Reports, data & databases

Numbers you don't have to double-check.

If a report takes longer to prepare than to read, or someone checks the totals by eye because nobody trusts them, the problem is usually how the data is collected, not the people. I build reports, imports and reconciliations where the process checks the numbers, so people can spend their time on what the numbers mean.

What I do

Automate recurring reports

Scheduled reports that pull from every source, apply the same rules every time and arrive as Excel files, PDFs or emails.

Make imports safe

Imports that check every row, hold back what doesn't fit with a clear reason, and never load the same data twice.

Reconcile automatically

Matching rules that compare systems or periods and list only the differences that need a person.

Clean up and migrate

One-off cleanups and migrations, tested on real data before anything is switched over.

Tidy the database

Document what is there, speed up slow queries, and give reports a stable set of views to read from.

  • SQL Server
  • T-SQL
  • C# / .NET
  • Excel
  • CSV / XML
  • Scheduled jobs

Excel, a database, or both?

SituationWhat usually makes sense
A few people, one report, rules that change oftenKeep it in Excel, but load the data automatically
Several sources combined the same way every timeA scheduled job that builds the report
The spreadsheet has become the databaseMove the data to SQL Server and keep Excel for reading and analysis
Totals must match between systemsAutomatic reconciliation with a list of exceptions
Managers want to explore the data themselvesClean views or tables that your reporting tool can read

Often the fix is not a new tool but a reliable way to get clean data into the one you already use.

How a data project usually goes

  1. Trace the numbers

    Where each figure comes from and what happens to it on the way.

  2. Agree the rules

    What counts as correct, what is an exception, and who decides.

  3. Agree the scope

    A written proposal: what changes, what it costs, how we check it.

  4. Build and compare

    The new process runs next to the old one until the numbers match.

  5. Hand over

    The working process, the rules written down, and a way to see that it ran.

Questions about this

Do you build dashboards?

My focus is the data underneath: collecting it, checking it and delivering it on time, usually as Excel files, PDFs or database views. If you use a reporting tool, mention it and I'll tell you honestly how I can help.

Which databases do you work with?

Mainly Microsoft SQL Server. If your data lives somewhere else, describe it and I'll tell you whether I'm the right person.

Our data is sensitive. How does that work?

We agree how data is accessed before any is shared. I can often start with the structure and a small sample with dummy values. If you need an NDA first, we can sign one.

Can you work with a database our software vendor set up?

Reading from it is usually fine. Writing to it depends on what the vendor supports. I check this before proposing anything, so the vendor's software keeps working as it should.

Which report takes longer to build than to read?

Tell me what goes into it, where the data comes from and how it's put together today.

Describe your problem