Work repeated every week
People spend hours downloading, copying, reconciling, formatting, and emailing the same information on a schedule.
Data, Automation & Analytics Consulting
Data consulting that automates the work you keep doing by hand.
Independent data consulting for small and mid-sized businesses: cleaning up data you cannot rely on, connecting systems that do not talk, and turning the reports someone rebuilds every week into something that runs itself.
Based in the Minneapolis-St. Paul area, working remotely.
01
Done by hand today
Spreadsheets, exports, files, rekeyed data
02
Cleaned and connected
Validate, model, integrate, warehouse
03
Runs on a schedule
Dashboards, reports, alerts, automation
Common problems
The data exists. Getting a useful answer out of it is the hard part.
Data you don't trust
Information is spread across systems, spreadsheets, exports, databases, and reports that don't always agree.
Systems that don't talk
Important information exists, but getting it from the source system into something useful takes too much manual work.
What usually gets automated
The boring work, specifically.
Most projects start with one of these. None of them need a platform migration or a data team, just someone to look at the process and build the version that runs on its own.
- 01
The report rebuilt every month
Someone exports the same files, pastes them into the same workbook, fixes the same formatting, and emails it. That whole sequence can run on a schedule instead.
- 02
Files carried between two systems
A download here, a cleanup there, an upload somewhere else. If the steps are the same every time, a pipeline can do them and tell you when something looks wrong.
- 03
Data typed in twice
The same information keyed into a second system because the two do not talk. An integration or a small internal tool usually removes the second pass entirely.
- 04
Reconciliation done by eye
Two lists compared in a spreadsheet to find what does not match. That comparison can run automatically and surface only the exceptions worth looking at.
- 05
Month-end assembled from exports
Close that depends on one person remembering the order of a dozen steps. Documented, automated, and repeatable beats institutional memory.
- 06
Numbers nobody quite trusts
Two reports that should agree and do not. Usually a definition problem before it is a technology problem, and worth settling before anything gets automated.
I spent seven years building this kind of thing inside a bank, and I keep building it on my own time. Open the tools and data products I run, or read on for how the work is usually scoped.
Services
Data first. Automation where it helps. Reporting people can use.
The work starts with the business problem, not a list of tools. I can fix one stuck process or build the data foundation behind many of them.
- 01
Data Foundations
Get the data organized first.
I clean up inconsistent data, settle the business definitions everyone argues about, and build a structure that reporting and automation can depend on. This is information management and data governance without the committee.
- Data cleaning and quality
- Warehouses and database design
- Reliable metrics and models
- 02
ETL & Data Integration
Get the right data where it needs to go.
If someone downloads a file from one system, cleans it up, and loads it somewhere else every week, that is a pipeline waiting to be written. I build reliable ETL and data integration between the systems a business already runs.
- ETL and data pipelines
- Files, databases, and APIs
- Scheduled, monitored workflows
- 03
Analytics, Reporting & Dashboards
Make the numbers useful.
I turn reliable data into business intelligence people actually open: dashboards, KPI reporting, and recurring financial and operational reports that show what happened, what needs attention, and where to focus next.
- Executive and department dashboards
- Power BI and KPI reporting
- Financial and recurring reports
- 04
Automation & Busy-Work Elimination
Stop doing the same work over and over.
I look for the bottlenecks, repetitive tasks, and manual handoffs in a process, then automate the parts that do not need a person. The goal is less copying, reconciling, reformatting, and chasing files.
- Reporting automation
- Recurring extracts and file processing
- Workflow alerts and internal tools
- 05
Agentic AI & Intelligent Automation
Use AI where it adds something useful.
I fix the data and the process first, then use AI for the specific work it does well: finding approved information, processing documents, or moving a research workflow along. Generative and agentic AI are tools here, not the pitch.
- Internal knowledge assistants
- Document and research workflows
- Human-in-the-loop AI agents
How I work
A practical approach
- 01
Understand the problem
Figure out where the data lives, what people are doing manually, what isn't reliable, and what the business actually needs.
- 02
Build the useful version
Clean the data, connect the systems, automate the process, or build the reporting needed to solve the problem.
- 03
Make it maintainable
Document the important logic and make sure the finished system is understandable and maintainable.
What you get
Work that is useful after the project ends.
- Cleaner, more reliable data
- A better-organized data foundation
- Repeatable ETL and integrations
- Less recurring manual work
- Professional dashboards
- Consistent business metrics
- Reusable reporting logic
- Documentation and maintainable systems
Professional Work
Systems built inside a real, regulated business.
A sample of the data, analytics, automation, and internal systems work completed over seven years in community banking. Details are generalized to protect confidential business information.
Data Engineering & Integration01
Enterprise Data Warehouse
Designed and built a centralized data environment that brings core banking and other operational systems into modeled, analytics-ready datasets. It replaced disconnected one-off extracts with a shared foundation that reporting, analysis, and automation across the organization are built on.
Analytics & Business Intelligence02
Executive Management Dashboard
Built the reporting behind a daily executive view that consolidates financial, lending, deposit, customer, and operational indicators into one consistent picture. The objective was a decision-support system leadership could rely on each morning, not another report someone had to reconcile first.
Financial & Risk Analytics03
Customer Profitability Model
Developed a customer-level profitability framework combining interest earnings, funding costs, fee income, and estimated servicing expense into a consistent view of relationship value. The calculations and reporting were then automated, so profitability became an ongoing capability rather than a periodic analysis project.
Regulatory & Reporting Automation04
Finance & Management Reporting Automation
Automated recurring financial and management reporting covering general-ledger trends, income and expense analysis, funding, fee activity, and board-level packages. Work that had been gathered, calculated, formatted, and distributed by hand each cycle now runs on a schedule.
Start a conversation
Have a data problem that should be easier?
Whether you’re rebuilding the same report every week, working around messy data, or trying to get multiple systems to work together, let’s talk through it.