the eight systems every trades company runs on

If you walk into the back office of a $10M to $100M-revenue construction, HVAC, restoration, or field-services company, you will find, with remarkable consistency, eight kinds of software running. The brand names change. The shape does not. This piece walks through the eight, names the usual suspects, and — most importantly — shows where the work lives in the gaps between them.

This is the pattern we see when we’re called in. It’s also why integration work, not “replace your ERP,” is usually the highest-return project on the table at this stage of a company’s life.

the eight

1. accounting / GL. QuickBooks in maybe 70% of shops this size, with Sage 100 / 300 and Foundation in the runner-up slot. This is the financial source of truth. Everything eventually flows here; nothing flows out without manual export.
2. job-cost / project accounting. Sometimes the same system as accounting (Sage 300 CRE, Procore, a custom build), sometimes a separate one bolted on. Holds budgets, commitments, POs, change orders, labor burden.
3. scheduling / dispatch. ServiceTitan or HousecallPro in service trades; Procore or Buildertrend in construction; often a whiteboard, a spreadsheet, or a custom tool built by the nephew in restoration and smaller shops.
4. CRM and sales pipeline. HubSpot, Salesforce, or — very commonly — a Google Sheet the owner guards. Bids, proposals, and customer relationships live here.
5. field app (crew-facing). ServiceTitan mobile, Procore mobile, a custom app, or, in half of shops, text messages and phone calls.
6. document storage. A Dropbox or Google Drive or Sharepoint folder with a thousand PDFs of signed contracts, permits, and change orders. Nobody can find anything without asking Sharon in the office.
7. payroll and HR. ADP, Paychex, Gusto. Often disconnected from job-cost, so crew hours are re-keyed.
8. the vendor portals. The supply house portal, the equipment-rental portal, the city-permit portal, the insurance portal. Each one is a website an office person logs into weekly and copies data out of.

the actual map (what we draw on the whiteboard)

On paper, these should be a connected graph. In practice they’re a hub and spokes, and the hub is a person. A typical shop has one office employee (let’s call her Sharon) who:

• Takes a timesheet from the scheduling tool, re-types it into payroll.
• Pulls job-cost from the project system, re-types it into QuickBooks.
• Downloads a supply-house invoice PDF, matches it to a PO, and enters it into QuickBooks by hand.
• Copies customer info from the CRM into the job-cost tool when a bid gets signed.
• Emails a daily spreadsheet to the owner because no single system can produce it.

Sharon is the integration layer. When she goes on vacation, five things stop working. When she retires, five years of tacit knowledge about which fields matter and which ones lie goes with her.

where the leverage is

The instinct — usually from a consultant — is to replace one or more of the eight tools with something newer, more modern, or more all-in-one. This is almost always the wrong call at this stage. The tools themselves mostly work; the data inside them is mostly correct; the team mostly knows how to use them. Ripping out and replacing introduces huge retraining costs and breaks the business rules encoded in the current workflows.

The leverage is in the connectors — the seven or eight small integrations that move data between the tools automatically, so Sharon doesn’t have to. In order of typical ROI:

the seven high-leverage connectors

1. scheduling → payroll. Crew hours pulled from dispatch directly into payroll. Usually eliminates the biggest source of data entry and one of the biggest sources of error in the shop.
2. job-cost → GL. POs, labor burden, and commitments posted to accounting as they happen instead of at month end. Gets the owner a real cash position mid-month.
3. supply-house portals → job-cost. Vendor invoices auto-coded to the job and posted against the PO. Closes the biggest leak in margin on most jobs.
4. CRM signed-bid → job-cost project creation. When a bid gets signed, the job setup happens automatically, with the customer, budget, and schedule pre-populated.
5. document folders → job-cost. Tagged PDFs automatically linked to the job record. This is the “can we find the signed change order” problem, solved.
6. field app → daily log → owner’s email. The foreman’s end-of-day notes become a one-paragraph summary by job, delivered to the owner at 6pm. This is the single most popular automation we deploy.
7. AR aging → alert. Customers who cross a threshold (45 days, 60 days) trigger a message to the PM. Recovers real money in the first month.

why this is the shape of the work, not custom software

A shop at this size rarely needs its own ERP. It needs its eight systems to cooperate, plus a thin layer on top that answers the questions nobody can answer today: what’s my cash position, what’s the margin on job 4412, who is the customer I’m owed the most money by. Once the connectors exist, that thin layer becomes cheap to build — because the data is finally in one place.

The AI-driven front end sits on top of that layer. It’s not replacing any of the eight tools; it’s the thing that finally lets the owner ask their business questions in the language they’d ask Sharon, and get answers faster than Sharon could give them.

where to start

Not with all seven. Pick the one connector whose absence bothers you most this week. That’s the one that will pay for itself first. If you don’t know which one, it’s usually scheduling → payroll, because you can measure the savings in hours saved the first pay period.

We’ve written about three starter connectors in more detail, with rough prices, in a separate piece.

---

Want to talk through this in the context of your shop? Talk to a builder. No pitch deck, no sales motion — just a conversation.