Blog · Published · Updated
AI agents for clarifying product data with suppliers
Short answer
For product data, an AI agent links missing catalogue fields with the supplier's correspondence, checks whether a reply concerns the right product and parameter, and writes a more precise question for the employee. The person responsible for the catalogue approves the completed product card. The agent costs from €3,000 net at Syntalith and, with 500 gaps in product cards a month at 10 minutes each, pays back in 1.5 months.
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 agent for one process, build paid once
- from €3,000
- from signing the contract to acceptance
- 6–16 weeks
- free prototype on your cases
- €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
Product catalogues and AI: statistics 2026
1.5 mo.
payback of a €3,000 agent with 500 gaps in product cards a month: €2,908 of work today, €883 with checking and the model. Scenario.
15.5%
According to Eurostat (2025), 15.5% of EU retail trade enterprises with at least 10 employees used AI.
Source: Eurostat, Artificial intelligence by NACE Rev. 2 activity (isoc_eb_ain2) (2025) ↗
34.7%
According to Eurostat (2025), 34.7% of EU enterprises that use AI use it for marketing or sales.
Source: Eurostat, Use of artificial intelligence in enterprises (2025) ↗
−87%
According to Anthropic (2025), AI cuts the time of a task on invoices, memos and other documents by an estimated 87%.
Source: Anthropic, Estimating AI productivity gains (2025) ↗
14–56%
According to Anthropic (2025), controlled trials found smaller time savings, from 14% to 56%.
Source: Anthropic, Estimating AI productivity gains (2025) ↗
Figures checked: 22 October 2026
Which dimensions we need
A shop has a confirmed mapping of the supplier's code to a product in its catalogue. For the product card it needs the dimensions of the assembled product, and the supplier's file holds only the carton dimensions. An employee asks, the supplier replies and again gives the packaging dimensions. With several hundred open questions, the employee has to remember whom they wrote to, whether a reply concerns the same product and what is still to be clarified.
The agent shows the data next to the missing field and proposes a question: „Please send the dimensions of the assembled product for the code below, with the unit of measure. The file gave us the carton dimensions, and we need the dimensions of the product itself”. When the supplier sends the packaging dimensions again, the agent points out that the reply does not explain the product's size and prepares a clarification. Only a reply with the dimensions of the assembled product becomes a proposal to complete the card, which the person responsible for the catalogue approves.
- 01Gapan empty or suspicious field in the product card
- 02Questiona specific request to the supplier with the code and the unit
- 03Replythe agent checks whether it concerns the right parameter
- 04Clarificationa further question when the reply does not resolve the gap
- 05Carda proposal to complete it, for approval
When a colleague takes over the case, they see the latest question and the supplier's explanation on the product card. A later correction from the supplier appears next to the previous value.
What it costs and when it pays back
- Today
- 500 gaps in product cards × 10 min = 83.3 h (€2,908)
- With the system, a person checks the output in 30% of the time
- €873
- Model, 2 calls per case, Claude Sonnet 5.5
- €11
- Cost of one case
- €5.82 today, €1.77 with the system
- Saving
- €2,025 a month
Payback of €3,000
1.5 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: 4,000 input and 400 output tokens per call at Anthropic's rates, converted at the ECB rate of 1.1205 USD per euro on 22 October 2026.
The count covers the work around a case: searching, reading, writing and passing things on. What remains is the time to check the output, 30% of today's work in the scenario. Before the quote we measure both times on 20 to 100 cases the team has already handled, in a €0 test that takes 5 working days.
Check how much work your current PIM can take on
A PIM is the system where a company keeps product information. Akeneo Supplier Data Manager lets suppliers submit data in one place. If the supplier uses the form and enters the right data straight away, an agent is unnecessary. With few gaps, a list of questions with a responsible person is enough. An agent makes sense when replies arrive in different emails and attachments. When you first have to work out which supplier code matches a product, the article on matching supplier products is the closer fit.
Suppliers enter data in the PIM portal
the supplier portal in the PIM
licence
A few dozen gaps a month
a list of questions with an owner
no build
Hundreds of gaps, replies in emails and files
an agent that clarifies data with suppliers
from €3,000
Where to start
The first scope covers one type of missing data and the sources the catalogue team uses. On a finished case we check whether the employee reaches an approved card faster. We build this as part of AI agents. For the free process review, a description of one case from recent weeks and the tools it ran through is enough. You do not need to send confidential documents, and you get a written plan within 2 working days.
Worked example · Scenario
How we count the payback: AI in e-commerce
We run the same count for your catalogue in the free process review.
€2,792
a month for handling 800 orders that bypass the checkout
Payback: 1.3 mo.
6 minutes to retype an order or answer a question and checking in 15% of the old time are our assumptions. €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: reading the order in our system (about €0.0006) plus matching the lines (Claude Sonnet 5.5, about 3,000 + 500 tokens), at the ECB rate of 1.1205 USD per euro on 22 October 2026.
See the full calculation →
