Amazon ships products before you order them. And it’s been doing it for over a decade.

My wife ordered something from Amazon one morning. Before sunset, the item was sitting on our porch. She was impressed.

We don’t live in a large metropolitan area. Our town has a population of less than 50k people. The item she ordered wasn’t something you’d buy daily. It wasn’t on any top sales chart.

Yet, Amazon had at least one of those items in a nearby warehouse, ready to deliver within hours.

When I told her that Amazon had most likely shipped it before she even clicked “Buy Now,” she didn’t believe me.

But it’s true.

What makes it even more astonishing is that this isn’t new technology. I wrote extensively about it in one of my PhD papers two years ago. This technology has been around for over a decade.

How Anticipatory Shipping Works

In January 2014, Lance Ulanoff wrote an article titled Amazon Knows What You Want Before You Buy It. It breaks down how Amazon’s patented algorithm, named “Anticipatory Shipping,” works.

The concept is straightforward: Amazon ships packages to a destination geographical area without a specific address. The final destination is determined while the package is en route.

Here’s my take on how it works: The algorithm uses millions of data points Amazon collects—browsing history, past purchases, wish lists, shopping cart additions—to create a confidence level ranking. When a customer reaches a certain threshold in that confidence level, indicating a purchase is highly likely, the shipping process is triggered.

The item moves to a regional warehouse near the customer. Then, when the customer finally clicks “Buy,” the package is already nearby, ready for same-day or next-day delivery.

Of course, this wouldn’t be sustainable for most businesses. Amazon collects more data per hour than the average company collects in an entire year.

But that doesn’t mean you can’t use your data to make predictions.

You don’t need millions of customers or petabytes of data. You just need to look at the data you’re already collecting—and most service businesses are sitting on years of untapped insights.

The Service Industry

Let’s use an HVAC company as an example. The internal data alone could increase your sales and provide you with more information than you know what to do with.

During every service call, you’re already gathering information: the HVAC unit brand, size, model, age, service history, and repair frequency. Depending on how long you’ve been in business, you now have years of data.

Start there.

Let’s say you’ve been servicing HVAC units for five years. Your records show:

→ Customer A installed a Carrier unit in 2015. Average lifespan for that model? 12-15 years. They’re approaching year 10.

→ Customer B’s unit has had three compressor repairs in the last 18 months. It’s on borrowed time.

→ Customer C always calls in July when temperatures hit 95 degrees. It’s June, and it’s already 90 degrees.

You now know:

  • Who to call about replacement units (Customer A is entering the replacement window)

  • Who needs a proactive conversation before their unit dies during a heatwave (Customer B)

  • Who to schedule for preventive maintenance next month (Customer C)

That’s not magic. That’s just paying attention to patterns in data you already have.

The impact:

If you call 50 customers approaching the 10-year mark and 20% convert to new installations at an average of 8, 000perunit, that’s80,000 in revenue from data you were already sitting on.

If you’re focusing on maintenance, you can use the same data to optimize your operations. If 70% of your customer base uses Carrier units, stocking Carrier parts on every truck reduces return trips and improves turnaround time.

This is exactly how Amazon thinks—except they’re predicting what you’ll buy, and you’re predicting what your customers will need.

How Other Industries Can Apply This

The principle is the same across service businesses: use the data you’re already collecting to predict needs and prevent problems.

Plumbing Companies

You’ve serviced water heaters for years. Your records show patterns:

→ Customer A installed a 40-gallon tank in 2016. Average lifespan? 8-12 years. They’re in year 9.

→ Customer B has called three times in 18 months for “no hot water” issues. That unit is failing.

→ Customer C calls every winter when their pipes freeze. It’s November.

Predictive action: Call Customer A about replacement before their heater fails on Christmas morning. Offer Customer B a diagnostic before the next emergency. Send Customer C winterization tips and a service discount now, before the freeze.

Result: More planned installations, fewer emergency calls, happier customers, better cash flow.

Property Maintenance Companies

You track every work order: HVAC repairs, plumbing calls, electrical issues, appliance replacements. Patterns emerge:

→ Building A has had four HVAC calls in six months. Those units are either undersized or poorly maintained.

→ Property B replaces dishwashers every three years like clockwork. They’re buying cheap units that don’t last.

→ Complex C has seasonal plumbing issues every winter. Likely a pipe insulation problem.

Predictive action: Propose an HVAC assessment for Building A before the next failure. Show Property B the ROI of better appliances. Offer preventive pipe winterization for Complex C in October, not January.

Result: You shift from reactive—“fix what breaks”—to strategic partner—“prevent what will break.” That’s where the real money is.

Auto Repair Shops

Your service records contain goldmines:

→ Customer A is at 95,000 miles. Timing belt replacement is typically needed at 100,000 miles.

→ Customer B gets oil changes every 5,000 miles like clockwork. They’re due in two weeks.

→ Customer C’s brake pads were at 30% during their last visit six months ago. They’re probably due now.

Predictive action: Send Customer A a reminder about the timing belt service with a discount. Email Customer B their oil change reminder. Call Customer C about a brake inspection.

Result: Customers appreciate the heads-up. You fill your schedule with planned work instead of waiting for emergencies. More importantly, now the customer is not looking for the first available place to fix their brakes; they are coming to you.

These examples all follow the same pattern: recognize what you’re already tracking, identify predictable needs, and reach out proactively.

It sounds simple, because it is. But most businesses never do it because they’re too busy reacting to what’s already broken.

I learned this lesson the hard way in my own business.

How I Used Data to Find $10K in Waste

My company was painting apartments. I noticed our paint usage had drastically increased, which was shrinking my already slim profit margin.

Using the data I had already collected, I analyzed where the usage had increased.

It didn’t take long to figure out.

In one of the projects, a painter was throwing away good paint. When his sprayer stopped sucking paint from the bucket—usually when about 10% remained—he’d toss the bucket and grab a new one. For the sake of speed.

Once I identified the problem, I had a conversation with the painter. We calculated he was wasting about 200perweekonpaint.Over10,000 annually.

A simple change fixed the issue. I would have a helper pick up the buckets and combine them, saving thousands of dollars. But I only caught it because I was tracking paint usage per unit.

These types of inefficiencies happen every day in service businesses. The difference between profitable companies and struggling ones often isn’t a matter of skill or effort. It’s visibility.

The Takeaway

You don’t need Amazon’s budget or a PhD in data science. You just need to start looking at the data you’re already collecting.

Three questions to ask yourself:

  1. What patterns exist in my service history that I’m ignoring?
    Are there predictable timelines for replacements? Seasonal issues that repeat? Equipment that fails at specific intervals?

  2. Which customers are approaching a decision point?
    Who’s nearing the end of their equipment’s lifespan? Who’s had multiple repairs indicating imminent failure? Who’s showing early warning signs?

  3. Where am I bleeding money without realizing it?
    What inefficiencies are hiding in my operations? What waste exists in materials, time, or processes? What problems repeat that could be prevented?

The data is already there. You just have to look.

Amazon didn’t revolutionize retail by inventing new technology. They revolutionized it by paying attention to patterns everyone else ignored.

Your business has the same opportunity. The question is: are you looking?

What patterns have you discovered in your own business data? Drop a comment—I’d love to hear your stories.

*If you need help turning your operational data into actionable insights, that’s exactly what I do. Check out my services.