Freya Burrell · West Yorkshire

Hello. I make data look like someone cared.

I'm Freya — Fine Arts graduate, NHS operations by day, enthusiastic gardener and plant parent the rest of the time. I'm not a programmer: I design the thing, then build it with AI doing the typing and me doing the arguing. Turns out that's enough to ship.

The plan is simple: build tools that make my own job easier first, then test them in the real world and see if they help anyone else. Everything below has survived that test — or is still in progress.

tools in daily use by real teams 3,400 formulas I'd never write by hand designed first, built second self-taught · AI-assisted · honest about it
See the work How I got here
absence-tracker.xlsx
Column1Column14abs_dFLAG??
SMITH,J03/02/2641
khan_a4496220
Whitaker#VALUE!71
o'brien M19/03/2610

Same numbers. Same file. One of them you'd read. Invented data · about this

↑ every spreadsheet deserves this. go on, press it

Web apps & dashboardsin daily use by NHS ops teams
100k-row pipelinescleaned, matched and charted
Templates on Etsybought by strangers, which was the test
One plantthat wilts when I forget my salt
Selected work

Things I've made

Every one of these started the same way — something was annoying me and nobody had fixed it. The demos are real and running on this page; go on, poke them.

opening soon

Rooms

web app · in progress · opening soon

Every building books its rooms the same way: a calendar that's wrong, a printed grid by the kettle, and somebody's memory.

Rooms does the obvious part properly — pick a site, a room, a time, and if it clashes it tells you while you're typing and offers the next free slot. But the part I'm actually pleased with is the numbers underneath: utilisation by room, load by hour, and how much capacity sat there bookable and nobody booked it. Most systems tell you what's taken. This one tells you what you're wasting.

Ask me about Rooms →
try booking 12:00, I dare you →
Rooms · todayinvented data
Green is free, grey is taken. Pick one.
free · always will be

Bradford Factor calculator

free · runs entirely in your browser

Somebody asked me what their absence score meant and I couldn't find a single page worth sending them.

The Bradford Factor is spells × spells × days. The number is meaningless until someone tells you that 200 is a conversation and 500 is a process — so the calculator says that part too. Nothing you type is sent anywhere, stored, or remembered. No email box.

Open the full calculator →
Quick scorelive — type in it
162
Medium — worth an informal chat3 × 3 × 18
for sale · on Etsy

The trackers

Excel · Google Sheets · on Etsy

Every team has the same file. Someone made it years ago, one column is called Column14, and nobody dares touch it because nobody knows what breaks.

So I built a good one. About 3,400 formulas doing the arithmetic underneath, so the person using it types dates into boxes and gets an answer. No pivot tables, no training day. It looks like something you'd put in front of a board — because that's usually where it ends up. I put it on Etsy to find out whether it was just me. It wasn't.

See them on Etsy →
Absence rate by monthinvented data
JanFebMarAprMayJunJulAug
3.4%absence rate
11open spells
2scores over 200
in daily use

Ledger

web app · in daily use

Ops teams run on three lists that never agree: projects on the go, ideas nobody's picked up, and the priorities leadership keeps asking about.

Ledger puts them in one place. Projects have an owner and a state. Ideas sit in a pile until someone clicks pick this up and it becomes theirs. The quarterly priorities get their own board with a present mode. It's in daily use — which is the only test of this sort of thing that counts.

Ledger · ideasinvented data · go on then
Nobody's picked this up

Sort out the shared drive

Nobody can find anything and there are four folders called "final".

in the pipeline:
Locum booking & planningoffer spare sessions to the GP team, see what’s already booked, end the email ping-pong — sketching it out now
Built at work

The heavyweight builds

My biggest builds happen at the day job — planning tools an NHS ops team actually runs on. Those stay at work, on work systems, behind work logins. What's below are reconstructions with invented numbers: the ideas travel, the data never does.

in daily use

Same-day capacity planner

web app · multi-site

Every morning, six sites need to know one thing: can we see today's patients? This reads the plan, checks it against what's actually rostered, spots the gaps and the name mismatches, and turns findings into actions — before the phones open, not after.

Today · same-day coverinvented data
Practice 192% Practice 2104% Practice 378% Practice 496% Practice 561% Practice 699%

⚠ Rota check: Dr J. Example has a rota at Practice 3 but is booked on annual leave — please check.

in daily use

Demand & capacity dashboard

data pipeline · run charts

Hundreds of thousands of triage records, cleaned and fed into weekly run charts of demand against capacity, with an ideal line so a board can see the gap in one glance instead of forty tabs. Built so a colleague could open one sheet and just see it.

Weekly demand vs idealinvented data
5,120contacts this week
4,900weekly ideal
+4.5%vs ideal
5.4k 4.9k 4.4k ideal 4,900 5,120 w1w3w5w7w9

Demand has run above the ideal line for 3 of the last 4 weeks — the chart says so before anyone has to.

well established

Workforce capacity planner

workforce model · ~50 staff, five pools

A whole nursing, HCA and phlebotomy team modelled: contracted hours minus breaks, admin, leave and training — against what clinics actually need. Its one big finding changed the conversation: enough hours, not enough trained hands for specific clinics. A hiring problem became a training plan.

Trained cover by clinicinvented data
2,120appts needed / month
1,764provided
356short — all in two clinics
Diabetes reviewson target
Asthma reviewson target
Vascular checks40 of 580 provided
COPD reviews150 of 180 provided

Plenty of staff hours overall — the gap is training, not headcount. That sentence saved a recruitment round.

after hours
Quiet software

Tools I built for my own body and brain

No accounts, no streaks, no cloud. Your data stays on your device, and nothing makes you feel guilty at 11pm. I build these because I need them; they're here because they show what I care about.

Pottable

iOS · in progress

An app for people with POTS. After years of random symptoms and an unexpected diagnosis, I built the thing I needed to navigate it: a pot plant called Basil who wilts when you haven’t had enough water or salt. No score, no streak — just a plant to keep alive, which turns out to be exactly the right amount of pressure.

Basil · perked up — 6 glasses, salt ✓

Gut Log

in progress

Track what you ate, and how it went. I monitor obscure gastro symptoms, and the app uses AI to map triggers, symptoms and patterns — so my clinicians and I get accurate data instead of scribbled notes on the back of an appointment letter.

pattern found: dairy → symptoms within 3h · 78% match · 4 wks of data

The book

Mac app · in daily use

An interactive journal with sealed letters you can’t peek at, a gratitude jar you shake, and the real moon phase. Lives entirely in one file on your machine, and forgets nothing while telling no one.

247 gratitudes in the jar · shake to be handed one back

Story

How I got here

I studied Fine Arts. Then I ended up in operations, which is about as far from a life drawing class as you can get, and found myself staring at spreadsheets that were completely correct and totally unreadable.

That bothered me more than it probably should have. So I started fixing them. First formulas, then scripts, and then I discovered that AI could write the code if I could describe the problem well enough. I still can't really code. I can, however, make computers do what I want — which turns out to be the useful part. The skill isn't the syntax; it's knowing what to build, seeing why something looks wrong, and being too stubborn to stop before it works.

The art degree wasn't a detour. Most data is presented badly, and knowing why something looks wrong is a thing you can't pick up from a formula reference. It's the same eye, pointed at a different canvas.

Furrell is a side project, not a company. Some templates, some free tools, and somewhere to keep the things I make. If one of them saves you an afternoon, that's the whole idea.

Freya Burrell me, probably about to water something
the quick version:
StudiedFine Arts
DoNHS operations
Learned byGetting it wrong first
Build withSheets · scripts · an AI that types
BasedWest Yorkshire
CurrentlyTeaching a plant to wilt
The shop

Templates you open and use

No formulas to write, no training day, no subscription.

Template

Absence tracker

Bradford scores, absence rates and monthly breakdowns, calculated as you type.

Template

Performance dashboards

Turn a month of numbers into something people will actually look at.

Template

Workforce trackers

Who's in, who's out, what it costs — instead of six tabs and a good memory.

See everything on Etsy
About the data on this site

None of it is real, and none of it is yours

The demos

Everyone in them is invented

The names, spells and scores are made up to show what the formatting does. They're not real people and aren't based on real people.

The free tools

Nothing leaves your browser

The Bradford calculator runs entirely on your machine. Whatever you type is never sent to me, never stored, and gone when you close the tab.

The templates

They're just files

You download a spreadsheet and it's yours. It doesn't phone home, doesn't sync anywhere, and doesn't need the internet to work.

My day job

The ideas travel, the data never does

The planning tools in "Built at work" were built at work, for work, and stay there — on work systems, behind work logins. What's shown here are reconstructions with invented sites, invented names and invented numbers. Nothing on this page comes from my employer's data, systems or files. That line matters to me more than the site does.

Got a spreadsheet that's ruining your week?

Tell me about it. I'm nosy about other people's data problems and I might just make the thing.

Say hello