Blog · Published · Updated
AI training for corporate tax teams: sources and review-ready memos
Short answer
AI training for a tax team teaches you to prepare a memo in which the reviewer sees at once the sources, the established facts and the questions for assessment. The key exercise is catching a draft that carries a conclusion from an older memo over to a new transaction. At Syntalith team training costs from €1,150 net a day. With 5 people each saving 1.5 hours a week, it pays back in 1.1 months after licence costs.
- net per day of closed training for the team
- from €1,150
- people in one training group
- up to 15
- after this time, we measure the effect on the same tasks
- 30 days
We belong to the OpenAI and Anthropic partner programmes and learn through their official courses. What this means ↓
Author: Syntalith · 3 min read
Statistics
Working with sources and AI: statistics 2026
1.1 mo.
payback of a €1,150 training at 5 people saving 1.5 hours a week each: €1,133 of time regained, €89 of licences a month. Scenario.
40%
According to a study published in Science (2023), ChatGPT cut the time 453 professionals spent on writing tasks by 40%.
Source: Noy and Zhang, Science (2023) ↗
25.1%
According to a Harvard Business School study (2023) of 758 consultants, tasks within the model's capabilities were done 25.1% faster with AI.
49%
According to Anthropic (2024), combining contextual embeddings with BM25 search cut failed retrievals by 49%, and by 67% with reranking.
15.1%
According to Eurostat (2025), 15.1% of people aged 16 to 74 in the EU used generative AI tools for work.
Figures checked: 22 October 2026
An older memo and a new transaction
A colleague from the business asks about a transaction in the current period. The draft prepared with AI repeats a conclusion from an older internal memo, although nobody has checked the material it was based on or whether it fits the present case. The reader will assume they have an answer about this transaction.
The corrected material points to the older memo and its source, and records what needs checking: whether the source covers the right entity, country and period, and whether the current transaction matches the one described before. The reviewer can go back to the document and judge whether the conclusion applies. We leave the tax assessment to a person with the right expertise, and the training covers preparing the material for that assessment.
When working with sources, we use the official EUREKA search tool of the Polish tax administration, which finds tax rulings and other tax and customs material by document type, phrase, reference number and date of issue. Finding a document is the starting point for checking what it means for the case.
- 01Questiona sharpened business question
- 02Factsconfirmed facts kept apart from missing ones
- 03Sourcesa document with its entity, country, period and date
- 04Draftan AI memo compared with passages from the sources
- 05Questionsa request to the business with no tax jargon
- Every statement with a reference to a passage in the source
- The entity, country and period of the case at the start
- Conclusions from older memos marked for checking
- Questions to the business that make sense without knowing the law
What it costs and when it pays back
- Time saved
- 5 people × 1.5 h a week × 4.33 = 32.5 h
- Value of that time
- €1,133
- Licences
- $100 (€89)
- Left after licences
- €1,044 a month
Payback of a €1,150 training
1.1 months
The hours saved are the scenario's assumption, which we check by timing the tasks before the training and after 30 days. €34.90 an hour is the EU average hourly labour cost for 2025 according to Eurostat. Claude Team licences at $20 per person a month, at the ECB rate of 1.1205 USD per euro on 22 October 2026.
We measure the hours in the scenario before the training and after 30 days, on the same team tasks. If after a month checking AI texts takes as long as writing from scratch, we change the exercises or the scope of the tool.
According to a study in Science (2023), ChatGPT cut the time 453 professionals spent on writing tasks by 40%. In tax memos we also count the memos that came back to the author with a question about a source.
A register and your current database may be enough
If the team loses time mostly on finding documents it already knows, an organised register or better use of the current database will help. Training answers the need to prepare clear material from those sources and to ask the questions that let you complete it.
Time goes on finding known documents
a register or a better database
no build
Memos come back with questions about sources and facts
training on working with sources with AI
from €1,150 a day
A large archive of your own opinions
search across company documents
from €3,000
Where to start
We set the programme on a memo that came back to its author with questions about sources or facts. A senior specialist says which explanations the review needs, and a person from the business checks that the request for facts is clear. We run the workshop as part of team training. For the free analysis, a description of one task the team does every week and of the tools it uses is enough. You get the written training programme within 2 working days.
Worked example · Scenario
How we count the payback: a tax team
We run the same count for your team in the free analysis.
€2,094
a month for preparing 20 reports
Payback: 1.7 mo.
3 hours to collect the data and prepare a report 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: checking the commentary in our reporting system (about €0.0002) plus compiling the data (Claude Sonnet 5.5, about 30,000 + 2,000 tokens), at the ECB rate of 1.1205 USD per euro on 22 October 2026.
See the full calculation →
