Today
A visitor from a brewery lands on the same generic page as one from a power plant and leaves, because nothing speaks to their application. Dedicated pages per industry would help, but each needs values, references, images and text, and the next data sheet revision makes it outdated. So there is one page for everyone, and enquiries from new industries arrive only by chance.
An example
Aldevo Wassertechnik has one page for all its filter systems. The agent finds 34 breweries, 19 dairies and 12 power plants among the customers in the ERP and CRM, plus 40 enquiries from the beverage industry last year, and proposes six pages. First it drafts “Filtration for brewing water”: the FA DN 150 filter system with 50 m³/h, PN 10, 0.1 µm filter fineness and DVGW approval from master data, two approved brewery references, photos from the media library and the terms brewers use in their enquiries, such as brewing water treatment, turbidity and CIP cleaning. Product management corrects one value, the beverage application engineer appears as the contact, and the page goes live two days later.
Your data, wherever it lives
What an industry wants to know is scattered across your company today: product values in the PLM, orders and reference customers in the ERP and CRM, application reports and presentations in SharePoint, photos in the media library, enquiries in e-mails and search terms in your website analytics. demi reads, understands and links them, with your rights. The AI drafts a dedicated page for every industry and application within hours, and marketing and product management decide what goes live.
Where the data lives today
- PLM
- ERP
- CRM
- E-mails
- SharePoint
- Photos, video & audio
- Web & portals
How demi solves it
Three layers in this order: connect the data first, then the rights, then the AI.
- 1 · ConnectData Layer
The case needs data from the PLM, ERP, CRM, e-mails, the file share, the media library and your website. demi replaces nothing and links orders and reference customers per industry with products, variants, data sheet values, application reports, photos and the terms these customers use in enquiries. demi hands approved pages to your website system as drafts.
- 2 · PermitPermission Layer
Only products and values released for publication go on the pages; never prices, margins or internal test reports. Reference customers appear only with an approval in the CRM, contact persons only with their consent. Product management and marketing review every page before it goes live, and the approval is recorded.
- 3 · ActIntelligence Layer
The agent sorts orders, references and enquiries by industry and application and shows where a dedicated page pays off. It writes the page in your layout, with values carried over unchanged from master data and the technical and search terms of the audience. When a data sheet changes or a reference is added, it proposes the update and reports which pages bring enquiries.
How it runs
- 1
The agent sorts orders, references and enquiries by industry and application and suggests where a page of its own pays off.
- 2
For each chosen application it gathers products, values, references, photos and the terms these customers search and write with.
- 3
It drafts the page in your layout: the customer's task, the solution, technical values, a reference, a contact person and common questions.
- 4
Product management and marketing check values and statements and approve the page.
- 5
When a data sheet changes or a reference is added, the agent proposes the update.
What changes
Every visitor finds a page about their application
Technical values that match the current data sheet
Enquiries that already fit the application
What does it bring you?
Check it with your own numbers
Every starting value is an assumption you can change. The calculation sits right below, step by step.
Your numbers
Result
+154 hfreed per year
Assumptions and calculation
demi drafts industry pages from master data, references and enquiries; you check and approve.
8 h → 4.8 h per case
How we calculate
4 industry pages × 8 h × 40% × 1.0 (manual) × 12 months = 154 h
Value per year, before demi costs
Freed per year+154 h
Assumption: labour cost of €49.50 per hour in German manufacturing in 2025 (Destatis), plus workplace, IT and overheads.
= Value per year, before demi costs
+€10,010
gained per year
154 h ÷ 1,760 h per role = 0.1 full-time roles
An estimate, not a promise. The starting values are assumptions for an industrial company with 150 to 500 people.
Combine with more cases and get the packageQuestions about this use case
How does demi know an industry's terms?
From these customers' enquiries, e-mails and specifications, from application reports and from the search terms that bring visitors to your website today. The agent shows each term with its frequency and source. You decide which ones go on the page.
How do the pages stay current?
Every value on the page is linked to its master data record. When the record changes, for example with a new data sheet revision, the agent proposes the change on every affected page, and it assigns newly approved references to the matching pages. It flags pages nobody has touched for a long time.
Are customers named without their consent?
No. The agent names a reference customer only if the CRM holds an approval; otherwise it writes neutrally, such as “a brewery in Franconia”. It uses photos from customer plants only if the approval is recorded in the media library.
Does demi work with our website system?
Yes. demi delivers the pages as drafts in your content management system's templates, with page title, description for search engines, internal links and structured data. Publishing happens in your system after approval, as before.
