Ask people what they spend on subscriptions and they say about $86 a month.
Measure it and the figure comes out near $219 (LowerMySubs, 2026).
That is a 2.5x gap between belief and reality.
Not a rounding error. Not a few forgotten apps.
Most people are wrong about their own spending by more than double.
This article explains why, and gives you the arithmetic to find your own number. 📊
🎁 Trial Before You Add Another Bill →
🧾 Ĉefaj Trovoj Unuavide
| Mezuro | Figuro | Fonto |
|---|---|---|
| What people estimate they spend | About $86/month | LowerMySubs (2026) |
| What they actually spend | About $219/month | LowerMySubs (2026) |
| Higher estimate | $273/month | West Monroe (2026) |
| Lower estimate | $84/month on 4.5 services | Trackery (2026) |
| Annual average | $1,887 | Resubs (2026) |
| Active subscriptions per person | 3.4 to 12+ | Varies by source |
| Wasted on unused services | $26.79/month | Fortunly (2026) |
🕳️ The Perception Gap
The 2.5x gap is the most important number here, and it is worth sitting with.
📊 What people think they spend, versus what they do
That gap is $133 a month. Roughly $1,600 a year that people do not know they are spending.
Why the gap exists
Three mechanisms, all mundane.
Auto-renewal removes the moment of paying. No decision means no memory.
Small amounts do not register individually. Nine dollars feels like nothing.
And subscriptions arrive one at a time, months apart, so you never see the total form.
The counting problem
When asked, people recall the big ones.
Streaming, phone, maybe a gym.
They forget the $4 storage upgrade and the $7 app renewed annually.
You cannot recall what you never consciously decided.
🔢 Why the Published Figures Disagree So Much
Look at the range across sources and it is genuinely wide.
| Fonto | Monthly figure | What it likely counts |
|---|---|---|
| Trackery (2026) | $84 on 4.5 services | Digital media only |
| Various surveys | $90–$118 | Self-reported |
| Resubs (2026) | About $157 | From $1,887 annual |
| LowerMySubs (2026) | $219 on 8.2 services | Measured, broad |
| West Monroe (2026) | $273 | Broadest definition |
A three-fold spread for the same question.
The definitional problem
What counts as a subscription?
Streaming, obviously. But what about your phone contract? Insurance? Gym? Cloud storage?
Narrow definitions produce $84. Broad ones produce $273.
Neither is wrong. They are answering different questions.
Self-reported versus measured
This is the bigger distinction, and it explains the pattern.
Surveys that ask people produce low numbers, because of the perception gap.
Studies that read actual transactions produce high ones.
Where a source sits in that range tells you its method.
Which figure to use
For your own planning, none of them.
Use your own statement, because the variation between people is larger than the variation between studies.
📈 How Creep Actually Happens
Nobody signs up for twelve subscriptions in one afternoon.
It accumulates through individually sensible decisions.
📊 How a subscription bill builds over three years
The four ways it grows
| Mekanismo | How it feels at the time |
|---|---|
| New service added | “It is only $9” |
| Price rise on existing service | An email you skimmed |
| Tier upgrade for one feature | Temporary, you told yourself |
| Trial that converted | You meant to cancel |
Only the first feels like a decision.
The other three happen without you choosing anything.
Price rises are the quiet one
Services raise prices gradually and announce it in routine email.
A $10 service at 8% a year becomes $12.60 in three years.
Across twelve services that is real money, and you approved none of it.
🔎 Compare Trials Before Committing →
💰 The Compounding Nobody Calculates
Monthly framing is what makes small subscriptions feel harmless.
Annual framing changes the conversation entirely.
| Ĉiumonate | Per year | Over 5 years |
|---|---|---|
| $5 | $60 | $300 |
| $12 | $144 | $720 |
| $30 | $360 | $1,800 |
| $86 (estimated) | $1,032 | $5,160 |
| $219 (actual) | $2,628 | $13,140 |
That bottom row is the honest version of the picture.
Over five years, the average subscription bill exceeds $13,000.
The waste inside that
Consumers waste about $26.79 a month on unused services (Fortunly, 2026).
That is $321 a year, or roughly $1,600 over five years.
Paid for nothing, by definition.
Why annual framing works
“$9 a month” invites a yes. “$108 a year” invites a question.
Same money. Very different decision.
Convert everything to annual before deciding, and half the decisions change.
🏷️ Where the Money Actually Goes
Subscription spending is not spread evenly across categories.
A few take most of it, and the rest hide in the long tail.
🥧 A typical monthly subscription bill by category
The long tail is where the waste sits
The big categories are visible and mostly deliberate.
You know you pay for streaming. You chose your phone plan.
The long tail is a dozen small charges nobody remembers agreeing to.
That is where most of the $26.79 monthly waste lives (Fortunly, 2026).
Why the tail grows fastest
Small charges never trigger a review.
Nobody sets a reminder about $4 a month.
So they accumulate indefinitely, and each new one is equally unremarkable.
Attack the tail first
Counterintuitively, start cancelling at the bottom of the list.
The small ones are the least missed and the most numerous.
Cutting eight charges of $6 saves more than agonising over one $30 service you actually use.
📋 The Audit, in Thirty Minutes
You do not need an app. You need thirty minutes and three sources.
| Paŝo | Where to look | Kial gravas |
|---|---|---|
| 1 | 12 months of card statements | Catches annual billing |
| 2 | PayPal recurring payments | Often missed entirely |
| 3 | Phone app-store subscriptions | Never itemised on cards |
| 4 | Write each with annual cost | Defeats monthly blindness |
| 5 | Mark: daily, weekly, rarely, never | Factual, not subjective |
| 6 | Cancel everything marked never | The whole point |
Step three catches more forgotten spending than any other.
App-store subscriptions appear as one lump from Apple or Google, not by name.
Why twelve months, not three
Annual subscriptions only appear once.
A three-month window misses three quarters of them.
This is the single most common audit mistake.
Rate by frequency, not by value
“Is this worth it?” invites justification.
“When did I last use it?” has a factual answer.
Facts are far harder to argue with than judgements.
Do it with someone else if you can
Household subscriptions are often duplicated across two people.
Auditing separately misses exactly the overlaps worth finding.
Twenty minutes together usually beats an hour apart.
What to expect
Most people find two or three charges they cannot identify at all.
Those are the easy wins, and they are usually the oldest.
👥 Sharing, Bundling and the Duplicate Problem
Three easy savings sit in most subscription lists, and all are commonly missed.
| Opportunity | Typical saving | Effort |
|---|---|---|
| Family or household plans | Often 50–70% per person | Malalta |
| Bundles you already qualify for | Varies, sometimes large | Malalta |
| Duplicate services | Full cost of one | Low once spotted |
| Annual instead of monthly | Typically 15–20% | Malalta |
| Student or age discounts | Often substantial | Meza |
The duplicate row is the one people are most surprised by.
How duplicates happen
Two people in a household subscribe separately to the same service.
Or you pay for cloud storage from three providers, each holding different files.
Nobody planned it. It emerged one decision at a time.
The bundle you already have
Many phone contracts, bank accounts and shopping memberships include services people also pay for separately.
Streaming, storage and music are the common ones.
Check what your existing contracts include before renewing anything.
Annual billing, with one caution
Paying yearly usually saves 15–20%.
But it only saves money if you would genuinely have kept the service all year.
An annual plan on a service you abandon in month three costs more, not less.
Switch to annual only after a service has survived a full quarterly review.
🧠 Why the Audit Does Not Stick
People audit once, save money, and creep straight back.
Within a year the bill is where it was.
That is because the audit fixes the symptom, not the mechanism.
The mechanism is the default
Auto-renewal makes “keep paying” the path of least resistance.
Unless something changes that, creep resumes immediately.
Nia analysis of why people do not cancel covers the psychology in detail.
Three habits that hold
One card for everything recurring. Makes the total visible in one place.
A recurring calendar review. Quarterly, fifteen minutes.
Annual figures only. Never let yourself think in monthly terms.
Why one card matters most
Spending spread across three cards and PayPal is invisible by design.
Consolidating it does not save money directly.
It makes the number impossible to avoid seeing, which does.
📆 Timing Your Cancellations Properly
When you cancel matters almost as much as whether you do.
Cancel at the wrong moment and you lose money you had already paid for.
| Situacio | Best time to cancel |
|---|---|
| Monata plano | Just before renewal date |
| Jara plano | Set reminder 4 weeks before renewal |
| Senpaga provo | Two days before it ends |
| Just renewed by mistake | Ask immediately — refunds are common |
| Mid-cycle | Usually keeps access until period ends |
The fourth row is worth knowing about.
Many services refund an accidental renewal if you ask within a few days.
They rarely advertise it, and most people never ask.
Cancelling does not always mean losing access immediately
For most services, cancelling stops the next payment rather than ending access now.
You usually keep it until the paid period runs out.
That removes the main reason people delay: fear of losing something they already paid for.
The four-week annual reminder
Annual renewals are the expensive ones to miss.
Four weeks gives time to evaluate, export data, and decide without pressure.
A reminder on the renewal day itself is already too late to think clearly.
Write the renewal date down
Most people cannot name a single renewal date from memory.
Add each to your calendar during the audit, while you have the statements open.
Five minutes then saves the whole problem recurring.
🎯 Deciding What Actually Earns Its Place
Cancelling everything is not the goal. Some subscriptions are excellent value.
| Testo | Keep if |
|---|---|
| Frequency | Used weekly or more |
| Replacement cost | Buying outright costs more |
| Time saved | Saves more than it costs in hours |
| Risk covered | Backup, security, insurance |
| Alternative exists free | Then probably cancel |
The time-saved test is the one most people skip.
A $20 tool saving two hours a month is excellent value at almost any wage.
The free-alternative check
For many categories, a capable free tier now exists.
Storage, note-taking, basic design, video calls.
Check before renewing, because free tiers have improved faster than most people noticed.
When paying once beats paying monthly
Some tools are sold outright or as lifetime deals.
If you will use something for years, the arithmetic often favours buying.
Nia lifetime versus subscription analysis works through when that holds.
🧮 What the Money Could Do Instead
Opportunity cost makes the numbers feel real in a way percentages do not.
The measured average is $219 a month, or $2,628 a year (LowerMySubs, 2026).
Suppose an audit cuts it by a third, which is a common result.
| Saved per month | Per year | What that equals |
|---|---|---|
| $27 | $324 | The waste figure alone |
| $50 | $600 | A short holiday |
| $73 | $876 | A third of the average bill |
| $100 | $1,200 | A month of rent, in many places |
None of that requires earning more.
It is money already leaving your account, redirected to something you chose.
Why this framing helps
“Cancel some subscriptions” is a chore with no reward attached.
“Fund a holiday by cancelling things you never open” is a different proposition.
Same action, different motivation, and motivation is what actually gets it done.
Make the saving visible
Move the saved amount somewhere you can see it.
A separate account, or a note of the running total.
Otherwise it dissolves into general spending and the effort feels unrewarded.
The compounding version
If you saved $73 a month and left it alone, that is $876 a year.
Over five years that is well over $4,000 before any interest.
Subscription creep compounds against you, so reversing it compounds in your favour.
🔬 How Reliable Are These Figures?
| Temo | Effect on the number |
|---|---|
| Definition of “subscription” | Largest single factor |
| Self-reported vs measured | Self-reported runs far lower |
| Mostly US samples | Limited elsewhere |
| Published by subscription trackers | They sell the solution |
| Averaĝoj kaŝas grandegan variadon | Your number may differ wildly |
That fourth row deserves stating plainly.
Much of this research is published by companies selling subscription-management apps.
That does not make it false. It does mean the alarming end of the range should be treated carefully.
Kio estas bone establita
The perception gap is the most robust finding here.
Every method that compares stated spending to measured spending finds a large gap.
The exact multiple varies. The direction never does.
🌍 Why Your Country Changes the Numbers
Almost all of this research is American, and that matters more than it seems.
| Faktoro | Effect elsewhere |
|---|---|
| Regional pricing | Same service, very different price |
| Bundled telecom packages | Common outside the US |
| Cancellation law | Stronger in parts of Europe |
| Currency and card fees | Foreign charges add 2–3% |
| Free public alternatives | Reduce need for some categories |
The currency row catches people out repeatedly.
A service billed in dollars can cost you a few percent more through card conversion fees.
Across a dozen subscriptions that is a real amount, and it never appears on the vendor’s price page.
Check whether local billing exists
Many services bill in your own currency if you ask or set the region correctly.
That removes the conversion fee entirely.
It is a five-minute check that can save a few percent permanently.
Cancellation rights vary
Some countries require cancellation to be as easy as signing up.
Others do not, which is why the same company behaves differently by market.
Knowing your local rules is worth ten minutes if a company is being obstructive.
🚫 Kion Ĉi Tiuj Datumoj Ne Diras al Vi
It does not give your figure. Individual variation dwarfs the averages.
It is heavily US-based. Pricing and habits differ by country.
It does not weigh value. A large bill may be entirely justified.
Definitions are inconsistent. Compare like with like before quoting.
Tracker-published research is not neutral. Check who benefits.
🏁 La Mallonga Versio
People estimate they spend $86 a month on subscriptions and actually spend around $219 (LowerMySubs, 2026).
Published figures range from $84 to $273 depending on what counts.
Roughly $26.79 a month goes to services nobody uses (Fortunly, 2026).
Creep happens through price rises and converted trials, not through decisions.
Find your real number by checking twelve months of statements, PayPal and your phone’s app store.
Then convert everything to annual figures, because $9 a month is $108 a year.
And put the next review in the calendar before you close the spreadsheet.
The wider point is not really about subscriptions.
It is that recurring costs are decided once and paid forever.
Any spending that renews without asking deserves a moment where you ask.
That moment is the whole discipline. Everything else here is just where to look. 📊
❓ Oftaj Demandoj
How much does the average person spend on subscriptions?
Around $219 a month by measured studies. Published figures range from $84 to $273, depending on what counts as a subscription.
Why do people underestimate so badly?
Auto-renewal removes the moment of paying, small amounts do not register individually, and subscriptions arrive one at a time.
How big is the perception gap?
About 2.5x. People estimate $86 and pay around $219, a difference near $1,600 a year (LowerMySubs, 2026).
How much is wasted?
About $26.79 a month on unused services, or roughly $321 a year (Fortunly, 2026).
Why do the published figures vary so much?
Definitions differ. Narrow ones count digital media only; broad ones include phone contracts, insurance and gyms.
How do I find my own number?
Check twelve months of card statements, PayPal recurring payments, and your phone’s app-store subscriptions. The last is missed most often.
Why twelve months rather than three?
Annual subscriptions appear only once a year, so a short window misses most of them.
Why does the audit stop working after a while?
Because it fixes the symptom. Auto-renewal still makes paying the default, so creep resumes unless you change the mechanism.
When is the best time to cancel?
Just before renewal for monthly plans, and four weeks before for annual ones. Cancelling usually keeps your access until the paid period ends.
Can I get a refund on an accidental renewal?
Often yes, if you ask within a few days. Many services refund quietly, but almost nobody asks.
What are the easiest savings I am probably missing?
Household plans, bundles you already qualify for through a phone or bank contract, and duplicate services nobody realised overlapped.
Should I switch to annual billing?
Only after a service has survived a full quarterly review. Annual billing on something you abandon in month three costs more, not less.
Do these figures apply outside the US?
Partly. Pricing, bundling and cancellation law all differ, and foreign-currency card fees can add 2–3% invisibly.
Should I cancel everything?
No. Anything used weekly, saving real time, or covering genuine risk usually earns its place. Test by frequency, not by feeling.
📚 Referencoj
LowerMySubs. (2026). Subscription statistics: What Americans actually spendPrenite la 8-an de aŭgusto 2026, de https://www.lowermysubs.com/blog/subscription-statistics
Fortunly. (2026). Subscription spending statisticsPrenite la 8-an de aŭgusto 2026, de https://fortunly.com/statistics/subscription-spending-statistics/
Trackery. (2026). Average subscription spending per monthPrenite la 8-an de aŭgusto 2026, de https://trackery.app/average-subscription-spending.html
Resubs. (2026). Subscription spending statistics: What the data showsPrenite la 8-an de aŭgusto 2026, de https://resubs.app/resources/subscription-spending-statistics
Free Subscription Tracker. (2026). Subscription costs in 2026: How much Americans spendPrenite la 8-an de aŭgusto 2026, de https://freesubscriptiontracker.org/blog/subscription-costs-2026
Rilata legado en ĉi tiu retejo
Nia analysis of why cancelling feels hard covers the behavioural side. For business software specifically, see our software spending research, kaj la Senpaga Provo Insider-adresaro lists trials with their terms.
Pri ĉi tiu analizo
Published monthly figures range from $84 to $273 because sources define “subscription” differently, and that range is shown rather than averaged. Much of this research is published by companies selling subscription-tracking apps, which is stated in the text. Figures were checked on August 8, 2026.
- AI Video Idea Generators: Turning 100 Ideas Into One Good One - Septembro 2, 2026
- YouTube Automation Tools in 2026: What Can Actually Be Automated - Septembro 1, 2026
- Subscription Creep: How Small Fees Become Big Bills - Aŭgusto 31, 2026