On the first Saturday of last quarter I sat down with my bank statement and a mug of barley tea and counted the recurring charges. There were nineteen. Two streaming services, two cloud-storage tiers, a meditation app, a meal-kit, a Pilates studio, a digital newspaper, two SaaS tools I used for work, a music subscription, an e-book service, a wine club, a magazine, a domain renewal, a productivity app I had forgotten subscribing to, a virtual phone line, and a backup tier for a backup. The total came to ¥38,420 per month. Annualised, ¥461,040.
I had not noticed the totals before. This is, I think, the central feature of the modern subscription economy: each charge is small enough to slip beneath the threshold of attention, and the system is designed for the slipping. The mottainai of subscriptions is not the cost. It is the way the cost has been engineered to be invisible.
This essay is about the quarterly audit I now run on my recurring charges, the four questions I ask of each one, and the small template I use to track the answers. It is not a teardown of the subscription economy. There are subscriptions I keep happily. The audit is about distinguishing them from the others.
Why quarterly, and why not monthly
I tried, for a few months, to audit subscriptions monthly. It did not work. The monthly cadence is too short to give the data shape; a service used twice in May and not at all in June might still be earning its place across the season.
Quarterly is closer to the right rhythm. Three months is long enough that genuine usage patterns emerge — the streaming service watched only in winter, the gym used only in spring — and short enough that you do not pay for a year of something you stopped opening in February. The four audits a year fall naturally on the first weekends of January, April, July, and October. I block ninety minutes for each.
If you have never audited subscriptions, expect the first one to take longer. The list is almost always longer than you remember.
The four questions
For each recurring charge, I ask, in order:
- Did I open this in the last ninety days? Not did I value it. Did I, in fact, open or use it. A binary question, answered honestly from memory or app history.
- If I cancelled today, what would I lose? Specific, not abstract. Two saved playlists. The 4K stream of one show I am halfway through. Nothing.
- Is there a free or cheaper version that does eighty per cent of this? Most categories now have one. The question is not whether you would prefer the cheaper version aesthetically; it is whether the cheaper version would, in practice, leave you no worse off.
- Would I sign up for this today, at this price, if I were not already subscribed? The strongest of the four questions. Most people protect existing subscriptions in a way they would not protect new ones, because cancellation feels like a loss. The question reframes the loss: every month you keep something you would not now buy is a small mottainai of the same shape as the unused gym pass.
Three "no" answers across the four questions is, in my practice, an automatic cancellation. Two "no" answers is a flag for the next quarter; if it is still flagged in three months, it goes. One "no" is normal and not a problem.
What the first audit revealed
Of my nineteen subscriptions, my first quarterly audit cancelled six. They were:
- The second streaming service. Not opened in eighty-three days. Two shows I had been meaning to watch were also available on the first service, with a small wait. ¥1,490/month, ¥17,880/year.
- The meditation app. Used twice in ninety days, both in a single week of high stress. A free alternative would have served identically. ¥980/month, ¥11,760/year.
- The wine club. The wines were good. The volume was wrong; bottles were accumulating. ¥4,500/month, ¥54,000/year.
- The productivity app I had forgotten subscribing to. Question one answered itself. ¥820/month, ¥9,840/year.
- The backup tier for the backup. Genuinely redundant once I traced the architecture. ¥600/month, ¥7,200/year.
- The virtual phone line, kept for a project that had ended. ¥1,200/month, ¥14,400/year.
The cancellations recovered ¥9,590 per month — ¥115,080 per year. The audit took ninety minutes. I have done four more audits since; subsequent ones recover less, because the obvious mottainai was caught in the first pass, but each one has produced one or two cancellations.
The tracker template
For the first year I ran the audit on a sheet of paper. This was fine, but the year-over-year comparisons were difficult, and I kept losing the previous quarter's notes. I now keep a simple Notion table with one row per subscription and a column for each quarter, marked with a usage tally and an audit verdict. The table has eight columns: name, monthly cost, annualised cost, category, last opened, kept/cancelled, the four-question verdict, and notes.
If you would like to start with a ready-made template rather than build one, the subscription tracker I use is published on Notion as part of a small set of household templates I maintain. It is free on Notion's free plan. The structure is intentionally minimal — there are no charts, no automation, nothing to maintain. The point is the discipline of writing the answers, not the complexity of the tool.
For people who prefer paper, the Mindful Yen Etsy shop has a printable quarterly audit sheet that runs on the same four questions. I sometimes use both — Notion for the running record, paper for the actual audit ninety minutes — because writing by hand still produces the slower thinking that catches the questions you would otherwise skip.
The categories that always need scrutiny
After several quarters of audits, certain categories have proven, in my household, to be persistent sources of subscription mottainai. They may be different in yours, but I will name mine.
Streaming services. The economics are designed to make adding cheap and removing emotionally expensive. A second service is almost always mottainai unless both are watched weekly. A third service is mottainai unless one is for a specific household member with distinct taste.
Productivity and wellness apps. The category most prone to signed up after a recommendation, used twice, never returned to. A subscription whose purpose is to change you is unusually good at convincing you that cancellation is giving up on yourself. It is not. It is admitting that the app is not the lever.
Cloud storage. Easy to over-purchase because the upgrade was prompted by a single full-storage notification. After three months, the actual usage is usually much lower than the tier you panicked into.
Gym, studio, or class memberships. The hardest to audit honestly because the unused pass is also a quiet self-judgment. The question is not am I a person who exercises. It is did I, in fact, attend. If the answer is two visits in ninety days, a pay-per-visit pass is almost always the kinder structure.
Renewals you do not see. Domain names, virtual phone lines, third-party email forwarders, subscription gifts that converted to paid. These are the most dangerous because they renew silently. A quarterly audit is the only structural defence I have found.
What the audit does over a year
Across four quarters, the audit has done three things in my household.
First, it has reduced recurring spend by roughly ¥14,000 per month, or ¥168,000 per year, with no perceptible drop in what subscriptions I value. The savings flow into the same place every kakeibo saves into: the slow, uncelebrated buffer that makes other decisions easier.
Second, it has changed how I sign up for new subscriptions. Knowing the audit is coming makes the moment of subscription a different kind of decision. I now ask, at the point of subscription, whether I would still be subscribed in ninety days. The answer is more often no than I expected, and the trial is more often left to lapse.
Third, and most usefully, it has surfaced a small number of subscriptions I had been quietly underusing because I had forgotten what they offered. The newspaper subscription, in particular, I had been opening once a fortnight; once I noticed, I started opening it daily, and it became one of the two subscriptions I value most. Mottainai works in both directions. Sometimes it tells you the object's mottai is being violated by neglect rather than excess.
The shape of the practice
If you adopt the quarterly audit, the shape is similar to other slow practices I have written about. The first audit will produce the largest cancellation list. The second will produce a smaller, more thoughtful one. By the fourth, the audit has become routine — ninety minutes, four times a year, with quiet but compounding effects on both the budget and the relationship between you and what you have agreed to pay for.
The mottainai framework I described in last week's essay on the monthly budget handles the small daily noticings. The quarterly subscription audit handles the larger structural ones. Together they cover most of the modern household's quiet leakage.
If you want to read more in this register — slow money, slow audits, slow noticings — most of my longer writing on these themes lives on the Mindful Yen Substack, where I publish a quarterly subscription audit of my own each season as a worked example.