← All posts
ServiceTitanCustom dashboardsApril 26, 2026Clint Research Team

How to Build a Custom Dashboard in ServiceTitan (Including Atlas / Titan Intelligence)

ServiceTitan custom dashboards combine report modules into a single owner view. Here is the full build path, plus how Atlas and Titan Intelligence change what a dashboard means in 2026.

10 min read

Key takeaways

  • ServiceTitan custom dashboards are built from report modules, which are saved reports configured to display as a chart, table, or KPI tile
  • Dashboard sharing is managed under Settings > Operations > Dashboards, and each dashboard is scoped to a user, role, or location
  • Titan Intelligence is the analytics engine that benchmarks your shop against similar businesses and ships a quarterly comparative report
  • Atlas is the natural-language layer on top of ServiceTitan, where you type a question and it runs reports, finds jobs, and dispatches techs
  • A custom dashboard answers the questions you already knew to ask. Clint answers the ones you did not, including ones that span email and calls
Contents
  1. 01What a custom dashboard is in ServiceTitan
  2. 02Step-by-step: create a custom dashboard
  3. 03Step-by-step: build a report module
  4. 04Step-by-step: share dashboards across the team
  5. 05Atlas: the AI sidekick that runs the reports for you
  6. 06Titan Intelligence: the benchmark layer
  7. 07Common dashboard configurations that actually work
  8. 08The honest tradeoffs
  9. 09Where Clint fits next to your dashboards
  10. 10Real contractor stories
  11. 11Sources
  12. 12Frequently Asked Questions

ServiceTitan invested more than $50M in research and development on its AI products alone before the Atlas launch. That is the budget context behind every dashboard, every Titan Intelligence benchmark, and the new natural-language Atlas layer. ServiceTitan has the most sophisticated reporting and dashboard stack of any field-service platform on the market, and it is also the most complex to set up.

This guide walks the full custom dashboard build inside ServiceTitan, where Atlas fits, what Titan Intelligence actually does, and where each one runs out of room.

What a custom dashboard is in ServiceTitan

A custom dashboard in ServiceTitan is a single page made up of report modules. Each module is a saved report that has been configured to display as a chart, a number tile, or a table. You arrange the modules in a grid, set the date range and location filters at the dashboard level, and share the result with a user, a role, or a location.

The defining feature is that every module is alive. The chart on your dashboard is reading the same data the report builder reads, so when you click into a module you can drill down into the underlying rows. There is no separate data warehouse, no overnight refresh, and no exported snapshot. This is what ServiceTitan's reporting infrastructure was built to do.

The cost of that power is the build path. Dashboards do not exist by default for most KPIs. You build them.

Step-by-step: create a custom dashboard

Run this once for the dashboard shell, then again for each module you want on it.

  • From the navigation bar click Reports
  • Click Dashboards in the left sidebar
  • Click Create Dashboard at the top right
  • Name the dashboard (e.g. "Owner Morning View" or "Service Manager KPIs")
  • Set the default date range (most common is "Last 30 days" rolling)
  • Set the default location filter (single location, all locations, or a specific business unit)
  • Save the empty dashboard

You now have an empty container. The dashboard is useless until you add modules.

Text Clint: "what should be on my owner morning dashboard for HVAC ops"

Step-by-step: build a report module

Each module is a saved report. The build path is the same as the report builder, with one extra step at the end where you choose the display format.

  • From inside the dashboard, click Add Module
  • Choose Create New Module
  • Pick a report template (Job, Invoice, Lead, Membership, etc.)
  • Click columns to add them to the module
  • Add filters (date range, location, status, business unit)
  • Run the report to verify the data
  • Click Configure Display
  • Choose chart type: KPI tile, bar, line, pie, or data table
  • Set the X-axis, Y-axis, and aggregation (sum, count, average)
  • Save and the module appears on the dashboard

Module display options are tighter than full report exports. A KPI tile shows a single number with optional trend arrow. A bar or line chart works for time-series. A data table works for top-N lists like top 10 techs by revenue. Pie charts are technically supported and operationally useless.

Text Clint: "build me an owner KPI list for the next 30 days, what numbers should I check daily"

Step-by-step: share dashboards across the team

Sharing is managed separately from creation. To manage custom dashboards, go to the navigation bar and click Settings, then in the side panel go to Operations > Dashboards, and click the Custom Dashboards tab.

From there:

  • Find the dashboard in the list
  • Click the Share icon
  • Choose share scope: specific users, a role, or a location
  • Pick view-only or edit access
  • Save

Edit access lets the recipient add or remove modules. View-only locks the dashboard to its current shape. Most shops give the owner edit access on every dashboard, give managers edit on their own, and give techs view-only on the techs-leaderboard dashboard.

The same Settings > Operations > Dashboards page is where you rename, duplicate, or delete custom dashboards. Duplicating is the right move when you want a "manager view" that is structurally similar to the "owner view" but with different filters.

Atlas: the AI sidekick that runs the reports for you

Atlas is the AI sidekick layer that ServiceTitan launched in late 2025. Atlas can handle tasks, surface insights, and even take action so it feels like working with a trusted copilot who knows ServiceTitan inside and out. Contractors interact directly with Atlas in the ServiceTitan platform, typing or speaking in plain English where they can ask Atlas to run reports, find jobs, dispatch technicians, or guide them through workflows.

Where Atlas fits with dashboards:

  • You can ask Atlas to run a report you have not yet built into a dashboard module
  • Atlas can summarize a dashboard view in plain English ("revenue is up 12% week over week, driven by HVAC service")
  • Atlas can take action, including throttling marketing spend or triggering campaigns when demand is light
  • Atlas surfaces anomalies on your dashboard before you click in

The honest tradeoff: Atlas is excellent at the questions whose answer is fully inside ServiceTitan. Atlas does not see your Gmail, your Google Calendar, or your CallRail recordings. If your question requires correlating a missed call with an email reply with a job that was booked the next morning, Atlas cannot answer it because the email and call data never reach ServiceTitan.

For a fuller take on the boundary, see what Titan Intelligence cannot tell you.

Titan Intelligence: the benchmark layer

Titan Intelligence is the analytics engine inside ServiceTitan that analyzes aggregate data from ServiceTitan's end-to-end system tailored for trades and arms you with the insights you need to make decisions quickly. It includes a quarterly report which compares key metrics of your business to businesses like yours and gives you data-driven insights on where you can make improvements.

What you actually get:

  • Quarterly benchmark report (your numbers vs comparable shops)
  • Anomaly detection on your KPIs
  • Predictive scoring on which jobs convert to memberships
  • Campaign-spend recommendations rolled into Marketing Pro Ads
  • Tech-performance scoring with peer comparison

Titan Intelligence is a paid layer on top of the base ServiceTitan platform. The pricing is bundled into The Works package or sold as a separate add-on. For shops doing more than $5M in annual revenue across multiple locations, the benchmark data alone usually justifies the line item.

For shops under $3M, the benchmarks are interesting and the actionable layer is thinner. The same insights are often available from your CSR's gut feel and your own weekly numbers.

Text Clint: "average ticket by tech last 30 days vs last 90 days, which techs are slipping"

Common dashboard configurations that actually work

After watching hundreds of ServiceTitan implementations, four dashboard archetypes emerge:

Owner Morning View: revenue MTD, booked jobs today, leads aging more than 7 days, AR over 30 days, tech utilization. Five tiles. Read in 30 seconds before the morning meeting.

Service Manager Daily: open quotes, scheduled vs completed today, callbacks open, parts backorders, the day's first-trip-fix rate. Eight modules. Driven all morning.

Marketing View: cost per booked job by source, leads by source week-over-week, conversion rate by campaign, ad spend MTD. Six modules. Reviewed weekly with the marketer.

Tech Leaderboard: top 10 techs by revenue, average ticket per tech, callbacks per tech, membership conversion per tech. Four modules. Visible in the dispatch room.

Most shops over-build their first dashboard. Resist the urge. Five to eight modules max, and every module has to answer a question someone asks at least weekly.

For the metrics most owners ignore that are still worth surfacing, see contractor dashboard metrics owners ignore and the home service KPIs complete metrics playbook.

The honest tradeoffs

ServiceTitan's dashboard infrastructure is the best in field service. It is also where the most contractor frustration shows up. The G2 and Capterra reviews are consistent: powerful platform, steep learning curve, dense interface, expensive add-ons.

A common Reddit comment is that ServiceTitan is "almost like it's too big to where my people are scared to even start learning it, so I end up only getting the bare features from it." Dashboards are usually the unbuilt feature. Most $1M-$5M shops have one or two stock dashboards and never build a custom one because the office manager does not have the time, and the owner does not know which modules to ask for.

The escape hatches:

  • Hire a ServiceTitan consultant ($3K-$10K/month) to build dashboards on retainer
  • Use Atlas to run ad-hoc reports without building a dashboard module
  • Run reports manually each week instead of dashboards
  • Use Clint to ask the question over text and skip the dashboard build entirely

Text Clint: "build me a quick view of jobs over $5K booked this week and which techs sold them"

Where Clint fits next to your dashboards

Clint is the AI brain that connects to ServiceTitan, Gmail, Google Calendar, and CallRail. Owners text it questions and it answers from real data across all four systems.

Clint does not replace ServiceTitan dashboards. Dashboards are the right tool for the questions you already know to ask. Clint is the right tool for the questions you ask once.

Examples Clint handles that a dashboard cannot:

  • "Of the leads that came in last week, which ones got an email reply but no callback?" (needs Gmail)
  • "Show me jobs scheduled today where the customer left a voicemail this morning" (needs CallRail)
  • "Which calendar appointments tomorrow are with customers who have an unpaid invoice over 30 days?" (needs Google Calendar)

For the boundary between in-house ServiceTitan AI and standalone AI assistants, see ServiceTitan AI vs standalone AI for contractors.

Real contractor stories

ServiceTitan's Fall 2025 Release introduced Atlas as a flagship AI sidekick, and the rollout coverage in Roofing Contractor and Landscape Management gave a window into early adopter responses. Operators report Atlas handles 70% of report-pulling tasks they used to ask their office manager for, but the same operators note Atlas does not yet handle multi-system questions that span ServiceTitan plus email plus calls.

The Owned and Operated podcast covers ServiceTitan implementation regularly. John Wilson and Jack Carr have run multiple episodes on dashboard build-out, and the recurring theme is that owners need to commit to fewer dashboards, not more. Five high-quality dashboards beat 25 unbuilt ones.

Sources

Frequently Asked Questions

6 questions home service owners actually ask about this.

  • 01Are custom dashboards free in ServiceTitan?

    Custom dashboards are included in every plan that has the full Reports module, which is most plans. Atlas and Titan Intelligence are paid add-ons. The dashboard infrastructure itself does not cost extra.

  • 02Can I share a custom dashboard with someone outside my company?

    No. ServiceTitan custom dashboards are scoped to users inside your tenant. You can export a snapshot as PDF or PNG and share that, but the live dashboard is not shareable externally.

  • 03Does Atlas work on every ServiceTitan plan?

    Atlas is rolling out across plans through 2026. As of the Fall 2025 release it is available in The Works package and being added to additional tiers. Confirm with your CSM whether your plan includes it.

  • 04What is the difference between Atlas and Titan Intelligence?

    Titan Intelligence is the analytics and benchmarking engine. Atlas is the natural-language interface that uses Titan Intelligence under the hood. You ask Atlas in plain English, and it runs the report or surfaces the insight that Titan Intelligence computed.

  • 05Can I get my dashboards into Power BI or Looker Studio?

    Not directly. ServiceTitan dashboards are an in-platform view. To rebuild them in Power BI or Looker, you pull the underlying data through the ServiceTitan API into a warehouse, then build the visuals in your BI tool. The walkthrough in how to export reports from ServiceTitan covers the 10 paths in detail.

  • 06How long does it take to build a useful dashboard?

    For a single dashboard with 5-8 modules, expect 2-4 hours of build time if you already know which reports you want. If you are figuring out the metrics from scratch, double it. Most shops underestimate the spec phase, not the build phase.

See Clint in action

Clint is the pre-built AI for home service shops. Connect your CRM, email, and phone system in minutes and the agents run on your real data.