By Greg Allan, tradesman and creator of Data First Coaching
Why does quality fall apart once another cleaner is on the roster?
Quality falls apart because the standard the client was promised never left the owner's head. The owner knows the office needs the kitchen bench wiped a certain way and the bond clean needs the tracks and skirting checked before it is handed back, but if that knowledge only lives in the owner, every new cleaner on the roster is guessing.
Most cleaning businesses run jobs the owner physically cannot stand over. The end of lease clean gets handed back at five o'clock without the owner there. The office is done at midnight by whoever is on that night's roster. The mould job or water damage callout happens in someone's home while the owner is quoting the next job across town. Without something holding the standard in place, the standard only exists when the owner happens to be watching.
Why do new cleaners not do it the way I showed them?
Because a new cleaner does not arrive blank. They carry habits from whatever company trained them last, and those habits do not match the way this business promised the client the job would be done. In a trade with constant staff turnover, that is not a one-off problem, it repeats every time someone new joins the roster.
An owner can show a cleaner the right way to finish a carpet job or the right chemicals to use on a stone bench once, twice, three times, and still find the old habit creeping back in on a site the owner never checks. That is not the cleaner being careless. It is the business having nothing in place to override what they already know. Repeating the same instruction over and over is a sign the business is relying on memory, the cleaner's and the owner's, instead of something that holds regardless of who is rostered on.
What actually holds a cleaning business together when the owner isn't on site?
What holds it together is a system for each job type, built from the owner's own standard, that does not depend on which cleaner picked up the roster that week. That system needs a way to check the job was actually done the promised way, not just trust that it was.
This is where the owner's own data tells the truth that memory cannot. Which contracts have had complaints in the last run of jobs. Which cleaner's sites get the most callbacks for a missed spot or a redo. Which job type, carpet, commercial office, end of lease, restoration, has the thinnest margin once the chemicals, equipment and rework are counted. A feeling about who the good cleaners are is usually a few weeks out of date. The complaint and rebook pattern in the business's own numbers is not.
This is also where AI agents change what the owner carries. An AI agent is software that does a job the way the owner would do it, using the owner's prices, wording and rules. Instead of the owner chasing a quality check photo after every bond clean or following up a strata manager about a missed inspection, an agent can hold that step in place and flag it the moment it is missed, before the client ever has to raise it. That is data first: the business runs on what the job and complaint data actually shows, not on the owner's memory of how last month went.
How do you know which cleaner needs more training before a client finds out?
You know by breaking the role down into its exact steps and checking where a cleaner's result actually drops off, rather than waiting for a client complaint to tell you. Picture a cleaning business owner who has just taken on a new commercial contract with a strata manager, and two cleaners are sharing the roster across three sites.
One cleaner's sites keep coming back clean on every check. The other's keep needing a redo on the same kind of area, every time. Without breaking the job into its steps, that looks like one cleaner being worse than the other. Looked at properly, it is usually one missing step, one habit from an old job, repeating itself. Finding that gap early, from the job data rather than from a lost contract, is what lets an owner upskill fast instead of being found out on a client's site.
Can off-the-shelf cleaning software run all of this, or does it need to be built around how we work?
Some off-the-shelf tools genuinely suit parts of a cleaning business, rostering across multiple sites or logging recurring cleans, and the data on how they are actually used shows which ones earn their place. The case for owning systems built on your own job and client data is strongest where your contracts, your quality checks and your pricing are specific to how you run commercial work, bond cleans or restoration callouts, which no tool built for every trade at once was shaped around.
As a cleaning business takes on more commercial contracts and rosters more staff, the admin behind that growth has to be built to carry it, otherwise it lands back on the owner as more phone calls, more chasing, more checking. AI agents taking on that non-billable office work, following up a missed quality check, chasing a contract renewal date, logging an insurance job's paperwork, means the people in the business spend their time on the client and the crew, not on administration that exists only to hold the business together.
How does Data First Coaching approach this?
Data First Coaching starts from the cleaning business's own job and complaint data through the Trade Business Intelligence, not from the owner's description of what usually goes wrong. It looks at which contracts, sites or cleaners are actually producing the rebooks and the redo callouts, because that is where the real gap sits, not necessarily where the owner assumed it was. Greg Allan built this from running his own trade business, and his teaching asks “Why don't they follow my instructions?” because the answer usually sits in a system that was never built, not in the cleaner. The how of building that system is worked through inside the Academy, with members, because the Trade Business Intelligence is powered by every member for every member, and that only works for owners who are open minded and willing to share what they learn.
