How to Automate Google Drive

Drive is where documents go to be found later

Google Drive is the storage layer of Google Workspace, and its real job is not storing things. It is answering the question somebody asks nine months later: where is that document.

Every business accumulates paperwork it did not create. Supplier invoices. Bank statements. Signed contracts. Delivery notes. Certificates and inspection reports. Council correspondence. None of it is written in Drive. All of it needs to be in Drive when an accountant, an auditor or an angry customer asks.

Drive is good at the finding, if the filing was done. Search covers document contents, sharing is granular, and shared drives mean a file survives the person who saved it leaving the company.

Everything depends on the file having arrived, been named sensibly, and been put somewhere.

Filing is the tax on every finished job

The document almost never starts in Drive. It starts behind a login somewhere else.

The statement is in the bank's portal, available as a PDF once a month. The invoice is in the supplier's account area. The signed contract is in the e-signature tool. The compliance certificate is on a testing company's website. Each one has to be logged into, found, downloaded, renamed from something like statement_final_3.pdf, and dragged into the right folder.

It is nobody's actual job, so it gets batched, and batching is where it fails. Twenty statements downloaded in one sitting at the end of a quarter, half of them named wrong, two of them missed entirely because the portal only keeps three months.

The cost is invisible until the moment somebody needs a specific document quickly, which is always a moment that already has some pressure attached to it.

Google Drive

Statements collected on time The monthly documents downloaded from each portal that holds them, before somebody needs them for a deadline.

Named the same way every time One naming rule applied to every file, so search works and nobody opens six documents to find the right one.

Filed where it belongs Straight into the client, month or job folder, instead of a downloads pile somebody sorts out later.

Drive organises files, it does not go and fetch them

Drive is more automatable than most people realise, and Workspace adds a fair amount on top.

Sharing can be set by folder and inherited. Shortcuts put one file in several places without copies. Version history means nothing is really lost. Apps Script runs on a timer against your Drive, so moving, renaming and tidying can genuinely be scripted by anyone willing to write a little code. Shared drives keep ownership with the company rather than an individual.

All of that operates on files that are already in Drive. The scripting reaches the folder structure, not the outside world.

The bottleneck is upstream of all of it. A script can file a statement beautifully once the statement is in Drive. Nothing in Workspace can log into your bank, open the statements section, and download last month's PDF.

The folder can fill and file itself

A login, a click and a download: that is the whole of collecting a statement. A person is only involved in it out of habit.

Every monthly statement pulled from each bank and card portal on the day it appears, named to one convention, dropped in the right year and month folder. Supplier invoices collected from the accounts area of each supplier you buy from. Signed documents pulled out of the e-signature tool and filed against the client they belong to. Certificates collected from the testing company before they expire.

The naming is worth as much as the collecting. One convention, applied every time, is the difference between search working and a folder of files called document.pdf.

It is worth pointing the collecting at a staging folder that only the automation writes to, then filing from there. If a portal changes and a run brings back the wrong thing, it is sitting in one place rather than mixed into folders your accountant is about to read.

Somebody still has to log into the portal

The filing rules can be written once. Logging into eleven different portals every month is the part that keeps costing a person an afternoon.

WebRun is an AI agent that works a real Chrome browser, signed in as you. It opens each portal, finds the document, downloads it, names it to your convention and puts it in the folder where it belongs.

It runs on your schedule in your own private environment, and you can watch a run and stop it. Sessions stay in your own environment rather than being shared with anything else.

The workflows below are already built, and each one names the sites it opens.