Build the list from twelve months of bank and card statements rather than from memory, since the subscriptions costing you most are the ones nobody would think to list.

The list you cannot write from memory

Reviewing subscriptions is straightforward advice, and it assumes something that is not true: that you can produce a complete list.

You can produce the ones you use, which are the ones that are earning their money.

The expensive ones are the opposite: a tool nobody has opened in a year, a service billed annually that nobody sees eleven months out of twelve, a per-seat licence for somebody who left, and a trial that converted and has been charging ever since.

None of those appears on a list written from memory, which is why the review has to start somewhere else.

Start from the statement

The method, and it is the whole difference.

Take twelve months of bank and card statements and go through every line, marking anything recurring.

Twelve months rather than three, because annual charges are the ones that hide, and they are frequently the largest.

Do the business account and any personal card that has ever been used for business, which is where a surprising number of these live.

It is a tedious hour and it produces a genuinely complete list, which nothing else does.

Where they hide

The second is worth expanding: app store charges appear as a single vendor line, so several subscriptions can be concealed inside what looks like one small payment.

Check the seat count

The item that produces the largest single savings and is almost never examined.

Software billed per user grows as people are added and does not shrink when they leave, because removing a seat requires somebody to remember.

Open each per-seat tool and count the licences against the people who actually work here.

Businesses regularly find they are paying for two or three departed staff on several tools simultaneously, which compounds into a meaningful annual figure.

This is also a security check rather than only a cost one, since a paid seat frequently means an active account.

A worked example

A firm listed the subscriptions they knew about and counted eleven.

Going through twelve months of statements found nineteen.

The eight they had missed included two annual charges, three inside a single app store line, one on the owner's personal card from a trial three years earlier, one for a service replaced eighteen months ago and never cancelled, and one paying for two departed employees.

They cancelled six, reduced the seats on two, and kept the rest.

The annual saving was several times what they had expected, and roughly matched the cost of a piece of equipment they had been deferring.

The owner's observation was that they had reviewed subscriptions twice before, both times from memory, and had found nothing.

The three questions per line

Once the list exists, the decisions are quick.

When did somebody last open it, which is usually answerable and usually decisive.

What would break if it stopped, which distinguishes a tool from a habit.

And is there a cheaper tier that fits what you actually use, since most businesses are on a plan chosen when they expected more growth than arrived.

Downgrading is frequently better than cancelling, and it is the option people forget exists.

Cancel properly

The step where savings evaporate.

Cancel through the service rather than by stopping the card, which produces debt collection rather than a cancellation.

Export anything you need first, because access usually ends immediately or at the end of the period, and data is frequently deleted afterwards.

Get confirmation in writing and diarise the renewal date to check the charge has actually stopped.

And watch for a charge the following month regardless, since cancellations do fail and a subscription you believe is cancelled is worse than one you know about.

Prevent the next eight

The part that stops this being an annual archaeology exercise.

One card for all subscriptions, so they appear together and can be reviewed in one place.

A single note recording what each is, who owns it, what it costs, and when it renews, updated when anything is added.

Adding removal of licences to whatever you already do when somebody leaves.

And a rule that trials are diarised on the day they start, with a reminder two days before they convert.

None of that is difficult and all of it fails without one person being responsible for the note.

Count what it adds up to

Worth doing explicitly, because individual amounts are small and the total is not.

Add every recurring charge you found and express it as an annual figure rather than a monthly one.

Nine services at modest monthly amounts is a number most owners have never seen stated, and seeing it changes how the individual decisions feel.

Compare it against something concrete: a piece of equipment, a month of advertising, or a week of somebody's time.

That comparison is what turns a tedious audit into a decision, and it is why the total belongs at the top of the list rather than in a column nobody sums.

The counter-case

Cancelling can cost more than it saves.

Some services are cheap and quietly load-bearing: a backup tool, a monitoring service, or a domain-related subscription that nobody thinks about precisely because it works.

Cancelling something because nobody has logged in is a poor test for anything running automatically, and that is exactly the category worth keeping.

There is also a false economy in cancelling a tool and then spending hours doing manually what it did, which happens most with accounting and scheduling software.

Cancel what nobody uses. Keep what runs quietly. Ask what it does before assuming an unopened tool is idle.

The hour

  1. Take twelve months of statements, not three.
  2. Include any personal card used for business.
  3. Expand app store lines into individual services.
  4. Count seats against people who actually work here.
  5. Ask when each was last opened.
  6. Downgrade before cancelling.
  7. Cancel through the service and confirm next month.

Step one is the only thing that produces a complete list, and every review done from memory finds the subscriptions that were never the problem.

Reviewing the renewals you know about is covered in renewals that all land in January.


Frequently asked questions

Why can I not just list my subscriptions?

Because the ones you remember are the ones you use, which are earning their money. The expensive ones are the ones nobody would think to list.

Where should the list come from?

Twelve months of bank and card statements, marking anything recurring. Twelve rather than three, because annual charges hide and are frequently the largest.

Where do subscriptions hide?

Annual charges, app store billing aggregated into one line, a personal card used for business, per-seat licences for departed staff, converted trials, and a former employee's account.

What produces the biggest saving?

Seat counts. Per-user software grows when people are added and does not shrink when they leave, so businesses often pay for several departed staff across multiple tools.

Should I cancel or downgrade?

Check for a cheaper tier first. Most businesses are on a plan chosen when they expected more growth than arrived, and downgrading is the option people forget.

What should I not cancel?

Anything running quietly: backups, monitoring, domain-related services. Nobody logging in is a poor test for something that works automatically.

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.

Reviewed subscriptions from memory before?

Do it from twelve months of statements instead. Most businesses find roughly twice what they listed.

Start a Conversation