A custom Meridian build-out that finds the work, answers the phone, books the route, prices the job, tracks what each job actually earned, grades every tech and every rep — and puts a commission-only door team on the street to feed it. One system. Not seven subscriptions that don't talk.
Nobody in this trade loses a job to a competitor's better pitch. The job dies in a gap — an unanswered phone while you're mid-correction, a quote you meant to send, a customer whose coating came due and nobody called. Every leak below is a system that doesn't exist yet.
The reason this works is not any single module — it's that they share one price book, one calendar, one customer record. The voice agent quotes the same numbers the door rep quotes. The margin tracker costs the job the scheduler routed. Nothing is retyped, so nothing drifts.
Consumer detailing is won on the phone and at the door. Email is where you win the accounts — the fleets, dealers and property managers who hand you repeatable, route-dense, invoice-paid volume. This system is built for that target, and it is built to be boring and deliverable.
Recurring volume
Used-car dealers needing reconditioning, rental and car-share depots, body shops overflowing post-repair details, marinas and RV lots.
Route density
Trade fleets — plumbing, HVAC, electrical, delivery. One yard, twelve vans, one address, one invoice. The best economics in the trade.
High ticket
Property managers and HOAs booking resident amenity days, corporate offices buying on-site employee perk details, exotic and collector owners.
Reply rate is a vanity metric. The engine reports on the only chain that matters, per segment, so you can kill what doesn't pay and double what does.
| Measured | Why it's the one that counts |
|---|---|
| Deliverability | Landed in inbox vs. spam, per sending domain. If this slips, nothing downstream is real. |
| Positive reply rate | Split by segment, so "fleets" and "dealers" are judged separately rather than averaged into mush. |
| Booked jobs | Replies are not revenue. This is the first honest number in the chain. |
| Revenue per 1,000 sends | The number that tells you whether to buy more list or fix the copy. |
| Margin of won accounts | Fed back from the margin tracker. A channel that wins unprofitable work is a channel to turn off. |
Compliance is built in, not bolted on. Business addresses only, a real physical postal address in every footer, one-click unsubscribe honoured instantly, and a permanent suppression list that no future campaign can override.
The instinct is to buy a list of people who own expensive cars and email them. That doesn't work, and understanding why changes who you point the machine at. Every luxury vehicle in the market already sits under a business that has a reason to want a detailing partner — and unlike the owners, those businesses are findable, reachable, and route-dense.
One thing to confirm before the first send. Consumer cold email is treated very differently by jurisdiction — the US regime is opt-out, Canada's is opt-in and prohibits it outright. Worth settling which market applies, with someone qualified, in week one rather than week five.
Ranked by ticket against ease of reach. Each one is a business entity with a findable address and a commercial reason to answer.
| Target | Why they buy |
|---|---|
| Exotic & luxury dealers | Reconditioning, pre-delivery prep and delivery detailing — recurring volume against one invoice. |
| Marque specialists | The independent Porsche or Mercedes shop whose entire client list is your target, and who wants somewhere to refer detailing. |
| Collector storage | Car condos and climate storage — dozens of high-value vehicles, one address, one decision maker. |
| Exotic rental fleets | Turnover cleaning between rentals. Lower ticket, but the highest frequency in the trade. |
| Luxury body shops | Post-repair detail overflow they currently do badly in-house or turn away. |
| Clubs & concierges | Country clubs, gated communities and luxury-building concierges booking on-site amenity days. |
The list is small enough to do properly. A few hundred to a couple of thousand records in most markets — sourced from mapping data by category and radius, state business registrations and marque club sponsor pages. Small enough to verify every address and personalise every send for real.
The instinct is to sequence all the dealers, then all the storage facilities. Don't. A campaign batched by segment succeeds and hands you forty bookings scattered across the entire metro — which is the scattered day the margin tracker exists to prevent, manufactured on purpose.
Batch by postcode cluster instead. One cluster per week. The replies that come back land in the same part of the map inside the same window, so the same forty bookings arrive already dense.
This is the seam between the two systems. Outbound decides where the work is; routing decides whether it's profitable. Sending in waves is what makes those the same decision instead of two.
If you can't buy a list of who owns what, the alternative is to observe it — and a rep standing on a street is already looking directly at the cars. That observation is free, and it produces the one asset in this entire build that a competitor cannot purchase at any price: a map of which specific driveways in the market hold which specific vehicles.
Step one
Public property data picks the street
Assessor records are public and legal to use. Home value is a good enough proxy to choose which territories are worth a rep's day.
Step two
The rep records what's in the driveway
Make and model captured as a first-class field on the knock, whether or not it sold. A "no" that logs a Porsche is still a valuable knock.
Step three
The CRM remembers it forever
After one season you own a vehicle map of your market — for re-approach, for coating campaigns, and for deciding where the next rep walks.
Never ask for the appointment in a cold email. Stranger to calendar slot is too large a jump, and asking for it is most of what makes an email read as spam. Ask for something smaller — one vehicle, as a trial, on a day you're already working their area. The booking happens on the reply, either through route-aware slots or by handing the thread to the voice agent to close by phone.
This is the highest-leverage module in the entire build, because in mobile detailing the person best qualified to answer the phone is always the person least able to — the one currently holding a polisher. The agent picks up on the first ring at 9pm on a Sunday, quotes from your real price book, and books into your real calendar.
| Beat | What the agent actually does |
|---|---|
| Qualify | Vehicle year, make, model and size class. Current condition — pet hair, smoke, oxidation, swirls. Where the vehicle will be sitting. |
| Locate | Checks the address against your service radius and applies the correct travel band, or declines politely and logs the out-of-area demand so you can see where to expand. |
| Price | Quotes from the shared price book — size class times condition tier times package, plus add-ons. The same arithmetic your door reps use, so a customer never gets two different numbers. |
| Book | Offers slots weighted toward days you're already working that area. Route density is created at the moment of booking, not patched afterward. |
| Secure | Takes a deposit on the call. Texts confirmation with prep instructions — water and power access, clear the vehicle, where to park. |
| Escalate | Anything off-script — a fleet contract, a damage dispute, an unusual restoration — is handed to you with the call summary already written. |
Reactivation of customers who've gone quiet, coating-maintenance reminders that come due on a date the system already knows, no-show recovery, and review requests timed to the moment the customer sees the finished car.
Transcribed, summarised and attached to the customer record. The objections you lose on become visible as a pattern instead of a feeling, which is what lets you fix pricing or packaging with evidence.
The agent cannot invent a price, promise a date the calendar doesn't have, or approve a discount outside its band. Anything it isn't certain of becomes a callback task for you, not a guess.
Generic CRMs model a contact. Detailing needs to model a car — its paint condition history, what coating went on it and when, which tech touched it last, and the before-and-after photos that are simultaneously your best marketing and your only defence in a damage dispute.
Almost every detailer prices from a feeling about what a job costs. The margin tracker replaces the feeling with a per-job profit and loss that includes the costs nobody counts — the drive, the chemical burn, the card fee, the commission owed on that specific ticket.
Every tech gets a weekly scorecard built entirely from things the system already records — no manager opinion, no clipboard. It is visible to the tech, which is the entire point: people correct what they can see being measured.
| Measure | Reads on | Why it's on the card |
|---|---|---|
| Redo & complaint rate | Quality | The single most expensive failure in mobile — a redo costs a full drive plus a full job and earns nothing. |
| Hours on site vs. quoted | Speed | Consistently over means either the tech is slow or your pricing is wrong. The card tells you which. |
| Chemical cost per job | Waste | Product burn varies enormously between techs doing identical work. It's invisible until it's measured. |
| Photo compliance | Discipline | No before-and-after means no marketing asset and no damage-dispute defence. Non-negotiable. |
| Review capture rate | Growth | The tech is standing next to a delighted customer and a finished car. That is the only moment a review is easy. |
| Upsell attach rate | Revenue | Engine bay, headlight restoration, interior protection — offered at the vehicle, or not offered at all. |
| On-time arrival | Reliability | Late arrivals are the most common complaint in mobile service and the easiest to fix once it's counted. |
| Average job margin | Money | Pulled straight from the margin tracker. Revenue without margin is not performance. |
Grades tie to pay bands, and the ladder is published. A tech should be able to read their own card and know exactly what moves them up. That converts the scorecard from surveillance into a raise they can go and earn — which is the difference between a system people route around and one they use.
Door-to-door is unusually well suited to this trade. The product is visible in the driveway, the buyer is standing next to the exact vehicle being sold, and a rep who books four houses on one street has manufactured route density for free — the thing that quietly determines whether the day is profitable.
The rep app is not a separate product — it is the same core, on a phone, in a driveway.
| In the app | What it does |
|---|---|
| Territory | Geo-fenced assignment so two reps never work the same street, with burn tracking so streets aren't re-knocked too soon. |
| Disposition | Every door logged — not home, no, callback, sold. This is what makes coaching possible instead of guesswork. |
| Vehicle observed | Make and model in the driveway, captured whether or not it sold. This is what builds the market vehicle map — see Targeting. |
| Instant quote | Size class and condition tier in, price out — the same price book the voice agent reads from. |
| Book on the spot | Real calendar, real slots, weighted to the street the rep is standing on. |
| Capture | Photo of the vehicle, e-signed agreement, deposit taken at the door. |
| Ledger | Their commission, live, with pending and cleared separated. |
The same scorecard logic as the crew, on the metrics that predict a rep's survival. Most reps who quit do so in the first three weeks, and the leading indicator is almost always knocks per hour — not talent.
Two things to settle with a professional before the first rep knocks. First, worker classification — a genuinely commission-only field force generally has to be structured as independent contractors, and the tests for that are strict and vary by state; getting it wrong is an expensive correction. Second, local solicitation rules — many municipalities require a permit and honour do-not-knock registries. Both are cheap to handle up front and painful to retrofit, so the build sequence puts them in Phase 3 before recruiting, not after.
Outbound goes first, in week one. Not because it pays first — it doesn't — but because sending reputation is the one thing in this build that runs on a calendar you can't argue with. Domains have to age and mailboxes have to warm at a pace no amount of urgency will speed up. Start that clock on day one and it matures quietly in the background while everything else gets built; start it in week four and you've added a month to the whole programme for no reason.
Week 1 · Outbound live, foundation set
Day one: sending domains registered separately from your primary, SPF, DKIM and DMARC authenticated, mailboxes provisioned, warm-up started. In parallel, the price book — size classes, condition tiers, packages, add-ons, travel bands — and the real cost inputs behind them. First segment list built and verified, first sequences written, first sends going out at warm-up volume before the week is over.
Low volume by design. Warm-up is a ramp, not a switch, and the ramp is the reason this phase is first.
Weeks 1–3 · Capture
Deployed together, because they are one loop: the agent books into the calendar, the calendar creates the job, the job produces the margin record. The timing is deliberate — this loop comes online while the mailboxes are still warming, so by the time reply volume is real there is somewhere for it to land and something to cost the resulting jobs with.
Weeks 3–4 · Scale & paper
Domains are mature enough to push real volume, and the segments that replied in week one get widened while the ones that didn't get rewritten. Off the critical path, in parallel: contractor agreements, commission plan, territory map and municipal permits — all settled before anyone is recruited.
Weeks 4–7 · Field
Start with two or three reps, not ten. The first cohort is how you calibrate the comp plan and find out what your real close rate is. Scale the headcount only once a rep can hit the tier that makes the economics work.
Week 7 onward · Compounding
By now the margin data is thick enough to act on. Crew and rep scorecards go live, prices get corrected from evidence rather than nerve, and any channel producing unprofitable work gets cut. This is the phase where the system starts paying for itself.
Short list, but the build genuinely stalls without them.
Your customer list, your vehicle history, your pricing and your margin data. The system is built around your business, and the data in it is exportable in full at any point. No part of this build creates a dependency you can't walk away from.
This is a calculator, not a projection. Nothing below is a claim about what the system will produce — every figure is an input you control, and the defaults are deliberately conservative placeholders. Replace them with your actuals and see what the shape looks like.
— jobs / month
— jobs / month
— jobs / month
—
Read this as a sizing tool, not a forecast. Close rates in particular are the numbers most worth distrusting until your own first sixty days of real data replace them — which is exactly what Phase 2 exists to produce.
The first working session is the price book and the cost bands — a couple of hours with your real numbers. That single artefact unlocks the voice agent, the door team's quoting, and the margin tracker simultaneously, because all three read from it. Everything after that is sequencing.