Blog · Published · Updated
An AI knowledge base for technical procurement
Short answer
An AI knowledge base in technical procurement finds an engineer's earlier decision on a substitute part and shows which assembly revision it covered, so the buyer asks a specific question in place of rebuilding the history. At Syntalith an AI app on company documents costs from €3,000 net. At 300 questions a month with 20 minutes of searching each, it pays back in 1.0 month.
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
- net per application, fixed price before contract
- from €3,000
- from signing the contract to acceptance
- 4–10 weeks
- free prototype on your documents
- €0
We belong to the OpenAI and Anthropic partner programmes and learn through their official courses. What this means ↓
Author: Syntalith · 3 min read
Statistics
AI knowledge bases: statistics 2026
1.0 mo.
payback of a €3,000 app at 300 questions a month: €3,490 of searching today, €529 with the app. Scenario.
20.0%
According to Eurostat (2025), 20.0% of EU companies with at least 10 employees used AI.
Source: Eurostat, 20% of EU enterprises use AI technologies (2025) ↗
53.6%
According to Eurostat (2023), 53.6% of EU manufacturing companies used ERP software.
49%
According to Anthropic (2024), combining contextual embeddings with BM25 search cut failed retrievals by 49%, and by 67% with reranking.
50%
According to the European Commission (2026), half of EU SMEs and mid-caps name high investment costs as a barrier to AI.
Source: European Commission, Flash Eurobarometer 578 (2026) ↗
Figures checked: 22 October 2026
What the earlier substitution approval says
A buyer is considering part B in place of part A. A signed engineering note approves B in the assembly at revision R2, limited to R2. The current requirement concerns R3, and the note does not explain the technical reasoning. The buyer remembers that there was an approval once, but not what it covered, so today they go back to the engineering department.
- 01DecisionThe assistant finds the signed note and its scope.
- 02DifferenceIt shows that the current requirement concerns another revision.
- 03QuestionThe buyer asks the engineer: can B be used in R3?
- 04AssessmentThe person responsible for the technical assessment decides on use in the new revision.
- 05RecordThe new decision is filed with the assembly and the earlier case.
A new buyer sees the earlier order for part B, but the order alone does not say what the part was approved for, because the approval sits in the project folder. Joining these materials lets anyone open the decision without looking for the person who remembers that case, also when duties are handed over.
What a procurement knowledge base costs and when it pays back
- Today
- 300 questions × 20 min = 100 h (€3,490)
- With the system, a person checks the output in 15% of the time
- €524
- Model, 1 call per case, Claude Sonnet 5.5
- €4.59
- Cost of one case
- €11.63 today, €1.76 with the system
- Saving
- €2,961 a month
Payback of €3,000
1.0 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: 6,000 input tokens and 500 output tokens per call at Anthropic's rates, converted at the ECB rate of 1.1205 USD per euro on 22 October 2026.
The model is a small item here. The result depends on the time that searching and asking around take today, and on how much of it is left for checking the answer.
Before we quote a build, we run a test on 20 to 100 questions the team has already asked: we compare the app's answer with what an employee found by hand and measure the time from the question to the source. The test takes 5 working days and costs €0.
Use the change history in your existing system
Check the ERP or the product documentation system first. In Dynamics 365 Supply Chain Management, Microsoft describes product versions and engineering changes run through an approval workflow. If decisions with their reasons sit against the part and the assembly revision, a more convenient view is enough.
Decisions sit with the part and revision in the ERP
a better view and search
no new app
Purchase, assembly revision and decision in different places
an AI app that joins the sources
from €3,000
A question about availability and lead time of an approved part
an agent for parts requests
a separate scope
Connect engineering decisions with purchasing work
The first scope is a chosen group of parts and assemblies where buyers and engineers most often rebuild the same cases. The engineering department names the material and judges whether the answer reflects it, and buyers check whether it helps them prepare a specific question. Retrieval decides the quality: according to Anthropic (2024), contextual embeddings with BM25 cut failed retrievals by 49% (Contextual Retrieval). Questions about the availability and lead time of parts are covered in the post about an agent for spare parts requests.
Worked example · Scenario
How we count the payback: answers from company documents
We run the same count for your documents in the free analysis.
€3,723
a month for finding answers to 800 questions in documents
Payback: 1 mo.
Our assumptions are 8 minutes to find an answer in documents or ask a colleague, and checking in 15% of the old time. €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: Claude Sonnet 5.5, about 6,000 + 500 tokens per answer from documents, at the ECB rate of 1.1205 USD per euro on 22 October 2026.
See the full calculation →
