CRM and Pipeline Automation for Body Shops

A Repair Order Is a State Machine, Not a Deal

The stages that ship with every general-purpose CRM describe a sale: new lead, contacted, quoted, won, lost. A body shop's real object is not a deal that closes; it is a vehicle the shop takes physical custody of for two to five weeks, moves through a fixed sequence of physical states, and hands back. The correct collision repair pipeline is a state machine with stages that correspond to where the car actually is and who is currently holding it up. Estimate written. Scheduled for drop-off. Vehicle in, keys taken. Teardown. Supplement submitted. Waiting on insurer. Approved, parts ordered. Parts received. Body. Paint. Reassembly. ADAS calibration. Quality control. Ready for delivery. Delivered. That is fifteen states, and the difference between a shop that knows which of them each of its thirty cars is in and a shop that does not is the difference between a manager who can answer a phone call in four seconds and one who walks out to the shop floor to look. "Won" is the moment the work begins, not the moment it ends, and any pipeline that treats the signed authorization as the finish line is measuring the least interesting thing that happens all month.

Two Contacts, One Vehicle, Different Obligations

The record has to carry two counterparties. The customer is one: name, mobile, vehicle, deductible, rental arrangement, preferred contact method, and the running thread of what they have been told. The insurer is the other, and it is not the same insurer relationship twice — a claim has a carrier, a claim number, a named adjuster with a phone number that goes to voicemail, an assigned labor rate, a parts-usage policy, and a DRP relationship or the absence of one. Those are not custom fields bolted onto a customer record. They are a second party with its own obligations and its own clock. When the supplement goes out, the object that is now blocked is the repair, and the party responsible is the insurer, not the customer — so the follow-up cadence, the escalation path and the reporting attribution all belong to the carrier record. A collision repair pipeline that cannot say "this vehicle has been stalled for six days and the reason is a named adjuster at a named carrier" is a pipeline that will let the shop absorb the blame for it.

The Parts Sub-Pipeline Runs Underneath the Repair

A repair order does not wait on one part; it waits on the slowest one. That makes parts a pipeline inside a pipeline, and it needs its own stages: quoted, ordered, confirmed with ETA, shipped, received, inspected, mis-shipped, damaged in transit, returned, back-ordered indefinitely. Each line carries whether it is OEM or aftermarket, because the insurer's parts policy determines which one was approved, and a shop that installs an OEM fender when an aftermarket one was authorized will eat the difference. The stage that matters most is back-order, because it is the one the shop cannot fix and the one that most reliably destroys a delivery date. The automation that earns its keep here is dumb and reliable: the day a vendor changes an ETA, the repair order's projected delivery recalculates, the customer gets a revised date with a plain reason, and the rental note goes with it. What the shop must never do is hold a slipped date privately in the hope it recovers. It does not recover, and the customer finds out on the day they had arranged a ride home.

The Total Loss Branch Is a Different Business Entirely

Some percentage of the cars that come through the door will never be repaired. When the estimate plus the anticipated supplement approaches the total loss threshold — commonly 70% to 80% of actual cash value, depending on the state — the insurer is going to write the vehicle off, and everything about the shop's relationship with that customer inverts. There is no repair to schedule, no parts to order, no cycle time to manage. There is a tow bill, a storage bill, an authorization to release the vehicle to the carrier's salvage vendor, and a customer who has just learned their car is gone and who is now, whether they realize it or not, in the market for a different one. Most shops handle this badly because their pipeline has no branch for it: the record sits in "waiting on insurer" until someone notices. A pipeline that recognizes the total loss threshold as a real fork does three things instead — it stops the repair sequence cleanly, it triggers the storage and release paperwork, and it moves the customer to a short, genuinely useful sequence about what happens next with their claim. That last one is not a sales sequence. It is the reason that customer recommends the shop to the next person on their street even though the shop never turned a wrench on their car.

DRP and Non-DRP Cars Do Not Belong in the Same Lane

A direct repair program car arrives with a lot of the negotiation pre-settled — an agreed labor rate, an agreed parts policy, often a pre-authorized supplement ceiling and an electronic pathway to the adjuster. A non-DRP car arrives with none of it, and every one of those things is a phone call. The two require different documentation, different approval choreography, and different expectations set with the customer about how long the insurer side will take. Running them through identical stages guarantees that the shop under-communicates on the harder one, because the DRP process is the one the staff does forty times a week and the muscle memory wins. Splitting the pipeline by relationship type — and letting the DRP cars run a lighter, faster approval path while the non-DRP cars get a heavier chase cadence — is the single change that most reliably shortens the stall between teardown and approval on the jobs that stall the longest. Our reporting automation then measures the two separately, which is how a shop discovers what its non-DRP work actually costs it in days.

What Belongs in the Record Forever

The file has to survive the repair. Pre-scan and post-scan reports, the calibration documentation, the photos taken at drop-off before anyone touched the car, the photos taken at teardown that justified the supplement, the parts invoices showing OEM or aftermarket, the signed authorization, the final invoice, and the record of the deductible collected. Some of that is a warranty asset — a refinish complaint two years later on a $3,800 repair is defensible or it is not, and the difference is the paint code, the panel, the products and the date. Some of it is a liability asset, and the ADAS documentation is the clearest example: if a vehicle is later involved in a collision and the question of whether the forward-collision camera was properly calibrated is raised, the shop's answer is a document or it is nothing. A CRM that archives a customer as cold after eighteen months of inactivity is behaving exactly wrong here. The car has a decade left in it, and so does the file.

The parallel worth drawing is with an insurance agency's automation, which sits on the other side of the same claim and is optimized for a policyholder relationship measured in renewal years. A shop sees the same carrier as an approval gate on a vehicle it is holding right now. The two organizations are looking at one claim number and seeing completely different clocks. For the sequences that run while the customer is in the rental, see our multi-channel messaging automation.