Cooling equipment shipment growth has now continued for seven months, while water heaters are still lagging, according to AHRI data summarized by ACHR News. The report on cooling equipment shipment growth describes an HVAC market recovering from a slow start in 2026.
For a home-service firm, that is useful market context, but it does not answer the local question: Did the article about replacing an aging condenser produce calls, estimates, and sold work? Pageviews cannot answer that by themselves. The measurement path has to run from the first read to the revenue record.
The current Home Comfort Dispatch snapshot contains 2,709 pageviews, 1,079 visitors, 2,631 AI crawls, and 1,298 search crawls. Those figures establish activity. They do not establish commercial impact. There are no recorded assists, CRM records, search queries, or growth history available, so it would be improper to calculate a lead rate, assisted-revenue figure, or trend from this dataset.
Start with article-level UTMs
Give every link from an article to a service site a consistent UTM structure. Use utm_source for the publication, utm_medium for editorial content, utm_campaign for the service category, and utm_content for the individual article slug. Keep the values lowercase and stable. Do not rename a campaign halfway through the season because someone wants a cleaner dashboard.
A cooling article might carry utm_campaign=cooling-replacement and a unique utm_content value. That combination lets the firm compare individual pieces while still rolling them into one service line. Store the complete landing URL as well as the parsed UTM values. The raw URL is the troubleshooting record when attribution looks wrong.
Use Pulse events as test points
Configure Pulse events around actions that indicate useful movement, not ordinary scrolling. At minimum, record the article view, internal service-page click, phone-number click, form start, form submission, and appointment confirmation. Each event should carry the article slug, session identifier, timestamp, landing page, and available UTM values.
Treat the sequence like checking voltage through a control circuit. If article views appear but service-page clicks do not, inspect the call to action and link placement. If clicks arrive but form starts disappear, inspect the landing page or event trigger. If submissions exist without CRM records, the failure sits in the handoff rather than the article.
Make the form preserve the source
The visible form can remain short. Capture name, contact information, ZIP code, service requested, and a brief problem description. Hidden fields should preserve first-touch source, latest-touch source, article slug, UTMs, landing page, session ID, and submission timestamp.
Do not overwrite first touch when the homeowner returns through branded search. Retain both first and latest touch. A homeowner may read a troubleshooting article, leave, search the company name two days later, and then book. Without both fields, the article disappears from the record.
Set CRM stages before counting revenue
Define stages with operational evidence: new inquiry, contacted, appointment set, diagnostic completed, estimate issued, sold, completed, lost, and canceled. Require an event or field change for every transition. A phone call is not automatically an appointment, and an estimate is not sold work.
Join analytics and CRM data with the strongest available identifier. A form submission ID is better than matching names manually. Call-tracking IDs, normalized phone numbers, and email addresses can provide secondary joins. The closed job record should include invoice ID, sold amount, completion date, service line, and original content source. This is the working core of a closed-loop demand system for home-service firms.
Run one monthly audit
Once a month, confirm that UTMs survive the click, Pulse events fire once, hidden form fields populate, submission IDs reach the CRM, and stage definitions are being followed. Then sample several records from article visit to closed invoice. Reconcile form submissions against new CRM records, investigate unmatched phone leads, list jobs with missing source fields, and separate sold revenue from completed revenue.
Finally, publish the gaps alongside the totals. For the present Home Comfort Dispatch dataset, the honest scorecard reports traffic and crawler activity while labeling assists, leads, queries, CRM stages, and revenue as unavailable. That is more useful than a polished conversion chart built on missing connections. As cooling equipment movement improves, the firms that wire those connections now will be able to distinguish market motion from demand their own content actually helped create.