Conversion tracking alignment in Sammamish
After Hours IT provides conversion tracking alignment for businesses in Sammamish, WA and the surrounding area. Sammamish home offices and small suites along the plateau rely on reliable VPN, backup, and someone who can drive out same night.
For urgent seo & marketing needs, we answer live 24/7 with same-night on-site dispatch to Sammamish when remote troubleshooting isn't enough.
Conversion tracking alignment: the morning the numbers stopped matching
On a Monday morning, a regional service company called us because their ad spend had gone up, but the lead count in their reports had gone sideways. The owner and office manager were looking at Google Ads, GA4, and their CRM, and none of the numbers agreed. Calls were showing in one place, form fills in another, and chat leads were being counted twice in some reports and not at all in others. They weren’t asking for a marketing makeover. They just needed to know which leads were real so they could make decisions without guessing.
We treated it like a tracking problem first, not an ads problem. After a quick intake, our engineer asked for access to Google Tag Manager, Google Ads, GA4, the website CMS, and the call tracking platform. We also asked for a sample of the forms being used, the chat widget name, and the conversion actions currently imported into Ads. That first call usually tells us a lot: whether the issue is duplicate tags, broken thank-you pages, missing event parameters, or a simple mismatch between what the business counts as a lead and what the platforms are counting.
In this case, the business had recently updated their site and added a new chat tool, but nobody had re-checked the conversion setup afterward. The result was a messy stack of overlapping events. One form was firing both a native submit event and a GTM event. Phone calls from the call rail were being imported as conversions in Ads, while the same calls were also being tracked as GA4 events. The office manager had the right instinct: the tracking had drifted, and the reports were no longer trustworthy.
What we checked behind the scenes
We started in Google Tag Manager to map every tag, trigger, and variable tied to lead activity. Our engineers reviewed the container version history to see what changed during the site update, then tested the live pages with GTM Preview and Tag Assistant. That let us confirm which events were firing on page load, form submit, click-to-call, and chat open. We also checked whether the site was using gtag.js directly in addition to GTM, because double-installing analytics is a common cause of duplicate conversion data.
Next we moved into GA4 and Google Ads. In GA4, we reviewed the event stream, conversion markings, and parameter names such as form_name, lead_type, and phone_number. In Google Ads, we checked whether conversion actions were set to count every conversion or one, whether they were imported from GA4 or created as website tags, and whether enhanced conversions were enabled. We also verified attribution settings, lookback windows, and whether any offline conversions were being uploaded from the CRM.
For calls and chat, we checked the vendor settings and the event handoff. The call tracking platform was generating a lead event on call answer, but the business also had a separate event for calls over a certain duration. That meant one real conversation could be counted twice if the thresholds overlapped. The chat widget was firing a lead event on widget open instead of on completed chat submission, which inflated the numbers. We documented each source, the exact trigger condition, and whether it should count as a primary conversion or remain a secondary signal.
The fix: clean events, clear rules, and fewer surprises
Once we knew where the noise was coming from, we cleaned up the event flow. In GTM, we removed the duplicate form submit trigger and standardized the form success event to fire only on the confirmed thank-you state, not on button click. For forms that used AJAX, we switched to listening for the site’s custom dataLayer event instead of a generic submit trigger. That made the conversion fire once and only once, which is what the owner wanted when someone actually completed a request.
We then aligned the platform settings so each lead type had a single source of truth. Google Ads kept the primary conversion actions for qualified calls and completed forms, while GA4 retained the broader engagement events for reporting. The chat widget was changed to fire only on completed chat leads, not on opens or first messages that never went anywhere. We also reviewed consent mode behavior so tracking respected the site’s cookie banner and didn’t create gaps from tags firing before consent was granted.
For call tracking, we adjusted the duration threshold and excluded internal test numbers so staff calls would not pollute the reports. We also confirmed that dynamic number insertion was working correctly on mobile and desktop, and that the tracking script loaded after the consent state was known. Once the setup was stable, we published a new GTM container version, tested all key paths again in Preview mode, and ran live test conversions to verify they appeared in GA4 and Google Ads with the expected event names and values.
Follow-up: documentation, monitoring, and a clean handoff
After the fix, we didn’t just send a thumbs-up and disappear. We documented the full conversion map in plain language: which events count as primary leads, which ones are only for analytics, what fires them, and where each event appears in Ads and GA4. We also included screenshots of the GTM triggers, the final conversion action list, and the call tracking settings so the office manager had something usable if the site changed again later.
Because this was a scheduled alignment project, we set up a short monitoring window and watched the data for several days. We checked for duplicate fires, missing parameters, and sudden drops in event volume. We also compared the lead totals against the CRM and call logs to make sure the numbers were close enough to support planning. In this case, the reports settled down quickly: fewer phantom leads, no more double-counted calls, and a much clearer picture of which campaigns were producing actual inquiries.
By the end of the engagement, the owner could look at one dashboard and understand what was happening without cross-checking three different systems. That’s the real point of conversion tracking alignment. It’s not about chasing perfect numbers for the sake of it. It’s about making sure calls, forms, and chat events reflect the same business reality in ads and analytics, so the next budget decision is based on something solid.