Today
In mid-sized industrial companies, marketing is often a small team. Product texts are out of date and no longer match the data sheets. Every new variant means writing texts by hand, in every language.
An example
At Talwerk Antriebstechnik 60 new variants of a geared motor series are added. The agent writes web texts in German and English with power, ratio and protection class from the data sheets, such as 0.75 kW, i = 28 and IP55, plus a two-page brochure in the company template. Marketing changes two phrases and approves.
Your data, wherever it lives
Today performance values and variants sit in the PLM and ERP, data sheets as PDFs, images and project reports in folders on the drive, and the old texts on the website and in brochures. demi reads, understands and links them per product and variant, with your rights. The AI writes texts in every language to your style guide, with values taken unchanged from the master data, in minutes instead of an hour per text, and marketing reviews and approves.
Where the data lives today
- PLM
- ERP
- PDFs
- Photos, video & audio
- Folders & drives
- 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 PLM, ERP, CMS and file share. demi replaces nothing and attaches all values, variants, data sheets, images and project reports to each product. demi writes back only after approval.
- 2 · PermitPermission Layer
Marketing reviews and approves every text before it goes into the CMS, and customer names in reference stories appear only with the customer's consent. Every step is recorded.
- 3 · ActIntelligence Layer
The agent writes texts to the style guide and glossary and carries technical values over unchanged from the master data. When a value changes, it flags every text that mentions it.
How it runs
- 1
Clarify the data sources: PLM, ERP, file share and CMS
- 2
Add templates, brand guidelines and a glossary
- 3
The agent drafts texts per product and language
- 4
Marketing reviews and approves, the text goes into the CMS
- 5
When a value changes, the agent flags every text that mentions it
What changes
Consistent product information in every channel
More content from the same team
Texts that match the data
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
+288 hfreed per year
Assumptions and calculation
demi writes copy from master data and projects in your tone; you check and publish.
60 min → 36 min per case
How we calculate
60 product texts × 60 min × 40% × 1.0 (manual) × 12 months = 17,280 min = 288 h
Value per year, before demi costs
Freed per year+288 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
+€18,720
gained per year
288 h ÷ 1,760 h per role = 0.2 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
Can the agent use our corporate design?
Yes. You add templates, brand guidelines and tone of voice, for example for product pages, data sheets and brochures. The agent fills these templates instead of inventing its own layouts. Marketing can add new templates at any time.
How do the technical values stay correct?
The agent takes values such as power, dimensions or protection class straight from the master data and never rewords them. Every value in a text links to its source. When a value changes in the master data, the agent shows every text that needs an update.
Can it also write newsletters and trade articles?
Yes, based on product data, project reports and new releases. Reference stories are built from completed projects; customer names appear only once the customer agrees. Someone from your company always checks the content and approves it.
Which languages does it write in?
The ones you need. A glossary keeps technical terms consistent in every language. The technical values are identical everywhere because they come from the same master data.
