The Hidden Power of Automation

A client of mine runs a residential roofing company in southern Alberta. Forty-person crew, decent pipeline, healthy reputation. When we first sat down two years ago, he told me his biggest problem was that his estimators were drowning. They were good at their jobs and clients liked them, but every week one or two estimates would slip — go quoted but never sent, or sent but never followed up on, or followed up on too late. Each missed estimate was probably a $14,000 job. He was losing two of them a week, every week, and he thought the answer was hiring another estimator.

I asked him to walk me through what an estimator’s day looked like, in actual minutes. He pulled out a notebook. By the time he was done, the picture was clear: the people I was paying $80,000 a year to assess roofs and price jobs were spending about forty percent of their week on data entry, copy-paste between three software systems, and chasing signatures on PDFs. They were doing the work of administrative assistants for half their week, and the result was that the actually-skilled work was suffering for time.

This is what most small businesses look like when you take the lid off. Not understaffed. Not undertrained. Just leaking time into a hundred small, mechanical tasks that nobody set out to make a job, but that crept in over years.

The automation work I did for him was not glamorous. We connected his CRM to his estimating software, so when a lead came in by phone or web form, a job record materialized automatically with all the right fields populated. We connected his estimating software to his contract-signing tool, so when an estimate was approved internally, the contract draft appeared in the signing tool with the correct numbers and a routing rule for the right estimator’s signature. We connected the signing tool to his accounting software, so when a contract was signed, an invoice and a deposit-request were generated and queued for review. Three small connections. None of them visible to a customer. Each one took us about two days to build and test.

The result, six months in, was that his estimators got back the equivalent of a day and a half a week each. The work that was slipping stopped slipping, because the system was handling the parts that were causing the slippage. He didn’t hire the second estimator. He gave the existing two a raise out of the savings, and they were happier than they’d been in years.

I tell this story because the version of it most people imagine when they hear “small-business automation” is wrong in two specific ways.

It’s not about replacing people. The estimators in this story are still doing the estimator work. They are just doing it without the friction of being a part-time data clerk. Nobody got laid off. Nobody got their hours cut. The work the humans were good at got more space, and the work the humans were bad at — and resented — got removed.

It’s not about big technology projects. The work was three small connections between tools the business was already paying for. None of it required a developer on staff. None of it required a six-month implementation. The whole project was scoped, built, and live in under three weeks of part-time work, and most of that was waiting for vendor permissions.

The hardest part, in fact, isn’t the technology. The hardest part is the diagnosis — figuring out which fifteen-minute task, repeated forty times a week, is actually the source of the leak. Owner-operators don’t see those tasks because the tasks happen by reflex. The estimator doesn’t think of “copy ten fields from the CRM into the estimator” as a task; it’s just what you do before you start an estimate. The leak is invisible until someone with fresh eyes lists it back to them.

The team here at DataSmart does this kind of work as a service. We call it workflow automation, which is a deliberately boring name, because the work itself is boring on the surface — small connections between small tools — and transformative underneath. If your business has a feeling of “everyone’s working really hard and we’re somehow still behind,” the diagnosis is almost always not “we need more people.” It’s “we need fewer manual handoffs.”

Find the tasks that are repeated forty times a week. List them out, with how long each one takes. Multiply. The hidden tax always exceeds expectations. The savings, once you remove it, always exceed expectations too. The roofing client got a day and a half a week back per estimator. Yours will be different. But it will, almost without exception, be much more than you guess on the first try.