Let's talk






Withholding tax reclaim and relief at source
We support Kapitalverwaltungsgesellschaften (German fund management companies), depositaries and custodian banks through the rollout of reclaim and relief-at-source systems – from the as-is analysis until the team trusts the system. And as a UX consultancy we design the usability of such software for its vendors. Always on one side only per project.
Beabstracto is a UX studio in Munich for complex expert systems. In withholding tax reclaim and relief at source we work on two sides, never at the same time: we support fund management companies (KVGs), depositaries and custodian banks through the rollout of their reclaim system – as-is analysis, functional concept, status model, UAT, go-live, key-user training and hypercare. And we advise vendors of such software on UX: research with expert users, modelling of statuses and deadlines, UI definition, usability tests. First results of the as-is analysis are available after two to four weeks.
The pace is set from outside: electronic applications to the BZSt since 2024 (§ 50c EStG), MiKaDiv for capital income from 1 January 2027 (§§ 45b, 45c EStG) and the FASTER Directive from 2030. Keeping those dates takes a system that shows deadlines, documents and approvals per case in one place – and a team that trusts it more than its Excel list.
Rarely because of the software itself. Usually because deadline, document and status of a case are not visible in one place – and nobody translates between the business unit, fund accounting and the vendor.
The withholding tax reclaim system has been procured, the vendor training was months ago. Applications are still produced the old way. Nobody in the business unit has the time to think the process through to the end inside the system, and nobody has been named to do it.
Filing deadlines differ by source country and often run for years after the dividend is received. Who has to file what, where and by when is kept by one person in a list – because the system does not show the deadline per case the way the team needs it to decide.
Certificate of residence, power of attorney, tax voucher or dividend statement: an application is only complete when every document required by the source country is in place. Without a completeness indicator in the system, the gap shows up at dispatch – or when the authority queries the claim.
When a fund management company takes over the reclaim that used to sit with the depositary, the data set the system assumes is often missing: sub-custodians, central securities depositories, reallocations during the year. The system asks for data nobody in-house ever had to record.
Holdings, income and tax withheld come from the back-office or fund accounting system. If granularity, reference date or identifier do not match, the import gets stuck. Business unit and development then discuss the same problem in two different languages.
From 1 January 2027, capital income from shares and depositary receipts must be reported electronically to the BZSt. The format is manageable. What is usually open is who delivers the custody chain per holding, who checks and who approves under the four-eyes principle before the report goes out.
Not marketing figures, but the dates that shape daily work in the business unit.
AbzStEntModG of 2 June 2021, BGBl. I p. 1259
BZSt, §§ 45b, 45c EStG
Directive (EU) 2025/50, Art. 14
Directive (EU) 2025/50
Two sides, clearly separated – and the business analysis that both sides need.
We work on your side until the reclaim system is used in daily operations – not just until acceptance. We capture the real processes, translate between the business unit, fund accounting and the vendor, and accompany key users through UAT, go-live and hypercare. The goal: applications filed on time, fewer queries to the vendor, no more parallel list.
We work on your side until the reclaim system is used in daily operations – not just until acceptance. We capture the real processes, translate between the business unit, fund accounting and the vendor, and accompany key users through UAT, go-live and hypercare. The goal: applications filed on time, fewer queries to the vendor, no more parallel list.
Your users are tax and operations professionals, not casual users. We make the domain logic behind your product explicit – roles, end-to-end processes, risk points, deadlines – and translate it into status models, information architecture and interaction patterns that expert users understand without lengthy training. Validation happens with real users, not with assumptions.
Between procurement and daily use lies the work nobody usually has time for: describing requirements so that business unit, IT and vendor understand the same thing. We take the role of Business Analyst or Requirements Engineer with a focus on tax reporting and UX – from the functional concept through test concept and UAT to acceptance.
We work in the vocabulary of the business unit, not in that of an agency. These terms come up in every project:
The sequence is the same for every rollout. The scope of each step depends on how much is already running in the system.
We observe how applications really come about today: which Excel lists exist, which decision they support, where data from fund accounting arrives and where it gets stuck. The result is a process map from entitlement to application, with exceptions, and the list of gaps between process and system.
2–4 weeks
With key users and the head of tax or fund accounting we define which statuses a case passes through, which documents make it complete and which deadline drives it. The result is a functional concept that business unit, IT and vendor can all read.
Not everything has to work on day one. We order configuration, data deliveries and queries to the vendor by the dates that threaten first: source countries with short filing deadlines, MiKaDiv reporting dates, payment dates for relief at source.
We write test cases with the business unit, moderate the UAT and prepare key users so that they can guide their colleagues themselves. Training runs on the real process with realistic cases, not on the manual.
After go-live we stay until the parallel list is closed. We measure what counts – applications filed on time, cases with complete documents, queries to the vendor – and hand over to the key users once the numbers are stable.
From UX consulting for the vendor of a withholding tax reclaim software and from conversations with business units – without client data.
Domain knowledge is scattered: among developers, the product owner, tax specialists, management. Before we design an interface, we build a glossary and a regulatory mapping – which term means which obligation, which deadline, which document. Only when everyone understands the same thing by 'open' or 'approved' is design worth doing. Otherwise people discuss screens and mean processes.
In the business unit the question is never 'where is the button' but 'what is going on with this case, and do I have to do something today'. A status model that makes deadline, document completeness and approval visible per case replaces more Excel lists than any new feature. Where it is missing, the team builds its own – in a spreadsheet.
Users in tax and operations work with regulated workflows, audit trails and documents. Hiding complexity does not help them; it has to become manageable, auditable and safe. Filters, bulk actions and traceable validation matter more here than a tidy dashboard.
Every parallel list answers a question the system does not answer. Banning it loses the information; reading it reveals the missing status, the missing document or the stuck interface. That is why every as-is analysis of ours starts with the lists, not with the manual.
Stakeholders judge product direction by something that feels like software. A clickable prototype with realistic cases, deadlines and errors brings business unit, development and sales to a decision faster than any specification – and shows early where the status model does not hold.
Public cases and articles. We do not name banks, fund management companies or depositaries.
Withholding tax reclaim software (vendor)
New MiKaDiv reporting module and UX modernisation of the existing platform: glossary and regulatory mapping, personas, end-to-end processes, status model, information architecture and a prototype with realistic data that stakeholders at banks treated like production software.
→Article
Why parallel lists emerge at the interface between fund accounting and the reclaim system, which decision they support and what they reveal about missing statuses in the system.
→We orient our work on these texts; we claim no certification and no institutional affiliation.
Federal Law Gazette, Act of 2 June 2021 (BGBl. I p. 1259) · 2021
Moves the relief procedure from § 50d to § 50c EStG, limits exemption certificates to three years and makes electronic applications to the BZSt the rule from 2024 – the reason document and status logic has to live inside the system.
Bundeszentralamt für Steuern (BZSt, Federal Central Tax Office) · 2026 (handbook version)
Defines report types, XML format, deadlines and parties under §§ 45b, 45c EStG for income from 1 January 2027 – required reading for any functional concept of the MiKaDiv report.
Council of the European Union · 2024 (OJ L, 10 January 2025)
Introduces the eTRC, relief at source, quick refund within 60 days and a register of certified financial intermediaries; transposition by 31 Dec 2028, application from 1 Jan 2030. Sets the direction reclaim processes will take over the coming years.
OECD, Committee on Fiscal Affairs · 2013
The model for relief at source via authorised intermediaries on which FASTER builds; shows which role and evidence logic a system has to represent.
ISO/TC 159/SC 4 · 2020
Seven interaction principles, including suitability for the task, self-descriptiveness and robustness against use errors – our yardstick when we assess status displays and approval dialogues.
Kate Kaplan, Nielsen Norman Group · 2020
Names five layers of complexity, including the institutional one – exactly the layer at which regulatory obligations and the four-eyes principle reach into the interface.
Nielsen Norman Group · 2025
Experienced users do not fear change but the loss of productivity – the central argument for rollout support rather than training alone.
A 30-minute initial conversation: you describe where it gets stuck – we tell you honestly whether and how we can help.
Book an initial conversation




