Marketing automation is software that runs a repetitive marketing or follow-up task for you, triggered by something that already happened: a form gets submitted, a job gets marked complete, a customer hasn't booked in a year. For a contractor, the automations worth building first aren't the flashy ones. They're the ones that close a time gap a person can't close fast enough on their own: an instant reply to a brand-new lead, a review request sent while the job is still fresh, and a scheduled nudge to a past customer before they search for someone else. None of it replaces a real conversation. It just makes sure the chance for one doesn't get missed while someone is on a ladder or driving between jobs.
By the numbers
- Firms that tried to contact a new lead within one hour were nearly 7x more likely to qualify it than firms that waited just one more hour, and more than 60x more likely than firms that waited 24 hours or longer, per a Harvard Business Review audit of 2,241 U.S. companies. The study is older than our usual 2014-forward bar for new research, but it remains the largest published dataset on lead response speed and nothing of comparable scale has replaced it since. INDEPENDENT RESEARCH
- 23% of the audited companies never responded to the test lead at all, and among the ones that eventually did, the average response time was 42 hours, long after most buyers had already called someone else. INDEPENDENT RESEARCH
- 98% of homeowners say it matters that a business has recent reviews, and 70% won't request a quote from one rated under 4 stars, per Jobber's own 2024 survey of 500+ U.S. homeowners and renters, run alongside the launch of its native automated review-request tool. INDEPENDENT RESEARCH
- The single best time to ask for a review is right after the job, while it's still fresh. The next-best window is within 24 hours, per BrightLocal's own consumer research. Miss both, and the ask depends on someone remembering to send it manually. INDEPENDENT RESEARCH
The five automations worth building first
Most shops try to automate everything at once and end up finishing none of it. These five, built in roughly this order, cover the moments that actually cost booked jobs when they're left to memory.
| Automation | Triggered by | What it prevents |
|---|---|---|
| Instant lead reply | A new form, chat message, or missed call | A lead who's already moved on to the next search result by the time someone calls back |
| Review request | A job marked complete | A happy customer who meant to leave a review and simply forgot by the time anyone asked |
| Appointment reminder | A booking gets made | A no-show that costs the slot and the drive time, not just the one job |
| Win-back nudge | No job on file in 12+ months | A past customer who defaults to a fresh search instead of calling someone who already knows their address |
| Seasonal check-in | A set calendar date, e.g. pre-season | A maintenance job that goes to whoever the customer finds that week, because no one reminded them first |
Where the automation actually lives
Two different kinds of tools cover this, and most shops end up running some mix of both. Job-management platforms built for contractors are adding marketing automation directly: Jobber launched native Campaigns and Reviews tools in March 2024, letting a shop send segmented email campaigns and automated review requests without leaving the software it already uses for scheduling and invoicing. The limitation is usually scope: automation built into a job platform is strongest at the tasks tied directly to a job record, like a review request, and weaker at a full multi-step sequence across email and text.
A dedicated marketing automation platform, like the GoHighLevel system behind DHM's own website automation and AI Receptionist builds, or a general CRM like HubSpot, handles the fuller chain: a new lead reply, a nurture sequence, a review request, and a win-back nudge, all triggered off one connected contact record instead of several disconnected tools. The tradeoff is a separate monthly cost and, for shops not already using one, a setup step connecting it to wherever job data already lives. See the current integrations list for what connects without a rebuild.
Three real doubts about automating this
Only if the message reads like a robot wrote it. A short, plainly-worded text from a real person's name, sent fast, reads as attentive, not automated. The HBR data above points the other way from what most owners worry about: speed matters more to how a lead perceives the business than polish does, especially on the first reply.
Not necessarily. Check first whether your current platform already has a review-request or email add-on, the way Jobber, and increasingly its competitors, now do. The real question isn't which logo is on the tool. It's whether it can trigger off an event, a job completed, a year of silence, not just send a one-off newsletter to everyone at once.
Yes, if it isn't scoped correctly, which is why this guide doesn't repeat those rules here. A one-time, transactional text (a missed-call reply, a booking reminder) runs on a lighter consent standard than a recurring marketing message. The full consent and opt-out rules for email live on the email marketing guide, and for text on the SMS compliance guide, both worth reading before any sequence goes live.
Questions we get
Frequently asked questions
What is marketing automation, exactly, for a contractor?
Software that runs a repetitive marketing or follow-up task automatically once something triggers it: a form submission, a completed job, a customer gone quiet for a year. The benefit isn't creativity, it's consistency, the follow-up happens every time, not only on the days someone remembers.
What should I automate first if I'm starting from nothing?
The instant reply to a new lead. It's the one moment with the clearest, best-documented cost to getting it wrong: a Harvard Business Review audit found firms contacting a lead within an hour were nearly 7x more likely to qualify it than firms that waited just one more hour.
Does marketing automation replace an AI receptionist or staff answering the phone?
No. Automation handles the parts that don't need judgment, a text acknowledgment, a scheduled reminder, a review request. An AI Receptionist or a person still has to hold the actual conversation. The two work together, covered together on the website automation page.
Is an automated text message review request legal to send?
It depends on consent, not on whether the message is automated. The rules are the same ones that govern any marketing text, covered in full on the SMS marketing compliance guide, including the narrower, lighter standard that applies to a purely transactional message like a booking reminder.
How fast does a new lead actually need a reply?
As close to immediate as possible. The HBR audit behind this guide found the odds of qualifying a lead fall off within the first hour, and fall off sharply again after 24 hours. An automated acknowledgment buys the minutes it takes a real person to call back.
Does automating review requests violate Google's policies?
Asking every customer the same way, through an automated send, is not the problem Google's review policies target. What's against the rules is incentivizing, gating (only asking happy customers), or faking reviews. An automated request that goes to every customer, good experience or not, is the compliant version. Full detail on the actual lines not to cross is on the review guide.
Sources
- Oldroyd, J. B., McElheran, K., & Elkington, D., The Short Life of Online Sales Leads, Harvard Business Review, March 2011: the audit of 2,241 companies behind the lead-response-speed figures above.
- Jobber, Jobber's New Marketing Tools Help Service Pros Expand Their Businesses, March 12, 2024: the native Campaigns/Reviews launch and the accompanying homeowner survey figures.
- BrightLocal, Local Consumer Review Survey: timing data on when consumers expect to be asked for a review.
The 2011 HBR study is the largest published audit of lead response speed we could find; no comparable large-scale replication has been published since, so it's cited here as the best available evidence, not as current data.