ServiceTitan + CallRail Integration Guide
Native ServiceTitan and CallRail integration. Call attribution on customer records, source and keyword data, and channel-level CAC math.
What it does
CallRail does the call tracking; ServiceTitan becomes the system of record. A CallRail website pool dynamically swaps the phone number a visitor sees based on how they arrived (Google Ads, Local Services Ads, organic, a specific landing page), so each call carries its true source. When the call comes in, CallRail forwards it to your ServiceTitan tracking number and pushes the attribution to ServiceTitan through the ExternalCallAttributions endpoint. The call appears on the ServiceTitan calls page, matched to an existing customer or used to create a new one, populated with source, medium, campaign, and the PPC keyword that triggered the visit. From the Marketing Pro dashboard you can then line up web session data and the full customer journey against booked revenue, turning a lump-sum ad spend into channel-level ROI. The integration itself is free; you pay only for your CallRail plan and Marketing Pro Ads.
Setup
Enable call tracking in CallRail and create a website pool (not single source-tracking numbers) so you capture visitor-level keyword and source data automatically. Install the CallRail DNI JavaScript snippet on every page before the closing body tag, and point the pool to forward to your ServiceTitan tracking number. In ServiceTitan Marketing Pro Ads, activate the CallRail integration using your Tenant ID, Client ID, and Secret. Run one DNI source per site: do not also run ServiceTitan native Dynamic Call Tracking on the same pages, or the two scripts fight over the swap. For full PPC attribution, set the optimized tracking template in Google Ads and Microsoft Ads so campaign, ad group, and keyword data reach CallRail. Never hardcode a pool number anywhere on the site, as that freezes the swap and kills your source and keyword data. Validate with 5 test calls from different sources and confirm each lands in ServiceTitan with the right attribution.
Reporting patterns
CAC by source = total channel spend / new customers from that channel, filtered monthly. With clean attribution the channel gaps are stark. In current home services benchmarks, Local Services Ads run roughly $53 per lead and about $233 per paying customer with a 43.9 percent book rate, while traditional Google Ads sit near $104 per lead and $472 per paying customer. A handful of channels usually deliver most of your booked jobs, and the long tail often runs 2-3x your average CAC. Aim for a customer lifetime value to CAC ratio of at least 3:1; channels that miss it are candidates to cut or rework. Tie revenue, not just leads, back to source, because a channel with cheap leads and a low book rate can quietly cost more per booked job than a pricier one.
How Pineido helps
CRM cleanup or SEO sprint includes website-pool and DNI setup, source mapping, the Google and Microsoft Ads tracking templates, attribution validation across channels, and a CAC-by-channel dashboard tied to booked revenue. 30-day fixed scope, $8,500.
How agentic AI takes this further
CallRail pushing attribution into ServiceTitan is automation, a call happens, a source tag attaches, done. It is genuinely useful and it is also inert, it will never tell you that Local Services Ads quietly crossed your acceptable cost-per-booked-job three weeks running, because attaching a tag is not the same as watching a trend and deciding something needs to change. A marketing dashboard with an "AI insights" panel that narrates the same CAC chart back to you in prose is not further along than the raw numbers, it is the numbers with extra words, nobody asked it to act.
A real agent perceives the trend, not the single call. It reads booked-job revenue and source from ServiceTitan, per-call attribution from CallRail, and campaign-level spend from the Google Ads and Microsoft Ads APIs, the same three data points the CAC-by-source math in the section above already depends on, just pulled automatically instead of assembled by hand each month. Those are the right inputs because CAC is only meaningful measured against actual booked revenue, not raw lead count, and that means joining all three systems, not just reading CallRail.
Deciding when a channel has actually drifted, not just had one noisy week, is where an LLM with retrieved context matters, comparing this week's CAC against the shop's own target ratio and its history of past reallocation calls an owner has already made. A channel above target for a single week does not clear the confidence bar for a recommendation; the same channel above target for three or four consecutive weeks does, because that is the difference between normal variance and a real shift worth acting on.
Acting on a confirmed drift means drafting a specific, reviewable change, a proposed daily-budget cut on a named Google Ads campaign, prepared through the Ads API but never submitted unattended. Each proposal is idempotent per campaign and per week so a rerun does not draft the same cut twice, and a failed API pull (a campaign renamed, an account disconnected) gets retried and then surfaced as a data gap rather than silently treated as zero spend.
Spend decisions do not fire on their own. The agent proposes the reallocation with the CAC numbers attached and an owner or marketing lead approves it before any budget actually moves, because pausing the wrong campaign on a bad read costs real bookings, not just a reporting error. It also will not touch ad creative or landing pages, and it will not permanently kill a channel, only ever recommend a change for a human to make.
This runs on your own ServiceTitan and CallRail accounts, workflow and source handed over at the end, not a rented reporting tool you lose access to the day you cancel. The realistic outcome is catching a channel's cost drift within a few weeks of it starting instead of at the next quarterly review, while there is still budget left to redirect. Pineido builds AI agents for small business into your own systems, one owned workflow live in 30 days.
Frequently asked questions
How much does CallRail cost?
CallRail Call Tracking starts at $45/mo with 5 local numbers and 250 local minutes. Conversation Intelligence or Form Tracking is $90/mo, and the Complete plan is $175/mo. Overages add roughly $3 per extra local number and $0.05 per extra local minute, and calls round up to the next minute, so most shops budget 1.5-3x the base price once volume kicks in. Confirm current rates on CallRail pricing.
Does it require ServiceTitan Marketing Pro?
Yes, the native integration is part of the Marketing Pro Ads Package and uses ServiceTitan ExternalCallAttributions. You need call tracking and a website pool forwarding to a ServiceTitan tracking number, plus your Tenant ID, Client ID, and Secret. Without Marketing Pro you can pass call data via Zapier, but with reduced functionality.
What changes after setup?
Calls auto-attach to customer records with source, medium, campaign, and keyword, and new callers create contacts automatically. Marketing reporting becomes channel-level instead of lump-sum, so you can pin booked revenue to the campaign that earned it and compute real CAC per channel.
Ready to ship?
Tell us where you are stuck. We will review it and send a focused recommendation within one business day.
Get operations insights weekly
Join service-business operators who get one actionable insight on ops, automation, and growth every week.
No spam. Unsubscribe anytime.
You’re subscribed! Look for our next issue in your inbox.
That did not go through. Please try again, or email [email protected].