Produced Water Logistics Software: How Do You Stop Paying to Truck Barrels Past a Pipeline You Already Own?

The barrel that went the wrong way

A tank battery in a mature unconventional field fills faster than the oil does, because in these basins you produce several barrels of water for every barrel of oil. On Tuesday the level sensor trips, the pumper calls the hauling coordinator, and a vacuum truck is dispatched. The driver loads, drives 34 miles to a commercial disposal well, waits 50 minutes in line, unloads, and writes a ticket. That ticket becomes a line on an invoice that arrives three weeks later with a wait time charge nobody can verify.

Nine miles from that battery is a transfer point on a gathering line the company already pays a monthly commitment on. It had capacity that afternoon. Nobody knew, because the pipeline volumes live in a SCADA screen in a different office, the disposal well capacity lives in a spreadsheet the midstream group maintains, and the hauling dispatch lives in a phone call. Three systems, three owners, one truck going the expensive way.

That is the entire business case. Water handling is one of the largest recurring operating costs per barrel in a mature position, and the spread between a piped barrel and a trucked barrel is wide enough that a single misrouted week costs more than a month of software. The reason it keeps happening is not stupidity. It is that no one screen holds the network, the capacity and the demand at the same time.

What Sourcewater and Enverus actually leave you doing

Sourcewater is genuinely useful for what it does: market intelligence on water, disposal capacity and infrastructure across a basin, including who has capacity and who is moving what. If your question is where should we contract disposal next quarter, it earns its subscription. It is an outside in view. It does not run your dispatch, it does not hold your permit limits per well with today's remaining volume, and it does not reconcile your hauler's invoice.

Enverus is the same category at a larger scale, with a data and analytics centre of gravity. Excellent for benchmarking, acreage and market context. Not an operating system for the barrels moving through your own network this afternoon.

What operators actually run on underneath those subscriptions is a hauling coordinator with a whiteboard, a midstream engineer with a spreadsheet of injection volumes, and a back office clerk keying scanned tickets into an accounting system. Some run a field ticketing app that captures the ticket but knows nothing about capacity. The gap in every case is the same: nothing joins the source, the network, the sink and the cost in one place, at the point where the routing decision gets made.

What a custom build has to include

The network model comes first and it is the part everyone underestimates. Every tank battery with its storage volume and typical fill rate. Every transfer point and interconnect with the direction it can flow and the party that owns it. Every gathering segment with a capacity. Every disposal well with three separate numbers that people constantly conflate: what the permit allows per day, what the well can physically take at current pressure, and what is already committed to someone else today. Recycling facilities and storage pits with current inventory and quality. Once that graph exists, routing becomes a solvable problem rather than an opinion.

Then the ticket, digital and captured by the driver. Origin, destination, volume, load and unload times, wait time with a timestamp rather than a claim, and a signature. The wait time detail alone changes the invoice conversation, because demurrage disputes are currently one person's memory against another person's ticket. Capture location automatically so a ticket cannot be written for a battery the truck never visited.

Then the routing view. Given predicted volumes by battery over the next 24 to 48 hours, given available pipeline capacity and disposal capacity, and given the cost per barrel of each path including any minimum volume commitment you have already paid for, what should move where. This does not need to be a black box optimiser to be worth building. Even a ranked recommendation that flags every truck dispatch where a piped route was available at lower cost pays for the project, because it makes the exception visible in the moment instead of at month end.

Then the money. Hauler rate structures with per barrel, per mile and wait time components applied to the captured ticket automatically, so the invoice is checked against your own record rather than approved on trust. Interconnect and third party disposal billing where you are the one charging. Allocation of water cost back to the producing wells, which is what makes the operations team care about the number.

Then telemetry, and here the value is prediction rather than monitoring. Tank level sensors and transfer meters feeding fill rate forecasts mean dispatch happens on a projection instead of a phone call after the alarm. That single change is what converts a reactive trucking operation into a planned one, and it is where most of the recovered cost lives.

The data problem: every hauler sends something different

Field ticket data arrives in whatever the hauler's system produces. One sends a CSV nightly. One has a portal you log into. Two still send scanned paper. The gathering system operator sends a monthly statement in PDF. Your own SCADA holds transfer meter volumes on a historian tag structure nobody has documented since the engineer who built it left.

Plan for an ingestion layer that treats every one of those as an untrusted source, normalises to your ticket model, and quarantines anything that fails validation instead of silently averaging it. Volume reconciliation between what a battery says it produced, what the ticket says was hauled and what the disposal well says was received should be an explicit daily report with a variance threshold. Water disappears in the gaps, and so does money.

Regulatory reporting sits on top of this rather than beside it. Injection volumes per well per period, in the format your state agency requires, produced from the same records that drive your billing. In Texas that means the Railroad Commission's injection reporting, and other states have their own equivalents with different periods and formats. Operators who assemble those reports separately end up with two sets of numbers, which is a problem you only discover during an inquiry.

Cost, timeline and what moves them

The first release runs $60,000 to $140,000 and ships in 10 to 16 weeks. That is the network model, driver ticketing with offline capture, the daily routing and exception view, and hauler invoice reconciliation. Offline capture is not optional, because half of your batteries have no usable signal and a ticket that cannot be written in the field will be written badly later.

The full build with recycling inventory and water quality, telemetry forecasting, interconnect billing, cost allocation to wells and state reporting runs $160,000 to $380,000 across 6 to 12 months. What pushes it up: operating in more than one basin with different regulatory regimes, being a midstream party billing third parties rather than only an operator, and integrating live SCADA rather than daily files. What holds it down: starting in one area of interest with your top haulers and your own disposal wells, then extending.

When you should not build this

A small operator with a few dozen wells, one commercial disposal contract and no owned pipe does not need this. The hauler's tickets plus a spreadsheet are proportionate, and the routing decision has only one answer anyway. The build case appears when you have genuine choice in the network, which means owned or committed pipe alongside trucking, several disposal outlets, or recycling that competes with disposal for the same barrel.

How to choose a developer

Ask how they will model a disposal well. If the answer is a capacity field, keep looking. Permitted daily volume, current injection pressure headroom and committed volume are three different constraints and the routing is wrong if any of them is missing.

Ask how the driver app behaves with no signal for six hours. The right answer involves local storage, a sync queue and conflict handling on the ticket, and a developer who has built field applications will bring it up before you do.

Ask what they have integrated on the telemetry side. Tank level sensors from a specific vendor, a specific SCADA host, a specific historian. Water measurement is messier than oil measurement and the calibration and unit traps are real. Ask for the site and the tag, not a general claim.

Ask who owns the code, the repository and the cloud accounts, and settle it in writing before kickoff. At Digital Heroes the client owns all three from the first commit. Your network model with its permit limits and commercial terms is a description of how your water business actually works, and it should not live somewhere you cannot reach.