Today
Which cleaner, concentration, temperature and contact time fit an application is often known only to two or three experienced application technicians. Customers describe their problem by e-mail or phone, often incompletely, and wait days for a recommendation. The answers sit in lab reports, old recommendations, visit reports and technical data sheets. If material compatibility is missed, the cleaner attacks aluminum, seals or coatings, and the recommendation turns into a complaint.
An example
A delicatessen producer asks Orvelon for a foam cleaner for its filling line: fat and protein residues from salads and dips, stainless steel, conveyors on aluminum profiles, daily cleaning with water of 20 °dH at 40 °C. By soiling and process the strongly alkaline AR 240 at 3 % fits, but the material check rules it out because at pH 13 it attacks the aluminum. The agent recommends the inhibited alkaline foam cleaner AR 218 at 4 % with 15 minutes contact time, backed by a lab corrosion test on aluminum and two meat and delicatessen plants that have used it for years. The application technician adopts the recommendation and sends it to the customer the same day with technical data sheet and safety data sheet.
Your data, wherever it lives
The knowledge behind a recommendation is spread out today: customer inquiries in the mailbox, past recommendations and visit reports in the CRM, tests in the LIMS, formulations and product data in the PLM, technical and safety data sheets in the DMS, and much of it only in the e-mails and notes of senior technicians. demi reads, understands and links it per product and application, with your rights. The AI proposes product and dosage in minutes instead of days, and the application technician checks and decides.
Where the data lives today
- E-mails
- CRM
- PLM
- Databases
- DMS & archive
- ERP
- PDFs
How demi solves it
Three layers in this order: connect the data first, then the rights, then the AI.
- 1 · ConnectData Layer
Data sources in this case: customer inquiries in the mailbox, visit reports and past recommendations in the CRM, formulations and product data in the PLM, cleaning and corrosion tests in the LIMS, technical and safety data sheets in the DMS, and customer master data and delivery history in the ERP. demi replaces nothing and links every product to the soiling, materials, processes and conditions under which it was tested and used. For a new application you find the tests and customers where it already worked.
- 2 · PermitPermission Layer
Formulations are visible only to development and the lab; sales and application technicians see product data, test results and recommendations. Prices and terms stay with the responsible sales rep, a customer's data with the team that serves that customer. A recommendation goes out only after an application technician releases it; the release is recorded.
- 3 · ActIntelligence Layer
The agent reads soiling, materials, temperature, process and water hardness from the inquiry and asks for what is missing. It finds products whose tests and installations match these conditions and proposes concentration, temperature and contact time, each value with the test or data sheet it comes from. The material check runs against your compatibility tables, stored as rules, not phrased freely.
How it runs
- 1
The customer describes soiling, materials, temperature and process by e-mail, form or in a visit report.
- 2
The agent reads the details, asks for anything missing and searches your range for matching products.
- 3
For each proposal it shows concentration, temperature and contact time from tests and comparable customer plants.
- 4
It checks every product against the materials named and rules out anything incompatible.
- 5
The application technician checks, adjusts and sends the recommendation with data sheet and safety data sheet.
What changes
Recommendations in hours instead of days
Fewer complaints caused by incompatible materials
The knowledge of experienced technicians is available to all of sales
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
+162 hfreed per year
Assumptions and calculation
demi reads the application and proposes product and dosage with tests and data sheet; you check and send.
90 min → 63 min per case
How we calculate
30 application recommendations × 90 min × 30% × 1.0 (manual) × 12 months = 9,720 min = 162 h
Value per year, before demi costs
Freed per year+162 h
Assumption: labor cost of $48.62 per hour in US manufacturing in June 2026 (BLS), plus workplace, IT and overhead.
= Value per year, before demi costs
+$10,530
gained per year
162 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
What if the customer describes the application only vaguely?
Then the agent asks targeted questions, for example about materials, temperature, process or water hardness, and marks what is still open. A recommendation with missing details does not go out without a note. It also reads photos of the plant or the soiling.
How does demi know which product suits which materials?
From your compatibility tables, technical data sheets and corrosion tests in the lab. The agent checks every proposed product against every material named and shows the source. If no test exists for a combination, it says so and proposes one.
Does sales see our formulations?
No. Formulations stay with development and the lab. For sales the agent works with product data, test results and released data sheets; who sees what, you set in the access rights.
Does demi always attach the right safety data sheet?
The agent takes the current released version for the customer's country and language from the DMS and names the version. If there is no version for a country, it says so instead of attaching another. Nothing is sent before release.
