Integrations and onboarding
Starts with the data you already have.
No new stock system, no web project. DealerBrain loads your listing export, reads your websites and video channel, and plugs into each site with one line of code. We set it up with your team and keep tuning it.
Data sources
- Inventory feedDaily
Your listing export, refreshed daily, mapped to physical sites, duplicate-free, with images.
- New modelsPer brand
Model lines, trims, packages and colours for the brands you represent.
- Website contentFull text
Your magazine, articles and pages, imported as full text behind the answers.
- Video channelSynced
YouTube synced, transcribed, summarised and matched to stock cars.
- DocumentsUpload or link
Brochures, price lists and policies, added by upload or by link.
- DMS and CRMCustom
Custom connections through the REST API, behind the same data layer.
Where it shows up
- Brand websites
- Group website
- Campaign pages
- Partner sites
Your stock, loaded without losing anything
We import the listing export you already produce, so you can start without a new system.
- 01.01Listing export import
- Loaded from the export your portals or DMS already produce.
- 01.02Mapped to sites
- Every car belongs to a physical location.
- 01.03Duplicate-free
- The same car listed twice appears once.
- 01.04Images included
- Galleries come along with each vehicle.
- 01.05Scheduled refresh
- Daily refresh with instant indexing.
- 01.06Clean names
- Brand and model name variants are normalised.
Your videos, watched and indexed for you
Videos become searchable knowledge and appear next to the right cars.
- 02.01Channel sync
- New uploads are picked up automatically.
- 02.02Speech to text
- Every video is transcribed.
- 02.03Summaries
- AI summaries make content quotable in answers.
- 02.04Matched to stock
- Videos are linked to the cars they are about.
- 02.05Channel filters
- Choose which playlists and formats count.
- 02.06Cost cap
- A processing cap keeps AI costs predictable.
Open by design
Your data, your models and your systems. DealerBrain fits into the stack you already run.
- 03.01REST API
- Read and write conversations, contacts, leads and inventory from your own systems.
- 03.02Custom integrations
- DMS, CRM, marketing and reporting connections built to your needs.
- 03.03Customisation
- Branding, conversation flows, rules and interface adapted to your group.
- 03.04Bring your own key
- Connect your own AI provider account. Usage and billing stay with you.
- 03.05Your model subscriptions
- Use the models you already license, and switch when a better one appears.
- 03.06Your data stays yours
- Conversations, contacts and leads can be exported at any time.
One line of code per site
Works on any website and does not interfere with the host site.
- 04.01One-line embed
- No development project needed.
- 04.02Unlimited websites
- Group site, brand pages, microsites, campaign and partner pages.
- 04.03Isolated rendering
- No conflict with the host site’s design or scripts.
- 04.04Site key per website
- Every conversation and lead is tied to its source.
- 04.05Branding per site
- Colour, logo, position, corner radius, light or dark.
- 04.06Features per site
- Choose where lead forms, videos, articles or live agents appear.
- 04.07Page context and UTM
- Knows the page and attributes leads to campaigns.
- 04.08Languages per site
- A default language per website, more switchable.
Your AI, wherever it has to run.
DealerBrain works with many AI models and protocols. Run the AI layer on public APIs, on EU-hosted APIs, in your private cloud or on your own servers, and change your mind later without rebuilding anything.
Where the AI runs
Public APIs
The leading model providers, with your own key or through DealerBrain.
EU APIs
Models served from data centres in the European Union, so every request stays in the EU.
Private cloud
Models in your own cloud account, under the agreements you already have.
On-premises
The AI layer runs on GPU servers in your own data centre. Prompts and answers stay on your hardware.
Compatible with
- Gemini Enterprise Agent PlatformFormerly Vertex AI Platform
- Microsoft FoundryFormerly Azure AI Studio
- Amazon BedrockAmazon Web Services
Tested models
- GPT-6OpenAI
- Gemini 4Google
- Claude 5.5Anthropic
- DeepSeek 4.1DeepSeek
And many more models and protocols, including open-weight models running on your own GPUs.
AI infrastructure services
From GPU to gateway, we set it up with you.
Running AI on your own infrastructure takes more than a model. Our team helps with every layer, from the hardware to day-to-day operation.
Talk to our AI team- 01GPU server procurement
Sizing and sourcing the right hardware for your volume and your models.
- 02GPU cluster installation
Installation and configuration of the cluster in your data centre.
- 03Efficient model operation
Model choice, serving and tuning for speed and cost per answer.
- 04Monitoring
Load, latency, errors and costs, watched and reported.
- 05AI gateway setup
One controlled entry point for every model, with routing, keys, limits and logging.
Set up with your team, then kept sharp
DealerBrain is configured for each dealer group as a service on top of the platform.
- 01
Configure
- Persona, greetings and suggested questions
- Your tone, brand policy and sales rules in the rulebook
- Sites, brands, salespeople and contacts
- 02
Connect
- Inventory feed import and daily refresh
- Website content and video channel
- Widget placement on brand and site pages
- 03
Launch
- Training for agents and salespeople
- Inbox and lead flow rollout
- Opening hours, triggers and per-site display
- 04
Tune
- Fixes based on real conversations
- Data quality and content relevance checks
- Regular workshops with your team
Which feeds do you use?
We are prioritising listing portals and export formats together with our first customers in each market. Tell us what you run today.
Next step
See it answer for your own dealer group.
We walk you through DealerBrain with scenarios built around your brands, your sites and the kind of stock you carry. Thirty minutes, no slides.
