TRC Blog

The Search for Perfect Collection Data

Posted Nov 17, 2017 by Ryan Kiscaden

Recently, I read Brian Loftus’ September 2017 Data-Driven Newsletter.

Loftus, HARDI’s Market Research & Benchmarking analyst, made the general observation about the impact of weather on equipment sales. He was specifically referring to warm weather in August 2016, which saw higher sales of ducted air conditioners and ducted heat pumps when compared with August 2017. The culprit for the lower sales this year, he suggested, was an unusually warmer August the previous year.

Coincidentally, I also came across AHRI’s recent statistical release that tracked HVACR equipment shipments for August. It compared three shipment categories: numbers of air conditioners, air conditioners and heat pumps, and only heat pumps for August 2016 and 2017. Essentially, the AHRI data mimicked Loftus’ observation.

U.S. shipments of central air conditioners and air-source heat pumps in 2017 were down 5.5 percent when compared with 2016. Shipments of air conditioners decreased 9.4 percent, according to AHRI. (Admittedly, shipments of air-source heat pumps rose 3.3 percent).

Then came the cloudburst of an idea. I thought of TRC’s collection efforts and wondered how they would compare for August 2016 and 2017.

Our August 2017 TRC’s collections were slightly less than in August 2016, with a decrease of 1.2 percent.

What does all this mean? In a general sense, I wanted to know whether our collection efforts parallel the sales of equipment. If the linkage is inescapable, it can affect TRC’s marketing efforts and certainly our forecasting attempts.

Admittedly, we probably wouldn’t conclude that this is an immutable law (they all parallel each other), because there are several caveats to consider.

The first is an outlier effect. No matter the inference, something that happens out of the ordinary can skew any data average. Think of Hurricane Harvey, Irma, and Maria.

Second, there is the concern about narrow vision. You have the danger of looking at one data point, as we just did, and face the error of extrapolating a broader interpretation that simply doesn’t exist or is inaccurate.

Then there is the obvious connection to both weather and geography. Weather does have an impact on sales. And there is repeated evidence that geographic locations influence sales. At TRC, we know from years of collecting thermostats that the Midwest weather states are more bountiful at recycling. We call it the Goldilocks Zone.

The final element is the growing impact of market forces. The reliable method of predicting recycling efforts occurs when someone replaces a thermostat during a repair or retrofit.

Previously, we never would have thought that someone would run out to buy a thermostat because it was cool (forgive the pun) — until now. Whether it’s a Nest or a smart thermostat that allows you to connect to an embedded web server or receive updates via email, it’s one of the new, hip things to get. It exudes gadgetry and more options, and some people find that appealing.

At TRC, we soldier on, committed to our mission of collecting mercury-filled thermostats, regardless of weather, location or sales. However, we also know that the outside forces I have mentioned will affect our results.

You might infer from our efforts at examining these factors that we’re on the constant treadmill of self-examination. That would be a fair assessment. We’re always scrutinizing our recycling efforts and asking how we can improve our numbers. By viewing outside forces, it gives a more reliable approach to planning. And there is the motivation factor too. We like pushing the proverbial envelope that makes us work a little harder, a little smarter, to better succeed in the future. It’s a challenge we embrace.