Services
Automating manual processes: from Excel macros to web applications
Five service areas, each named after the problem behind it. This page gives an overview and helps you place your case; the details are on the individual service pages.
01
Excel automation
Reports, calculations and reconciliations done by hand in Excel again and again.
More on Excel automation- The problem
- Recurring Excel work costs hours every week and is error-prone.
- Typical situation
- A team maintains evaluations or calculations in Excel. The same steps every week, and only one person knows how the file really works.
- Possible solution
- The workbook is rebuilt so the recurring steps run at the push of a button: structured data, VBA macros with error handling, clear inputs and outputs. Excel remains the working environment.
- A good fit when
- … a clearly bounded, table-based workflow is handled by one person or a small team.
02
Data automation and web scraping
Information that is gathered from websites, portals, PDFs or files on a regular basis.
More on data automation and web scraping- The problem
- The data exists, but it is scattered and gathered by hand.
- Typical situation
- Staff open the same sources again and again, copy values into a spreadsheet and check what has changed.
- Possible solution
- A program retrieves the sources regularly – websites, APIs, PDFs, files – checks the data and delivers it in structured form. Usually implemented in Python.
- A good fit when
- … the research is recurring, the sources are accessible and gathering is the bottleneck.
03
Reporting and dashboards
Reports and evaluations assembled from several files every single time.
More on reporting and dashboards- The problem
- The report is ready long after the period is over – and nobody knows which numbers are current.
- Typical situation
- Exports from several systems are merged in Excel, KPIs updated, charts recreated and transferred into a template.
- Possible solution
- Merging and checking the data is automated and KPIs get a fixed definition. The result appears as an updated report or, where needed, as a dashboard.
- A good fit when
- … the same evaluation is needed regularly and the rules for it are known.
- The problem
- Several people, several versions, no permissions: the Excel file has become a liability.
- Typical situation
- A calculation or record runs in a file that has become business-critical. Simultaneous work is hardly possible, changes are not traceable.
- Possible solution
- First I check whether the existing solution can be simplified. If not, a lean web application with central data, user roles and change history is built – for independent operation.
- A good fit when
- … several users, permissions or a traceable history are needed.
05
AI automation
Recurring tasks with texts and documents that cannot be captured in fixed rules.
More on AI-assisted automation- The problem
- Categorising requests, pulling details out of letters, sifting documents – work that requires reading and understanding.
- Typical situation
- Someone reads documents or free text every day, categorises it and transfers details into a system. Fixed rules don't apply because the cases are too different.
- Possible solution
- A language model takes over the part rules cannot handle – embedded in conventional software with structured inputs and outputs, review steps and human control where needed.
- A good fit when
- … unstructured information has to be processed and an occasional error can be caught.
Orientation
Which tool fits which problem?
The technology follows from the process. As a rough guide:
- Excel/VBA
- A clearly bounded, table-based process handled by one person or a small team. The data comes from a few known sources.
- Python
- Data processing and automation in the background: fetching, checking, transforming and passing on data – without anyone having to press a button.
- Report or dashboard
- Recurring evaluation and visualisation. An automatically generated report if it is read at fixed intervals; a dashboard if numbers are needed continuously.
- Web application
- Several users, permissions, central data, change history – or a process that has become too complex for Excel.
- AI (language models)
- Unstructured information: texts, documents, free text. Always embedded in a controlled process, never as an end in itself.
These are guidelines, not rigid rules. The best solution is often a combination – say, a Python pipeline that collects data and an Excel file in which the team keeps working with it.
Frequently asked about services
Answers before you have to ask.
- How do I find out which service fits my problem?
- You don't need to know in advance. Describe the workflow that costs you time – I place it and propose the simplest solution that reliably covers the process. The five areas on this page are for orientation, not a choice you have to make.
- Can existing Excel files, macros or tools be taken over?
- Yes. Many projects start with a file that has grown over the years and is first stabilised and documented, then extended. Rebuilding only makes sense when the structure itself is the problem.
- Who owns the code at the end?
- You do. You receive all sources, the documentation and a walkthrough so your team or another provider can keep developing the solution. That applies to Excel macros as much as to web applications.
- Do you use AI on sensitive data as well?
- Only with a clear data concept: which data does the model see, where does it run, what is logged? Where necessary, locally hosted models or anonymised data are used. I don't make blanket promises about admissibility; that is clarified for the specific case.
Next step
Which process are you still doing by hand?
Briefly describe what is done today and how. You get an honest assessment of whether and how the process can be automated – and a concrete quote if it makes sense. Some projects take a few hours, others run over several phases. I only quote once I understand the process.