Flagship case study
Absolute Mobile Car Detailing
A mobile-only detailer with no shopfront, no walk-ins and a booking process that lived entirely in an Instagram inbox. We built the demand engine and the agent that answers it.
- Industry
- Mobile car detailing
- Location
- Brampton, ON, Canada
- Year
- 2025
- Pillars
- Marketing, AI & Automation
Image · awaiting upload
Placeholder for: Absolute Mobile Car Detailing van parked on a Brampton driveway mid-detail
absolute-coverProblem
The problem
Absolute Mobile Car Detailing is mobile-only. There is no unit to walk into and no signage on a main road - the entire proposition is we come to you. That is a strong offer and a hard one to distribute, because the business has no passive discovery at all. Every customer has to be reached deliberately.
Everything funnelled into one Instagram inbox. Enquiries arrived as DMs, usually outside working hours, and every one of them asked a version of the same three questions: do you come to my area, what does it cost, and when can you do it. Answering those meant the owner had to be free at exactly the moment a stranger felt like booking a detail.
- No storefront, so zero organic foot traffic to convert
- Demand concentrated in one Instagram DM inbox
- Repetitive qualifying questions consuming owner hours
- Lead response time tied to the owner being awake and free
What we built
- Market research and strategy document
- Meta Ads campaign (Advantage+ placements)
- Meta Pixel integration
- Instagram DM booking AI agent (n8n)
What we built
Two halves of one system. A paid acquisition engine to create the enquiries, and an automation layer to catch them the moment they land so that the first half is not wasted.
A market research and strategy document defining the buyer, the service area and the angles worth testing before a rupee or dollar of budget moved.
A Meta Ads campaign running Advantage+ placements across Facebook and Instagram, targeted to the 25-54 age band - the range that owns the vehicles and books the service.
Meta Pixel integration so on-site behaviour and conversion events feed back into the campaign instead of the account optimising blind.
An Instagram DM booking AI agent built in n8n: the AI Agent node driving a Gemini model, with Supabase behind it as the lead and booking store.
The artefacts
Image · awaiting upload
Placeholder for: Instagram DM thread showing the AI agent qualifying a detailing enquiry and confirming a booking slot
absolute-dm-threadImage · awaiting upload
Placeholder for: Meta Ads Manager view of the Absolute campaign structure with Advantage+ placements
absolute-ads-managerVideo · awaiting upload
Placeholder for: Screen recording walking through the DM booking agent end to end
absolute-walkthroughProcess
- 01
Research before spend
We wrote the strategy document first - who the customer is, which parts of the service area are worth paying to reach, and what a mobile detailer has to say differently from a fixed-location shop. The campaign brief came out of that document, not out of the ad account.
- 02
Instrument, then launch
Meta Pixel went in before the campaign went live so that no early data was lost. Events were mapped to the actions that actually indicate intent, so the optimisation had something real to chase from day one.
- 03
Campaign build
Advantage+ placements across Facebook and Instagram, 25-54 targeting, structured so that creative angles from the research doc could be read against each other rather than blended into one number.
- 04
Automate the inbox
In n8n, an Instagram DM trigger feeds an AI Agent node running Gemini. The agent answers the predictable questions in the brand voice, collects vehicle, location and preferred time, and writes the structured record to Supabase. Anything outside its remit is escalated to a human rather than guessed at.
- 05
Hand over the keys
The workflow, the Supabase schema and the campaign structure are documented and owned by the client. No black box, no hostage data.
Results
Results
The acquisition and response sides of the business now run as one system: paid demand lands in a DM, the agent qualifies and captures it in seconds, and the booking arrives as a structured row rather than as a conversation the owner has to re-read.
Campaign performance figures for this account are still being finalised with the client, and lead-quality benchmarks were not measured before the engagement began. Rather than estimate them, they are marked below and will be filled in once the client signs off on the numbers.
- Instagram DM enquiries answered and captured automatically, 24 hours a day
- Every lead stored as structured data in Supabase instead of inbox history
- Meta Pixel and conversion tracking live before the first ad impression
Keep reading
More work
Image · awaiting upload
Placeholder for: RENzia perfume vending machine finished in black and gold
renzia-coverRENzia
Perfume vending machines · renzialux.com
107 leads at 11 rupees a lead.
Read the case study
Image · awaiting upload
Placeholder for: Hands carving a wooden chess knight at the Boss Handicrafts workshop in Amritsar
boss-coverBoss Handicrafts
Premium handmade chess sets · Amritsar, India
Five content pillars, thirty reel ideas, one series arc.
Read the case study
Image · awaiting upload
Placeholder for: Meta Ads Manager results view for the Gurasis Media self-run test campaign
gm-first-campaign-coverGurasis Media
Self-run proof of work · Amritsar, India
A 3,000 rupee test budget. Ten leads. Our own money first.
Read the case study
Next step
Tell me where it is leaking.
One conversation. We find the bottleneck - demand, response, page or content - and you leave knowing which one it is, whether or not you hire us.