I find the work your team shouldn't be doing. Then I automate it.

The fundraiser faced the same problem again and again: finding the right grant funders, then filling out long forms that ask for the same charity information every time. Each form took a full working day of researching which funders would say yes, changing project descriptions to fit different questions, and typing in the same basic details over and over.

I built a system that stores the charity's key information and uses examples from past forms to keep the same writing style. When looking for funding, the fundraiser tells it what the project needs, and the system finds good matches and creates draft forms that fit each funder's questions while keeping the charity's usual voice.

Forms that took eight hours now need ten minutes, so six forms per year saves six working days that now go to building relationships with funders.

The editor couldn't answer basic questions about book status because five different spreadsheets held the information: editing progress in one, proofreading stages in another, indexing somewhere else. Finding out if a book would meet its print deadline meant searching multiple files and asking several people.

I built a single tracker that shows every book's stage, who's in charge, and the deadline in one place. A traffic light system flags books at risk, and percentage bars show if work is on target. Staff update it by hand as work finishes, which keeps things simple while showing current progress.

The editor used to spend two hours each week chasing status information across separate systems, and that time is gone. The system didn't prevent real failures since books already met their print deadlines, but it turned chaos into clarity so the editor can now focus on editorial decisions instead of detective work.

Management needed clear dashboards to make stock and pricing decisions, but the sales data made this impossible. The spreadsheet had 43,000 formulas spread across multiple sheets, with copied data and calculations hidden in beige and green cells. Staff ignored most data because it took too much effort to understand.

I rebuilt the entire database by changing the layout from horizontal to vertical and removing the formula layers. All calculations now happen in Looker Studio instead of the spreadsheet, so staff now use clickable filters to find specific data instantly instead of copying and pasting through complex sheets.

The finance team used to spend two hours each week finding their way through the mess, and that's gone. More importantly, non-technical staff can now pull useful insights in minutes because the data hasn't changed but how easy it is to use has, which turned an unusable file into a working decision tool.

Two staff spent an hour each week pulling sales data from tills and reformatting it for the main tracking spreadsheet - one managed fourteen shops, the other seven. The till system didn't connect with their data setup, which tracked sales alongside Gift Aid and other numbers.

I built automation in Google Sheets using scripts to process the raw exports, so the system pulls data from the till network and fills the main spreadsheet automatically. For shops in Northern Ireland this is now fully automatic, though two Scottish locations still need manual entry due to different till systems, taking twenty minutes each week.

The Northern Ireland manager saves an hour each week and the Scotland manager saves forty minutes, which is 100 minutes per week and recovers eleven working days each year.

Editorial staff check biblical quotes against multiple translations, verifying accuracy and formatting. The automation cross-references verses, flags translation differences, and ensures the house style is correct.

Seven staff turn credit card statements and receipt images into sorted expense entries. The system learns repeating patterns, processes statements, and pulls receipt data automatically.