Do You Actually Have That Key? Stock Checks, Key Blanks and Honest Phone Quoting (2026)

Do You Actually Have That Key? Stock Checks, Key Blanks and Honest Phone Quoting (2026)

There is a failure mode every automotive locksmith shop knows and almost nobody writes about. A caller describes a vehicle, whoever is holding the phone gives a confident price and a same-day window, the tech drives thirty minutes, opens the drawer, and the correct blank is not there. Or it is there, but it is the wrong chip family. Or the fob shell matches and the FCC ID does not. Now there is a customer standing in a parking lot, a tech with nothing to sell, and a shop that has to choose between eating the trip or charging for a visit that produced no key.

As of August 2026, this is still the most expensive quiet problem in the trade, and it is not a discipline problem. It is a data problem. Nobody forgot to check inventory. The shop genuinely could not check inventory in the eleven seconds available on a live call, because automotive key inventory does not behave like inventory in any other business. This guide covers why that is true, what the phone person actually needs to know before a price leaves their mouth, and the four honest ways to end that part of the conversation.

A key blank is not one SKU

Everything downstream of this misunderstanding goes wrong, so it is worth being precise about it.

When a parts catalog for almost any other trade says a part is in stock, that means one thing exists and it fits. An automotive key does not work that way. What a customer calls "a key for my truck" is really the intersection of at least five independent attributes, and getting four of them right still produces a key that does not work.

The blade profile. The mechanical cut, the keyway, and whether it is edge cut, high security laser cut, or a sidewinder. This is the attribute most shops track well, because it is physical and visible.

The transponder chip family. The immobilizer chip inside the head. Fixed code, rolling code, encrypted, and within encrypted there are generations that share nothing but a shape. Two keys that look identical on the bench can carry chips that are not remotely interchangeable, and a chip your programmer cannot write is functionally not stock.

The fob shell and button count. Three-button, four-button, with or without remote start, with or without a power liftgate. Same vehicle, same year, different trim, different shell. The customer will not know which one they have and will describe it as "the normal one."

The FCC ID. The regulatory identifier printed on the back of the remote, which pins the exact remote variant. Two remotes for the same model year with different FCC IDs are different parts. This is the single most reliable identifier a customer can read to you, and almost nobody asks for it.

The operating frequency. 315 MHz versus 433 MHz being the common split in the North American market. A remote at the wrong frequency programs to nothing. Radio equipment authorization and identifier conventions sit under the Federal Communications Commission, and the FCC ID on the back of the fob exists precisely so a specific device variant can be identified unambiguously. That is a gift to a locksmith and almost nobody uses it.

Multiply those out and a single popular model year can have six to ten distinct valid parts. That is before the part that really hurts.

One model year can straddle two systems

Manufacturers do not change key systems on January 1. They change them at a production break, which means a model year is frequently split. Vehicles built before the break use one system, vehicles built after it use another, and both are sold as the same model year with the same badge.

This is why a shop that quotes purely off year, make, and model will be wrong on a predictable share of jobs. The customer says a model year that sits on a split, the phone person prices the cheaper system because that is what the catalog row shows first, and the tech arrives to find the later system, which needs a different blank, a different chip, and often a longer programming procedure at a higher price. Now the shop is choosing between eating the difference or raising the price on arrival, and raising the price on arrival is the single fastest way to convert a booked job into a bad review.

The same problem appears in aftermarket form. A vehicle that has had an aftermarket remote start, a replacement immobilizer module, a salvage-title cluster swap, or a prior all-keys-lost job done by another shop may not match its own factory documentation at all. No catalog can know that. Only a tech standing at the vehicle can.

The practical consequence is that year, make, and model gets you to a price band and a probable part. It does not get you to certainty, and any process that pretends otherwise is manufacturing wasted drive-outs. The mechanics of how a structured year-make-model lookup actually behaves, and where it stops being reliable, are covered in how AI quotes car key replacement by year, make and model.

The three inventory locations, and why they always disagree

Ask a shop owner where a specific fob is and you will get a pause. That pause is real, and it comes from the fact that automotive key stock lives in three places that update on completely different cycles.

The shop shelf. Counted occasionally, usually accurate, and irrelevant to a mobile job unless someone is going back for it. Shelf stock is the number most shops mean when they say "we have it."

Each truck. This is where the job actually happens, and it is the least visible location in the business. A tech pulls a key off the shelf on Tuesday and uses it on Thursday. Nobody wrote either event down. Two techs can both believe they have the last one. A truck is a warehouse with no receiving dock, no cycle count, and a driver who is busy.

On order. Parts ordered against a specific job, parts ordered to restock, and parts that were ordered three weeks ago and never arrived. This is the location that generates the worst phone answers, because "we have it coming" gets spoken as "we have it."

These three disagree constantly, and they disagree in a specific direction: the shop believes it has more usable stock than it does. Shelf counts include keys that are already promised. Truck counts are stale. On-order is treated as present. The error always runs toward optimism, which is exactly the direction that produces a confident quote and an empty drawer.

You do not fix that with better intentions. You fix it with two rules. First, stock is consumed at the moment it is committed to a job, not when it is physically installed, so a key promised on a Tuesday call is gone from availability on Tuesday. Second, truck stock is a location with an owner, and it gets counted on a schedule, weekly at minimum, because a location nobody counts is a location nobody can quote from.

What "in stock" has to mean before you promise a time

Here is the operational definition worth adopting, and it is stricter than what most shops use.

A key is in stock for quoting purposes only when all of the following are true at once: the exact part is physically present in a known location, that location is either the truck assigned to the job or reachable before the promised window, the part is not already committed to another open job, and the shop's programming equipment can actually write that chip family. That last condition catches a real category of failure. Having the blank and not having the capability is the same as not having the blank, and it is worse because it feels like a stock problem when it is a tooling problem.

Anything short of all four is not stock. It is a probability. That distinction is the whole article, because a probability quoted as a certainty is how you end up on the phone with an angry customer at 4 PM.

The four honest phone outcomes

Every automotive call resolves into one of four answers on the parts question. Shops that name these four and give the phone person exact language for each stop guessing. Shops that do not, guess, and guessing skews optimistic every time.

Outcome	When it applies	What the caller must hear
In stock and confirmed	Exact part present, uncommitted, on or reachable by the assigned truck, and programmable with equipment on hand	A firm all-in price and a concrete arrival window, booked on the call
In stock, pending on-site verification	Catalog shows the part and it is on the truck, but the model year straddles a system split, the trim is ambiguous, or the vehicle may have prior aftermarket work	A price that is firm for the identified key type, said plainly as contingent on the tech confirming the key type at the vehicle, with the alternative price named out loud before the truck moves
Must be ordered	Part is not on hand or not on any truck, and has to come in	A real date, the deposit terms if the part is vehicle-specific, and no same-day language of any kind
Not something we service	Vehicle, system, or job type outside the shop's tooling or skill set	Told immediately on the call, with a referral if you have one, and no appointment created

Four rows. Most shops use two of them, and the second one they use is a soft version of the third.

In stock and confirmed is the easy one, and it is the only one that earns a firm same-day promise. Say the all-in number, including the trip charge and any after-hours surcharge, and book it.

In stock pending verification is the row that gets skipped, and skipping it is what creates the mid-job price increase. The correct handling is not to hide the uncertainty. It is to price the identified key type firmly and then name the fork out loud: if the tech finds the other system at the vehicle, the price is the second number, and the caller can decide then with no trip charge if they walk away. That sentence costs you a small number of bookings and saves you every argument you would otherwise have in a driveway. A customer told about a fork in advance treats the higher price as a known possibility. The same customer surprised by it treats it as a bait and switch, and is right to.

Must be ordered requires a real date, not a hopeful one. "A few days" is not a date. The failure here is subtle: shops soften an order into a same-day maybe because they are afraid of losing the caller to a competitor who will say yes faster. What actually happens is the caller books, waits, gets a call moving the appointment, and cancels with a worse opinion of you than if you had been straight in the first place. If the part is vehicle-specific and non-returnable, that is exactly the situation a deposit exists for, collected on the call through a texted payment link rather than a promise, which is covered in the SMS communication guide.

The fourth answer is the most valuable sentence on the call

"That is not something we can do" is the highest-value thing a locksmith shop can say on the phone, and it is the sentence shops avoid hardest.

The avoidance is understandable. It feels like turning down money. The phone person does not want to be the one who lost a job, especially if they are unsure and the owner might have said yes. So the call ends with something soft: we can send someone out to take a look, we will have a tech evaluate it, let us get you on the schedule and we will see. All three of those are the same sentence, which is: we will drive to you and find out.

Price that out honestly. A drive-out that produces no job costs fuel, an hour or more of a tech's time at a real loaded rate, the trip charge you will almost certainly refund because you could not perform the work, and the job you declined or delayed in that slot. Then add the part that does not show up in any ledger. The customer who watched a tech arrive, look at the vehicle, and leave now believes you do not know your own capabilities, and that belief goes in a review and into what they tell the next person who asks for a locksmith.

Compare that to the cost of a clean no on the phone. You lose a job you were never going to complete, in twenty seconds, at zero cost, and you frequently gain something: a caller who got a straight answer and a referral, remembers it, and calls you first for the job you can do. Straight answers are cheap and they compound.

The prerequisite for saying it is knowing where the line is. That means the shop has to have actually written down what it does not service: vehicle systems outside its tooling, job types it does not perform, security products it does not stock, and the geographic edge past which the drive does not pay. Most shops have never written that list, which is why the phone person cannot use it. Writing it takes an hour and it is the highest-return hour available.

What a bad stock answer actually costs

It helps to hold the whole cost in view, because shops tend to count only the fuel.

The wasted drive-out is the visible piece: fuel, vehicle wear, and typically sixty to ninety minutes of a tech's day, including the drive back. Then the refunded trip charge, because you cannot in good conscience bill a customer for a visit where you produced nothing, and if you do bill it you have bought a chargeback and a review.

Then the opportunity cost, which is usually the largest line. That truck was unavailable during the window. Whatever call came in during that time went to a competitor or waited. On a busy Saturday, a single dead drive-out can cost you two jobs, not one.

Then the reputation cost, which is unbounded. A customer who was quoted a price and a time and got neither writes a review that specifically says you quoted a price you did not honor. That review is going to sit on your profile for years, in front of every emergency caller who is choosing between three shops in ninety seconds.

Then the last one, which nobody counts: that customer now calls someone else first. Forever. You did not lose a job. You lost a lifetime of jobs and every referral attached to it.

Set against that, a slightly slower phone call with a truthful contingency in it is not a cost. It is the cheapest insurance in the business.

How a price-book-driven AI receptionist handles this

Nothing above is solved by an AI answering the phone unless the AI is wired to the shop's actual catalog and the shop's actual boundaries. Quoting from an invented average is worse than a human guessing, because it is confident and it is wrong at scale. What makes it work is that the price book is the shop's own, and the rules around it are the shop's own.

Three things change when the phone is answered by a system that reads from a real catalog.

The quote is tied to what the shop actually services. If the vehicle and key type have no row in the catalog, the correct behavior is not to improvise a number. It is to say the shop needs a tech to confirm, or that this is outside what the shop handles, and to route accordingly. The catalog stops being a pricing reference and starts being a boundary. How to structure that price book so it can carry that weight is covered in the locksmith price book guide, and the quoting mechanics are in automated quoting.

Ambiguity gets flagged instead of buried. A model year that straddles a system split, a trim that changes the fob, or a caller who cannot describe the key are all conditions the shop can define in advance. When one of them fires, the caller gets the second outcome from the table above, spoken plainly, with the fork named. The system does not need to be smart about the vehicle. It needs to be honest about the uncertainty the shop already knows exists.

The expectation is set on the call, not in the driveway. Whatever contingency was named on the phone travels with the job to the tech and to the customer's confirmation text, so the tech arrives at a customer who already knows there is a verification step. That is a completely different conversation than a surprise. The dispatch and scheduling side of holding that together is in scheduling and dispatch software evaluation, and the automotive-specific tooling picture is in automotive locksmith software.

The one thing worth being blunt about: none of this makes the parts problem go away. If truck stock is not counted, no phone system on earth can tell a caller whether the key is on the van. The AI enforces the rule consistently. It cannot invent the data. A shop that wants accurate stock answers on the phone has to do the unglamorous part first, which is treating each truck as a real inventory location with a real count.

A short implementation order

If you are starting from nothing, do it in this sequence, because each step makes the next one possible.

Write the do-not-service list first. One page. Vehicle systems, job types, and the geographic boundary. This is the only step that pays off the same day, because the phone person can use it immediately.

Then count the trucks and keep counting them weekly. Assign each truck an owner. Until truck stock is a known number, every stock answer is a guess.

Then mark stock as consumed at commitment, not at installation. This is a one-line policy change that removes an entire class of double-promise.

Then define your ambiguity rules: which model years, makes, and key types require on-site verification. Every experienced tech in your shop already knows these. Get them out of their heads and onto paper.

Then, and only then, wire the phone to it through AI call handling. A system reading a catalog that reflects reality gives honest answers all day. A system reading a catalog nobody maintains gives confident wrong answers all day, faster than a human could.

The bottom line

Automotive key inventory is hard for reasons that have nothing to do with organization. A key is five attributes that must all match, a single model year can carry two incompatible systems, and the stock lives in three locations that update on different clocks and always err toward optimism. The result is a phone quote that sounds certain and a drawer that is empty, and that gap costs a shop far more than fuel: a refunded trip charge, an unavailable truck, a review that names the price you did not honor, and a customer who now dials somebody else first.

The fix is not more confidence. It is a stricter definition of in stock, a written boundary for what the shop does not do, truck counts that actually happen, and four honest phone outcomes with the language written down in advance, including the one that ends the call in twenty seconds with a clean no. An AI receptionist working from the shop's real price book can apply those four consistently on every call, flag the ambiguous ones for tech confirmation instead of over-promising, and put the contingency in front of the customer before the truck moves. It cannot count your vans for you. Do that part yourself, and the phone stops writing checks the drawer cannot cash.

Frequently asked questions

Why can a locksmith not confirm a car key is in stock over the phone?

Because a key blank is five attributes that must all match, not one part number: the blade profile, the transponder chip family, the fob shell and button count, the FCC ID, and the operating frequency. A single model year can also straddle a production break where the immobilizer system changes mid-year, so the badge on the car does not determine the part. The honest answer on many vehicles is that the catalog identifies a probable key and a tech confirms it at the vehicle.

What does in stock actually mean for an automotive locksmith job?

In stock means four things are simultaneously true: the exact part is physically present in a known location, that location is the assigned truck or reachable before the promised window, the part is not already committed to another open job, and the shop can program that chip family with the equipment on hand. Anything less than all four is a probability, not stock. Quoting a probability as a certainty is what produces wasted drive-outs and refunded trip charges.

How should a shop answer when it cannot do the job at all?

Say so immediately on the call, in plain words, and offer a referral if you have one. Shops avoid this sentence because it feels like turning down money, so instead they send a tech to look, which costs fuel, an hour of tech time, a refunded trip charge, the job that truck could not take, and a customer who now doubts you know your own capabilities. A clean no costs twenty seconds and frequently earns a caller who remembers the straight answer.

Why do locksmith trucks and the shop shelf never agree on inventory?

Because a truck is an inventory location with no receiving process and a driver who is busy, so parts leave the shelf and get consumed with nothing written down. Shelf counts also include keys already promised to open jobs, and parts that are on order get spoken about as if they are on hand. All three errors run in the same direction, toward believing you have more usable stock than you do.

How does an AI receptionist avoid over-promising on parts availability?

It quotes only from the shop's own price book and treats a missing catalog row as a boundary rather than an invitation to improvise. When the shop has flagged a vehicle, model year, or key type as ambiguous, the caller is told the price is firm for the identified key type and contingent on the tech confirming it at the vehicle, with the alternative price named out loud before the truck moves. It applies the rule identically on every call, but it cannot invent stock data the shop does not maintain.

How much does TheKeyBot cost for a locksmith shop?

TheKeyBot's Core plan is $500 per month with 500 AI minutes and 45 cents per minute overage, Pro is $750 per month with 1,000 minutes at 40 cents overage, and Elite is $1,200 per month with 2,500 minutes at 35 cents overage. Every plan includes bilingual 24/7 answering, quoting from your own price book, calendar booking, GPS-aware dispatch, payment links, and automated review requests, with no per-seat fees. Plan details are at https://www.thekeybot.com/pricing.

Sources

1. Federal Communications Commission - equipment authorization and FCC identifier conventions for radio devices including vehicle remotes: https://www.fcc.gov/
2. Associated Locksmiths of America - professional standards, training, and industry practice guidance for locksmith businesses: https://www.aloa.org/
3. Federal Trade Commission - guidance on truthful pricing disclosures and advertising practices for service businesses: https://www.ftc.gov/