Good Equipment Is the Cheap Part: What Six Years of HVAC Procurement Taught Me

I manage purchasing for a commercial property company with 14 buildings and about 338,000 square feet. I am not an HVAC engineer, and I have never pretended to be one. My job is to approve replacements, watch the budget, and explain why a line item called Carrier air conditioner repair keeps showing up in my reports.

For the first few years, I blamed the equipment. That was my first assumption error. It was not the last.

Let me rephrase that: equipment is usually the result, not the cause. The cause starts before the purchase order.

The surface problem: new units should not create the same repair tickets

In 2023, one of our office suites had a cooling problem that had never been fully resolved. When I finally got the budget to replace the old system, I told the contractor I wanted more capacity. I assumed bigger meant safer, especially for a space that had always run warm.

The contractor did not push back hard enough. We installed a larger system in the same space, using the existing ductwork. The equipment was not defective. But it was too large for the actual cooling load.

What did we get? Short cycling, poor humidity control, a thermostat that never seemed satisfied. And before that cooling season ended, another Carrier air conditioner repair ticket. The invoice was not for a broken compressor. It was for a system operating under conditions it was never designed for.

The deeper problem: we choose a product before we understand the problem

After that, I started looking at every equipment decision differently. I noticed a pattern that has nothing to do with any one brand. We search for a product before asking an engineering question.

I see this in search phrases like what's the best hepa air purifier. The answer is meaningless until you know the room size, the CADR rating, and the specific air quality issue. Without that, the best purifier is just a fan in the wrong room.

For example, searching for carrier infinity heat pump tells you the category of equipment. It does not tell you whether the system is actually sized, matched, and set up for your building. That part decides the outcome.

The deeper cause is that our purchasing process rewards two numbers: the model number and the bid price. Nobody has long-term responsibility for what happens after startup. The installer gets paid when the system starts. The next problem becomes my repair budget.

The 115V mini split assumption

A second example: We had a small office where running a new high-voltage circuit looked expensive. On paper, the easy answer was a mini split 115v model. It seemed cheaper and less disruptive, and I wanted to approve it fast because the tenant was waiting.

What I missed was that the existing outlet was already on a loaded circuit. The mini split needed a dedicated circuit, not a shared outlet. The lower-voltage model did not remove the electrical work; it just delayed the discovery until after we ordered the equipment. In the end, that project cost more than a standard installation would have from the beginning.

I still kick myself for that one. I let the convenience of a quick search result override the site evaluation that we normally do first.

The smart plug search is the same mistake

That is also why I understand searches like alexa smart plug outdoor. People want remote control without paying for a controls package. The intent is reasonable, but a compressor is not a string of patio lights. If you connect a mini split to a smart plug that is not designed for motor loads, you are building tomorrow's repair call, not solving today's problem.

The better question is whether the equipment supports an approved Wi-Fi adapter or thermostat. That requires reading the manual, checking compatibility, and sometimes asking a trained technician. It is less convenient in the moment and much less expensive over time.

The cost of not looking below the surface

Here's the thing: a Carrier air conditioner repair call is not the real problem. It is a lagging indicator. The real problem lives in the sizing, installation, air distribution, control setup, or maintenance plan.

When I started asking repair technicians to record the probable cause, not just the failed part, our repeat call rate made more sense. A capacitor can go bad anywhere. But a brand-new system that loses its charge in the first year is a sign of a different issue.

The total cost of a bad decision also shows up in places that do not get coded as repair expense. It shows up in tenant complaints, wasted staff hours, and the quiet cost of interrupted work. That is why I keep saying that the lowest bid is not the real price.

What I do now: a short answer

If I had to fit this into a procurement policy, it would look like this:

1. Require a load calculation before any model is chosen. Ask for the number and the assumptions.
2. Make the startup report a condition of final payment. It should include static pressure, refrigerant charge, and the commissioning checklist.
3. Verify warranty registration in writing before the invoice is approved.
4. Ask for the manufacturer-approved control option before considering a smart plug or other workaround.
5. Treat the first repair ticket as a message, not a random event. Ask what decision caused it.

None of this is exciting. It does not make me look like the buyer who squeezed a vendor for the lowest price. But after six years of tracking invoices, the savings are real. Put another way: the best price is the one that does not create a second project later.