Blog · Published · Updated
AI in automotive: service request intake
Short answer
AI in a car workshop orders a client's request before it reaches the service adviser: the description of the symptom, the conditions the client gives and the information still to collect. It does not diagnose the fault or judge whether the vehicle is safe to drive. The workshop sets the rules for requests that concern safety.
How we workDetails ↓
- 01Call30 minutes with an engineer, free
- 02PrototypeA working system on your process and data, free
- 03ImplementationTarget and fixed price in the contract; 50% at signing
- 04Acceptance50% once the agreed target is met
- analysis with an engineer and a free prototype
- €0
- team in ChatGPT or Claude, up to 10 people
- from €1,150
- automation, agent or application for one process
- from €3,000
We belong to the OpenAI and Anthropic partner programmes and learn through their official courses. What this means ↓
Author: Syntalith · 3 min read
Statistics
Automotive and AI: statistics 2026
2.3 mo.
payback of €3,000 at 800 requests a month: €1,861 of work today, €563 with checking and the model. Scenario.
47%
According to the European Commission (2026), customer service is the most common use of AI among small and medium-sized EU companies that already use it: 47%.
Source: European Commission, Flash Eurobarometer 578 (2026) ↗
15%
According to Brynjolfsson, Li and Raymond (2025), customer support agents with AI help resolved 15% more issues an hour on average, in a study of 5,172 people.
Source: Brynjolfsson, Li and Raymond, The Quarterly Journal of Economics (2025) ↗
14.4%
According to Eurostat (2025), 14.4% of EU companies in the trade and repair of motor vehicles used AI.
Source: Eurostat, AI by NACE activity (isoc_eb_ain2) (2025) ↗
50%
According to the European Commission (2026), half of the small and medium-sized companies in the EU name high investment costs as a barrier to AI.
Source: European Commission, Flash Eurobarometer 578 (2026) ↗
Figures checked: 22 October 2026
- The symptom in the client's words
- When and in what conditions it occurs
- Information to fill in during the conversation
- Safety-related requests handled by the workshop's rules
A symptom is no name of a damaged part
In a fictional request, the client writes: „sometimes I hear a knocking after pulling away, then it stops”. A draft that names a specific part and the cost of replacing it adds a diagnosis that nobody has confirmed. The adviser needs the original description and questions about the circumstances.
A possible note says that, according to the client, the symptom appears after pulling away and does not always come back. Further details are missing. It is material for a conversation, with no repair instructions. The workshop sets the rules for handling requests that concern safety. A language model does not decide them.
A form can remove some of the gaps
First check the current form and the workshop's management system. Fields for the case identifier, the vehicle details and the contact can cut retyping. AI makes more sense where the client describes the situation in free text or sends several messages.
The result should point back to those messages. During a correction, keep a record of who changed the description and why. If the client sends a new detail, do not create a second request automatically only because the message came through another channel. The rule for merging cases has to be agreed with the team.
What should the first trial show?
Prepare a fictional request with an incomplete description, a follow-up to the same case and a similar description about another vehicle. The adviser checks the correct assignment, that the client's words were kept and the questions that need a further conversation. A confident-sounding diagnosis that is not in the material is a serious error.
Assess the review time separately. If the ordered note means rereading all the messages, fix the references or narrow the scope. This is a proposed acceptance trial; we present no rollout in a workshop and no result of a model test.
The next stage of the adviser's work is described in the AI offer for workshops: preparing material for a quote and linking the client's answer to the right version of the estimate. It is a proposed administrative scope, with no diagnosis of the vehicle.
When should you order an automation?
With a few requests a day, a template and the adviser's way of working may be enough. Connecting the mailbox with the workshop system is worth considering when the same material is retyped again and again and the assignment rules are already known.
Syntalith can build request handling automation. The service price levels are on the pricing page. The first scope can be limited to preparing a note for approval. Connecting to the workshop software and saving changes require the available interfaces to be confirmed. Searching manuals is a further, separate topic, described in the post on a knowledge base for multi-brand service. For the first call, a description of the path of a request from the client to the adviser is enough.
What it costs
- Today
- 800 requests × 4 min = 53.3 h (€1,861)
- With the system, a person checks the result in 30% of the time
- €558
- Model, 1 call per case, Claude Sonnet 5.5
- €5
- Cost of one case
- €2.33 today, €0.70 with the system
- Saving
- €1,298 a month
Payback of €3,000
2.3 months
The volume, the time and the checking share are the scenario's assumptions. €34.90 an hour is the EU average hourly labour cost for 2025 according to Eurostat. Model: 2,000 input tokens and 300 output tokens per call at Anthropic's rates, at the ECB rate of 1.1205 USD per euro on 22 October 2026.
- The time from receiving a case to the result, before and after the change
- Cases the application passed to a person
- Corrections by the person checking
- The cost of the model and maintenance at the target volume
Where to start
In the free review we choose one process and count the volume and the team's working time, and within 2 working days you receive a written summary with a price. We build the automation as part of AI automations, from €3,000 net.
Worked example · Scenario
How we count the payback: service requests
We run the same count for your workshop in the free review.
€2,327
a month for preparing 200 first repair estimates
Payback: 1.5 mo.
20 minutes for a first repair estimate and checking in 15% of the old time are our assumptions. €34.90 an hour is the EU average hourly labour cost for 2025 according to Eurostat, for companies with 10 or more employees. Model cost: reading the enquiry in our quoting system (about €0.0008) plus pricing the lines (Claude Sonnet 5.5, about 4,000 + 1,500 tokens), at the ECB rate of 1.1205 USD per euro on 22 October 2026. The extra margin is a scenario: we assume 1 in 20 estimates sent the same day ends in an order that a slow reply loses today, with a margin of €92. In the review we count it on your sales process.
See the full calculation →