Skip to content

Automation

Clean Taggy

Clean Taggy turns the repetitive, error-prone part of preparing manufacturing label and wire-tag data into a structured workflow. It runs from a link, works offline on a single device, and syncs each person's projects to the cloud once they sign in.

In production2025JavaScriptFirebaseFirestoreSingle-file app
clean-taggyPROJECTSLine 4 — Rev CLine 2 — Rev ABay 7 retrofitArchive 2024Synced · 4 devicesLine 4 — Rev CVALIDATEDExportTAGWIREFROMTOSTATUSTAG-0142W-2210RACK-ADSP-01OKTAG-0143W-2211RACK-AAMP-02OKTAG-0144W-2212RACK-BSPK-07CHECKTAG-0145W-2213RACK-BSPK-08OKTAG-0146W-2214PATCH-1CTRL-03OK

01 — The problem

What was actually happening

A quality control team was assembling label and wire-tag data by hand — reading values off drawings and exports, retyping them into spreadsheets, and reconciling versions between people. The work was slow, and every manual transcription was a chance to introduce a defect that would only surface on the floor.

  • Tag data was rebuilt by hand for every job, with no shared source of truth between team members.
  • Transcription errors were invisible until late in the process, when they are expensive to fix.
  • IT constraints ruled out installing desktop software or standing up a server, so most off-the-shelf tools were unusable.
  • Anything adopted had to work for non-technical users on their first attempt, with no training session.

02 — The solution

What I built, and why

Instead of fighting the environment, the constraint became the design: the entire application ships as a single HTML file. It can be dropped into SharePoint, published on GitHub Pages, or pinned as a Microsoft Teams tab — no install, no server, no admin approval.

  • Structured project workspace that replaces the ad-hoc spreadsheet with a consistent, validated data shape.
  • Local-first by default — the app is fully usable before anyone signs in, so adoption never blocks on IT.
  • Optional Google or email sign-in adds cloud sync, so each person's projects follow them across devices.
  • Firestore security rules scope every read and write to the signed-in user's own document.
  • Light and dark themes that follow the operating system, because the tool is used on a shop floor and at a desk.

03 — Process

How the work was sequenced

  1. 01

    Understand the workflow

    Sat with the actual process first — what gets read, what gets retyped, where the mistakes happen — before deciding what software could remove.

  2. 02

    Map the constraints

    Established the real boundaries early: no installs, no server, corporate SSO, and users who would judge the tool in the first thirty seconds.

  3. 03

    Design for zero friction

    Chose a single-file architecture so the app could be shared as a link, and made sign-in optional so the first run needed no setup at all.

  4. 04

    Build

    Implemented the workspace, the data model and the theming in vanilla HTML, CSS and JavaScript — no build step, so the file that is tested is the file that ships.

  5. 05

    Add sync

    Layered Firebase Authentication and Cloud Firestore on top of the working local app, with rules that isolate each user's data.

  6. 06

    Deploy and iterate

    Documented three deployment routes (GitHub Pages, SharePoint, Teams tab) so the team could pick whichever their IT policy allowed, then refined against real use.

04 — Technology

The stack, as decisions

Application

HTMLCSSJavaScript

Identity & data

Firebase AuthCloud FirestoreSecurity rules

Delivery

GitHub PagesSharePointMicrosoft Teams app manifest

05 — Results

What it changed

1 file

The entire application — no build step, no dependencies to break

0 installs

Deployed as a link, inside a locked-down corporate environment

Per-user sync

Projects follow each person across devices once signed in

The hardest constraint turned out to be organisational, not technical. Designing around what IT would actually approve — rather than what would be technically ideal — is what got the tool into daily use. It also proved how far a disciplined single-file architecture can go before it needs a framework.
What it taught

Have a problem worth solving?

Tell me what the process looks like today and where it breaks. If technology is the right answer, I will tell you what I would build — and if it is not, I will tell you that too.