The problem isn’t stock, it’s visibility

Most plumbing businesses don’t have an inventory problem in the traditional sense. They have parts in the warehouse, parts on trucks, parts in a technician’s garage, and parts that were bought for a job six months ago and never used.

The actual problem is that nobody can see the full picture quickly enough to make good decisions.

A technician gets to a no-hot-water call and needs a common gas valve. It isn’t on the truck. Dispatch calls the warehouse. The warehouse has the wrong model. The technician drives to a supply house, or returns the next day. The customer is frustrated, the job costs more to complete, and a second appointment blocks another revenue-producing slot.

That one event might only cost a few hundred dollars in direct time and materials. Repeated across five, 10, or 25 trucks, it becomes meaningful leakage. For a trades business doing $1 million to $25 million in annual revenue, we often see combined leakage from stockouts, emergency purchases, duplicate ordering, dead stock, and unnecessary return visits land somewhere in the $50,000 to $200,000 range.

The number isn’t only the cost of the missing part. It includes:

Technician time spent driving or waiting for supply
Emergency supply-house pricing
A second dispatch and fuel cost
Lost capacity for a new service call
Customer trust when a supposedly simple repair takes two visits
Admin time chasing what was used, ordered, returned, or misplaced

The best inventory system for plumbing service trucks gives you a live, usable answer to three questions:

What parts does each truck need for the work it actually performs?
What is on that truck right now?
What should be replenished before a missing part delays a job?

A spreadsheet can help at the start. A well-configured field service platform can help more. But the real lift comes when your inventory process is connected to job history, technician activity, purchasing data, and an AI agent that watches for the exceptions your team can’t realistically monitor by hand.

Start with the jobs your trucks actually run

Many truck-stock programs fail because they begin with a supplier catalogue. That approach creates a long list of parts that might be useful, rather than a disciplined list of parts that are repeatedly needed.

Start with the last 12 months of completed plumbing jobs. Break the data down by job type, geography, technician, equipment type, and part used. If your system doesn’t record part usage cleanly, use invoices, purchase orders, job notes, and supplier reports to build a first pass.

For most residential and light commercial plumbing businesses, the job categories tend to reveal clear patterns:

No-hot-water calls often require a narrow group of valves, igniters, thermocouples, fittings, and connectors
Drain work drives repeat demand for specific couplings, cleanout plugs, trap components, and repair materials
Toilet repairs create a recurring basket of fill valves, flappers, supply lines, bolts, seals, and wax rings
Leak calls require varied materials, but certain fittings and isolation components appear often enough to justify truck stock
Older housing areas can need a different range of pipe sizes and transition fittings than newer developments

That last point matters. A truck serving post-war homes with older copper and galvanised pipe shouldn’t necessarily have the same replenishment list as one working in new housing developments.

The objective isn’t to make every truck a mobile warehouse. It is to give each truck enough of the right inventory to complete a high percentage of first-visit repairs without turning shelves into a graveyard of slow-moving parts.

A practical approach is to define three categories:

Core stock is on every service truck. These are low-to-mid-cost parts with regular usage and a high cost of being unavailable.

Role-based stock depends on the technician or truck type. A drain specialist, water-heater technician, and general service plumber don’t need the same loadout.

Job-specific stock is picked for scheduled work based on the estimate, site notes, and equipment details.

This structure avoids the all-or-nothing choice between carrying everything and carrying too little.

Why min-max counts alone don’t solve it

A basic inventory system can set a minimum and maximum quantity for each SKU. When stock hits the minimum, someone reorders. That is useful, but it doesn’t account for the messiness of field service.

A standard reorder rule doesn’t know that your busiest season starts in two weeks. It doesn’t know a technician has been doing more water-heater work since another team member left. It doesn’t know a large commercial project used the same fittings that normally support service work. It certainly doesn’t know that 14 service calls are already booked for a part of town with older plumbing stock.

This is where AI-supported inventory management becomes practical.

An AI inventory agent can read completed job records, part usage, truck counts, open purchase orders, booked work, and supplier lead times. It then looks for patterns that are difficult for an owner or warehouse coordinator to spot while running dispatch.

It can flag things like:

A truck has used eight 3/4-inch ball valves in the last 21 days and now sits below its normal working level
A common part has been used faster than its 90-day average, likely because of seasonal demand or a local equipment issue
Three trucks hold more than six months of slow-moving stock while another truck repeatedly orders the same item
A part is technically in stock, but allocated to an upcoming job and shouldn’t be counted as available truck inventory
A scheduled run of water-heater jobs is likely to draw down a key part before the next delivery
A technician’s truck consistently needs a part that isn’t included in the current role-based loadout

The agent doesn’t replace a warehouse lead or operations manager. It does the daily checking that often falls through the gap between dispatch, purchasing, field work, and month-end accounting.

If you’re looking at where these workflows fit into a broader operating model, see Omni Ops. The aim is not another dashboard to inspect. The aim is a system that identifies the next action and routes it to the right person.

What an AI truck inventory agent looks like day to day

An effective inventory agent needs clear rules, connected data, and an owner who decides the commercial trade-offs. It shouldn’t be given free rein to buy whatever it wants.

Here is a practical end-to-end workflow.

At the end of each completed job, the field service system receives the parts used. This can come from a technician’s mobile app, barcode scan, job closeout checklist, or an office review process. The system deducts those parts from the truck’s assigned inventory.

Each morning, the AI agent checks four data sets:

Inventory by truck, warehouse, and committed job
Parts consumed over the previous seven, 30, and 90 days
Jobs already booked for the next one to two weeks
Purchase orders, supplier lead times, and substitution options

It compares stock levels with expected demand, not just a fixed minimum. If a truck needs replenishment, the agent creates a suggested pick list. The warehouse coordinator can approve it, make the transfer, and mark the list complete.

If a part is becoming a shortage risk across the business, the agent can prepare a purchase recommendation with the recent usage pattern, stock on hand, open demand, and supplier lead time. A purchasing manager or owner approves the order.

For dead stock, the agent identifies parts with low movement and meaningful value. It does not simply label every old item as waste. Some parts are held for warranty work, known seasonal work, or a planned installation. The review process should include those exceptions.

The useful output is short. For example:

Truck 14 is below its target stock of 1/2-inch isolation valves. Recent use is 11 units in 30 days versus a normal level of six. Four water-heater jobs are booked this week. Move 12 units from warehouse stock before 7:30 a.m.

That is far more useful than an end-of-month report showing total inventory valuation.

Reduce dead truck stock without creating more stockouts

Dead stock is not always obvious. A truck can look fully stocked while lacking the items needed to finish common jobs. Meanwhile, hundreds or thousands of dollars sit on shelves in specialist fittings, old product lines, duplicate materials, or parts bought against a job that changed scope.

The answer isn’t an aggressive purge. That can push technicians back into supply-house runs.

Instead, use an aging and usage review. Most businesses benefit from looking at truck inventory in bands such as:

Used in the last 30 days
Used in the last 31 to 90 days
Used in the last 91 to 180 days
Not used in more than 180 days

The correct action differs by category. A $4 fitting that is rarely used might stay because it saves a two-hour return visit when needed. A $250 component with no usage, no scheduled demand, and multiple units across trucks deserves attention.

An AI agent can prepare the list, but your operations lead should set the policy. They know which parts are insurance against a costly delay and which parts are simply leftovers.

Regular review also gives you a way to rebalance. If Truck 6 has five units of an item that has not moved for four months, while Truck 2 has been buying that item at the counter, a transfer may solve the problem without a new purchase.

This is where a connected inventory process earns its keep. The business spends less time arguing about anecdotal shortages and more time working from actual usage.

Inventory should connect to dispatch and customer response

Truck inventory doesn’t operate in isolation. It affects what your office can promise customers, how dispatch schedules work, and how quickly technicians can close jobs.

If a booked job requires a part that is already below threshold, dispatch should know before assigning the call. The system can either trigger replenishment, assign a better-stocked technician, or flag that the job needs a part check before confirmation.

This matters because missed calls and delayed jobs often feed each other. When a technician is tied up on an unplanned supply run, the schedule slips. Calls stack up. The owner gets dragged back into dispatch.

The 24/7 Dispatch Voice Agent helps protect the front end of that process. It answers calls, identifies emergency versus scheduled work, books directly into the dispatch tool, and sends a confirmation text. It means a customer calling about a burst pipe is not left at voicemail while the office is busy chasing a missing part.

The inventory agent can then use booked-job information to identify expected parts demand before the day starts. These are separate workflows, but together they reduce the gap between what was promised and what the field team can actually deliver.

For businesses with a steady flow of estimates, the Estimate Follow-Up Agent is another useful connection. It follows up on day 2, day 5, and day 14, based on job size and trade context. If follow-up increases accepted work, your inventory planning needs to account for that expected uplift. Otherwise, a healthier sales pipeline can expose weak truck stock processes.

Build the system in the right order

You don’t need perfect data to begin. You do need enough consistency that a recommendation has a reliable basis.

Start with these steps.

First, create a clean parts list. Consolidate duplicate SKUs and decide how you will name common parts. If three people call the same valve three different things, the reporting won’t work.

Second, define truck types and core stock lists. Keep the first version modest. Use completed jobs to justify each item.

Third, make parts capture part of job closeout. If technicians don’t record usage, truck counts will drift. Keep the process quick. A scan, preloaded parts list, or simple mobile selection is better than a paper form that gets completed on Friday from memory.

Fourth, set replenishment ownership. Someone needs to approve pick lists, receive stock, resolve discrepancies, and follow exceptions. AI can identify the work. It cannot count a box that was returned to the wrong shelf.

Fifth, review results every month. Look at first-visit completion, emergency purchases, return visits caused by missing parts, dead-stock value, and technician feedback. Adjust stock lists based on evidence.

For a broader view of how AI workflows can support field operations, read our guides for business owners. The practical question is always the same: which repeated work is slowing down a capable team, and what information is missing when decisions are made?

Find the leakage before buying more software

The biggest mistake is assuming a new inventory platform alone will fix the issue. Software can give you a better record of stock. It won’t decide your truck profiles, enforce job closeout discipline, connect booked jobs to expected demand, or resolve ownership between dispatch and the warehouse.

That is why we start with the operating workflow.

An Omni Audit maps the points where jobs, calls, estimates, stock, and follow-up break down. In 60 minutes, we identify the manual work consuming time, the data sources available, and the AI agents that can take on the repeatable parts. You leave with three outputs: a clear workflow map, a shortlist of agent opportunities, and a practical sequence for implementation. No deck and no vague transformation plan.

If you want to see how this applies across service operations, see Omni for trades businesses. If the issue is showing up as delayed jobs, supply-house runs, or a warehouse that can’t explain what is on each truck, Book a call with Sam.

Don’t ignore the calls you miss while fixing operations

Inventory control should free the office and field teams to serve more customers. It should not become another admin project that takes attention away from incoming work.

If after-hours calls, voicemail, and weak call handling are part of the same operational pressure, use the After-Hours Call Recovery Plan for Trades as a working checklist. You can also download the direct version here: After-Hours Call Recovery Plan.

The plan is useful because inventory and call recovery meet at capacity. There is little value booking every call if trucks arrive unprepared and create avoidable repeat visits. There is also little value improving truck stock if the business still loses urgent jobs because nobody answers the phone.

The Review and Reactivation Agent closes another part of that loop. It asks satisfied customers for a review the day after a completed job and reactivates previous customers at the right service interval. Better first-visit completion gives that agent a stronger customer outcome to work with.

The practical next step

A well-run truck inventory system does not require every part to be counted every day. It requires the business to know what moves, what is needed next, what is stranded, and who owns the action.

For a plumbing business, that can mean fewer counter runs, fewer delayed jobs, less cash tied up in dead stock, and a dispatch team that has more confidence in the schedule it is selling.

Start by reviewing the last 90 days of stockouts and supply-house purchases. Identify the 20 parts most often associated with return visits or urgent buying. Then compare that list with your truck loadouts and completed job data.

If you want help turning that review into an operating plan, the AI audit for trades businesses is designed for exactly this kind of problem. When you’re ready to map the workflow and find the leakage, Book a call with Sam.

Best Inventory System for Plumbing Service Trucks