Lonta AI | HVAC results: weather and operations review
https://lonta.io/blog/hvac-marketing-weather-adjusted-results/
Version: October 8, 2026. A claim review, not a causal or statistical model.

Claim being reviewed:
Outcome definition:
Earlier period / later period:
Equal follow-up cutoff for both acquired-inquiry groups:
Weather station, source and temperature method:
Degree-day base (the article uses 65 F):
Missing weather days and unresolved record gaps:

FOR EACH PERIOD
Advertising spend:
Heating degree days:
Cooling degree days:
Available technician hours:
Staffed phone hours:
Completed, collected jobs attributed to advertising:
Repair / replacement / maintenance split:
New in available records / returning / unknown customer split:
Collected revenue and contribution where available:
Advertising spend / completed, collected jobs:

CHANGE RECORD
Date / account or operational change / approver or evidence reference:
Conversion-definition changes that break comparability:

REVIEW
What improved in the recorded totals:
Alternative explanations (weather, spend, capacity, hours, mix, tracking):
Earlier periods / prior-year reference and differences:
Supported conclusion:
What remains unresolved:

Do not treat missing temperatures as zero. Keep heating and cooling separate.
Jobs divided by degree days are not automatically weather-adjusted performance.
No independent agency effect or statistical significance is established here.
Reuse this worksheet with attribution to Lonta AI and the source article.