AI in Automotive Aftersales: A Complete Guide (2026)

The first cold Saturday in October

A two-branch tire and fast-fit center opens at 8 AM. By 10 AM there have been forty calls. Half of them ask the same question: can I get my winter set fitted before the weekend. In Germany the "O bis O" rule, October to Easter, sends most of the country to the workshop in the same four weeks, and ADAC and the tire trade association BRV tell drivers every autumn to book early because workshops are full. Every call the counter cannot take is a set of tires fitted somewhere else.

Aftersales is full of moments like this: predictable peaks, thin staffing and a phone that decides who gets the work. This guide covers where AI is used in 2026, what the numbers say and how to start.

The market, in numbers

• The US light-duty aftermarket reached about $435 billion in 2025 and is forecast by the Auto Care Association and MEMA to grow 5.2 percent in 2026 and pass $500 billion by 2029.
• In the EU, the parts market is around 118 billion euros, of which the independent aftermarket supplies about 73 billion, according to the FIGIEFA and Roland Berger Panorama, which counts around 250,000 independent workshops and outlets.
• Cars are getting older. ACEA puts the EU fleet at 256 million cars with an average age of 12.7 years. S&P Global Mobility puts the US average at 12.8 years.
• Germany's ZDK counts 84,000 companies and 705,000 employees in the motor trade. Dutch workshops did a record 8.19 million periodic inspections on passenger cars in 2025, according to the RDW.

Older cars need more work. The constraint is not demand. It is people.

The people constraint

The TechForce Foundation's 2026 report estimates that US employers need about 241,842 new technicians a year through 2029 and that schools produce 101,743. For automotive alone it is 70,000 openings against roughly 42,000 graduates. The ZDK has reported a shortfall of around 10,000 skilled workers in the German trade. UWV counted 7,700 open vacancies in the Dutch car sector in the second quarter of 2026, with mechanics, body repairers, painters and receptionists all scarce.

That is why AI in aftersales is mostly about time: taking work off the counter and the shop floor that does not need a trained person. Diagnosis, warranty judgment and angry customers stay with people.

Where AI is being used

1. Customer contact

Voice agents answer the phone, understand the request and complete it: bookings, reschedules, opening hours, car-ready status, tire storage questions. The same rules run on web chat, email, SMS and WhatsApp. The phone is still the main channel: 64 percent of service customers in a 2025 CDK Global survey booked by phone, 19 percent online.

2. Scheduling and capacity

Booking the right job type with the right duration and technician, filling canceled slots from a waitlist, calling no-shows within hours and rebooking them, and finding every car due for a service, an inspection or a seasonal tire change and calling until it is booked. Vendor benchmarks such as AutoAlert's put dealership service no-shows at around 20 percent.

3. Status and proactive communication

Outbound car-ready calls and texts, parts-arrived updates and repair status for body shops. In Cox Automotive's 2025 Service Industry Study, 45 percent of dealership service customers had a frustration, most often that service took longer than expected. Most status calls exist because nobody told the customer.

4. Quotes, approvals and follow-up

Chasing open quotes, answering the questions and turning approvals into bookings. Cox reports that customers are 49 percent more likely to approve recommended work when they see photos or video.

5. Claims and estimates

For body shops, checking claims for missing photos, parts and approvals before they reach the insurer. Enterprise Mobility measured the average collision rental at 16.3 days in the first quarter of 2026, with a scheduling backlog of 1.8 weeks. CCC's Crash Course 2026 found that 28.3 percent of repairable estimates now include calibrations.

6. Asking questions of your own data

Plain-language questions over the planner and the call log: no-shows per location, unanswered calls last week, cars waiting on parts. Cox Automotive found 16 percent of consumers used AI tools during their service journey in 2026. Workshops can do the same with their own numbers. Ask Nekaf, for example, is free for 90 days and produces a baseline of six access metrics before any agent is switched on.

How to start

1. Measure first. Answer rate, speed of answer, abandonment, booking accuracy, calls per advisor, after-hours volume and bay utilization.
2. Pick one job. Usually the service desk, because that is where the leak is.
3. Write the rules down. Job durations, technician skills, loaner cars, breakdown handling. The workshop manager approves them.
4. Run in shadow mode. A week on real calls without acting, scored, before anything goes live.
5. Keep the number and the systems. Forward, do not migrate. Connect the planner when ready.
6. Review weekly. Sampled calls, hand-off quality, corrections tested before they go live. Add the next agent only when the first one reports results you can check.

Governance

Every call transcribed, every decision logged with the rule version that produced it, data held in the EU or the US as required, and a supervising review of sampled conversations. In Europe the EU AI Act and GDPR require it. It is good practice everywhere.