column is the part most people miss: the tell that a post is a live job even when it does not read like a request.

When local services demand turns seasonal

The second thing a year of watching makes obvious is the calendar. Demand does not trickle evenly. It ramps, hard, just before the weather that creates the need, then goes quiet again. And the ramp starts earlier than the peak, which is the whole opportunity: if you are already present when the first few posts appear, you own the thread before the rush shows up.

The rhythm is different per trade, but it repeats every year:

• Pressure washing wakes up in early spring. Driveways, house washes, and pollen season in the Southeast light the feed up before Easter. If you run a wash business, the channel side of this is worth its own read: how to get pressure washing customers without buying leads.
• Roofing has two waves: spring, when people assess winter damage, and a burst after any real storm, when half the street posts “who did your roof?” at once.
• HVAC is the sharpest. The feed is calm, then the first heat wave hits and “my AC just died” posts arrive in a cluster, and again on the first hard cold snap. Being the name that comes up in that window is the whole game for a small shop.
• Plumbing spikes when pipes freeze and again during heavy summer use. The freeze posts are urgent and time-critical, and the marketplaces share those exact leads with several shops at once, which is why winning plumbing work without paying for every phone call comes down to being early and local.
• Lawn care ramps at green-up in spring and rides mowing season. Cleaning clusters around moving season and the weeks before the holidays. Handyman work stays steadier than the rest, with a pre-listing bump when neighbors get houses ready to sell.

None of this needs a data science team. It needs someone paying attention year-round instead of only when their own schedule gets thin, which for most owners is the exact moment the feed has already moved on.

Why keyword alerts miss the ones that matter

Here is the pattern that ties it all together. Three of the four request shapes do not contain the trade keyword at all. The price-sanity check names a dollar amount, not a service. The vent names a feeling. The “do you also do X” is buried under a different topic. A tool that watches for the word “gutter cleaning” will only ever catch the one shape that is already the most crowded, and it will sail right past the three that convert best.

This is why I stopped trusting word-matching years ago. On Nextdoor® and every feed like it, the meaning lives in the whole sentence, not one term. “My gutters are disgusting and need to come off” is a real job. “Reminder to clean your gutters before winter, neighbors” is not, it is a helpful heads-up, and a word-matcher treats them identically. The only reliable filter is reading the post the way a person would, deciding what the neighbor actually wants, and ignoring the noise.

What most pros miss

Two things. First, these are public threads. When you reply, the neighbor who asked reads it, and so does everyone who opens that thread for months afterward. One genuinely helpful answer keeps working long after you type it. Second, speed decides who gets the running. The poster is usually ready to hire soon, they read the early replies, and they stop reading. Show up late and the best-written reply in town lands under a job that is already booked. That is the mechanism behind why the first reply tends to win local jobs, and it is the same reason the “my AC died” HVAC posts reward the shop that is watching over the one with the bigger ad budget.

To be clear about how I think this should work: the owner reads and sends every reply. Nobody should be auto-posting canned answers into a neighbor's thread. That is how you get replies deleted and your account flagged, and it reads as a bot to the one person you are trying to win. The job is to never miss the post and to never sound like an ad. Those are two different skills, and both of them start with actually reading what your neighbors are asking for.

The takeaway

If you only remember one thing, make it this: local services demand is seasonal and rarely spoken in keywords. It arrives as a problem described in plain frustration, a price someone is quietly worried about, a job buried in a comment, and it swells and fades on a calendar you can learn by watching. The pros who win are not the ones with the cleverest alert. They are the ones who read the feed like a neighbor, catch the three shapes everyone else scrolls past, and answer while the poster is still reading.