A thread about Amazon Prime benefits came across my feed this week, and the numbers in it were big enough that I went and checked the math myself.
The setup is a good one. A guy pays for Prime for nine years, uses free shipping and Prime Video, and never touches anything else. A coworker who used to work on the Prime membership team watches him buy allergy medication at a CVS over lunch and walks him through everything he has been leaving on the table. Eleven benefits later, the thread concludes he had been extracting about $260 a year from a membership holding $1,300 or more in value he never collected.
That premise holds up. Plenty of the specifics do not. Two of the eleven benefits are not included with Prime at all. The final tally does not match the thread's own line items in either direction. The most-quoted statistic in the whole piece appears to have been made up.
Strip out the wrong numbers and a better story is sitting underneath, particularly if you run any kind of membership.
The short version
-
A US Prime membership costs $139 a year or $14.99 a month. That price has not moved since February 2022.
-
J.P. Morgan ran a sum-of-the-benefits analysis in June 2026 and valued the membership at $1,437 a year, more than ten times what members pay.
-
Two of the eleven benefits in the viral thread, RxPass and One Medical, are paid add-ons that cost extra on top of Prime. They are discounted for members, not included.
-
The thread's own eleven line items add up to between $1,595 and $2,015. It reports $1,575 conservatively and approximately $2,200 aggressively. Neither figure matches its own math.
-
There is no credible public data on how many Prime benefits the average member uses. Amazon has never disclosed it, and the widely circulated "2 to 3 out of 15+" figure traces to nothing.
-
The closest usable figure comes from the September 2025 FTC settlement, where using no more than three Prime benefits in a twelve-month period is one of three requirements to qualify for a refund.
-
Every Prime retention figure in circulation is a survey estimate, and the most-cited one dates to 2019. Amazon itself has never put a renewal rate in public.
-
US Prime membership has flattened. That, more than churn, is why Amazon has spent the last several years stacking benefits into the bundle.
What Does an Amazon Prime Membership Actually Include in 2026?
Roughly a dozen distinct benefits, most of which have nothing to do with shipping.
Here is what Amazon documents as included at no extra cost. Free one and two-day shipping, plus free same-day delivery on orders over $25. Prime Video with ads. Unlimited full-resolution photo storage plus 5 GB of video storage through Amazon Photos. Prime Reading, a rotating catalog of over three thousand books, magazines, and comics. Amazon Music with more than 100 million songs. Amazon Luna, which absorbed Prime Gaming in October 2025 and still hands out PC games you claim and keep, along with one free Twitch channel subscription every month. A free Grubhub+ membership that Amazon valued at $120 a year when it announced the permanent version. Ten cents a gallon off fuel at roughly 7,500 BP, Amoco, Ampm, and Thorntons locations, which Amazon estimates at nearly $70 a year for an average driver. Prime member pricing at Whole Foods. Buy with Prime on outside merchant sites. Sharing with one other adult through Amazon Family.
Then there is Alexa+, which shows how much the bundle has changed. Amazon made it generally available in the US on February 4, 2026, at $19.99 a month, and free for Prime members. Annualized, that single benefit lists at $239.88, which is more than the entire membership costs for a year.
A few things sold as Prime benefits are not free. RxPass costs $5 a month on top of Prime and covers more than 50 generic medications, and it is not available in California or Washington. One Medical runs $9 a month or $99 a year for Prime members against $199 a year for everyone else, and scheduled visits are still billed to you or your insurance on top of that. Both are real value. Neither is included.
Amazon Music comes with a limitation that rarely makes it into these lists. The Prime tier gives you 100 million songs, and since November 2022 most of that catalog is shuffle only, with on-demand play limited to a curated set of All-Access playlists. Calling it a straight replacement for a paid music subscription overstates what it does. Amazon Music Unlimited, the on-demand tier, costs $11.99 a month for Prime members and $12.99 for everyone else.
How Much Are Amazon Prime Benefits Actually Worth?
$1,437 a year, according to J.P. Morgan, against a $139 price.
The analysts ran a sum-of-the-benefits analysis in June 2026 and concluded the membership is worth more than ten times what members pay. The components: $550 in delivery savings, $228 for Prime Video, $156 for Prime Gaming, $120 for Prime Music, with roughly $383 spread across groceries, restaurant delivery, and reading.
One caveat on that $550. It is Amazon's own figure, published in a February 2026 press release stating that US members saved an average of $550 on delivery in 2025, nearly four times the cost of a membership. J.P. Morgan adopted it. Those are not two independent sources agreeing with each other; they are one number appearing twice.
The bigger caveat is structural. A sum-of-the-benefits number assumes you use the benefits, and it prices each one at what a non-member would pay for the closest standalone product. Prime Music at $120 assumes the included tier substitutes for a paid music subscription, which is generous given that most of that catalog plays on shuffle. The same objection applies to several of the components. The total prices everything in the box at retail and says nothing about what any particular member walks away with.
Which Viral Amazon Prime Benefit Claims Are Wrong?
Three categories: benefits that are not actually free, arithmetic that does not survive checking, and one statistic that appears to be invented.
Start with the two that cost extra. The viral version counts RxPass at $252 a year in savings and One Medical at $100, both presented as value already sitting inside a membership you have paid for. Neither is. RxPass is a separate $5 monthly subscription, and One Medical is a separate $99 annual one. Strip those two line items out and $352 of the "included but unused" total disappears before you check anything else.
The arithmetic is its own problem. Here are the eleven annual values the thread itself publishes:
-
Amazon Photos, replacing Google One: $120
-
Grubhub+, membership plus saved delivery fees: $120 to $180
-
RxPass, generic prescriptions: $252
-
Prime Reading, books and magazines: $60 to $120
-
Prime Music, replacing Spotify: $132
-
Alexa+, included free: up to $240
-
Prime Gaming, free games plus Twitch subscription: $160 to $260
-
Fuel savings at ten cents a gallon: $72
-
Amazon Household, eliminating a partner's membership: $139
-
Whole Foods and Fresh grocery discounts: $200 to $400
-
One Medical, discounted healthcare: $100
Add them, and you get $1,595 at the low end and $2,015 at the high end, counting Alexa+ at the full $240 in both columns since the thread gives it no floor. The thread reports $1,575 conservative and approximately $2,200 aggressive. The high figure is not reachable from its own numbers, and the low figure is $20 short of them.
The opening post is further off than that. It puts the whole membership at "$780+" with $642 sitting unused, against the $1,300 in unused value the same thread arrives at eleven posts later. The first number is roughly half the last one, in a thread built entirely on the argument that people underestimate what they are paying for.
The smaller misses run in the same direction. Standalone Amazon Fresh delivery is free over $100, not $50, and it is a different program from the same-day grocery benefit. The free monthly Twitch subscription is valued at $4.99, which has been out of date since Twitch raised Tier 1 subs to $5.99 in July 2024, and neither Amazon nor Twitch publishes an official dollar value for it anyway. Valuing Amazon Music at $132 as a Spotify replacement ignores that the included tier is mostly shuffle.
The claim that matters most is the one with no source at all: that the average Prime member uses two to three benefits out of fifteen or more, while a strategic member uses eight to ten. I went looking for it. It is not in any Amazon disclosure, shareholder letter, earnings call, or analyst note. It is not in CIRP, eMarketer, Morning Consult, or Numerator. Its denominator is questionable too, since Amazon's own benefits page runs to about a dozen included items rather than fifteen.
What I did find is a category of site that ranks well for these exact queries and publishes confident-looking Prime statistics that, read closely, are labeled as the site's own internal estimates. Numbers like these get picked up, stripped of that label, and repeated until they look like research. If you go looking into this yourself, especially with an AI assistant, check whether the statistic you land on names a survey, a sample size, and a date. Most of them will not.
The thread's conclusion survives all of this. What it lacks is evidence that holds up when you check it, which is frustrating, because better evidence was available the whole time.
How Many Amazon Prime Benefits Does the Average Member Use?
Nobody outside Amazon knows, and Amazon has never said.
That is the honest answer, and it is worth stating plainly given how many articles assert otherwise. The closest usable figure turned up in a court settlement instead of a survey.
On September 25, 2025 the FTC announced a $2.5 billion settlement with Amazon over Prime enrollment and cancellation practices: a $1 billion civil penalty, the largest ever in an FTC rule violation case, plus $1.5 billion in consumer redress. The FTC described that as full relief for an estimated 35 million consumers.
The FTC's refund page lists three requirements, all of which have to be met. First, you have to be a US Prime customer. Second, you have to have enrolled through one of four specific flows, or been unable to cancel, between June 23, 2019 and June 23, 2025. Third, in the agency's exact words, you must have "used no more than three (3) Amazon Prime Benefits (including Prime Music or Prime Video products offered for free to Prime subscribers) in any 12-month period following Amazon Prime enrollment."
Read that third criterion for what it is. Two parties with opposite financial incentives had to write down a usage level low enough to mark someone who arguably never wanted the membership, and the number they landed on is three benefits in a year. Refunds are capped at $51 per person.
One thing this does not establish: it does not tell you 35 million people used three or fewer benefits. The three-benefit test is a filter applied on top of the enrollment criteria, not a finding about the population, and the 35 million is the FTC's estimate of consumers affected by the enrollment and cancellation conduct. What it gives you is a usage number that survived an adversarial process, which is more than any other figure on this subject can claim.
Why Does Every Amazon Prime Benefit Have Its Own Door?
Because activation is a separate action from renewal, and only one of them shows up in Amazon's revenue.
The viral thread gets this exactly right, and it is the one part that needed no fact-checking. Grubhub+ requires visiting a specific URL and linking two accounts. Amazon Photos requires downloading a different app. The Twitch subscription has to be manually reclaimed every thirty days. The fuel discount requires enrolling in a third-party loyalty program. RxPass and One Medical require signing up and paying separately. Amazon Family requires both adults to share a primary residential address and agree to share payment methods.
Every one of these is a door, and none of them are open by default.
The tempting conclusion is that Amazon hides this stuff on purpose. I looked for evidence and could not find it. There is a running post on lesser-known Prime benefits, an Explore Prime Benefits storefront, and a full benefits hub, all of them public and none of them hard to find. No reporting I came across establishes deliberate suppression, and what the FTC case surfaced was about enrollment and cancellation friction, not benefit marketing.
The explanation is duller and more useful than a conspiracy. A benefit you know exists and never open still earns its keep on renewal day, and it costs the company nothing to serve. Nothing is locked. Almost nobody at Amazon has a strong commercial reason to walk you through a door.
Does Amazon Prime Have a Retention Problem?
There is no public data either way, and the indirect evidence points to no.
I got into this with Richard, my business partner, when the thread came up. His read was that Amazon simply does not have a churn problem, and that the video and other benefits are aimed more at attracting people who do not shop much than at holding the ones who already do.
Proving the first half is harder than it should be, because Amazon has never publicly disclosed a Prime renewal rate. Not in a shareholder letter, not in a 10-K, not on an earnings call. The figures everyone quotes, 93 percent after the first year and 98 percent after two, come from a CIRP survey of 500 US consumers published in April 2019. That is more than seven years old, and the same study found free-trial conversion had fallen to 64 percent from a peak above 75 percent, which nobody ever quotes alongside it.
Directionally, Richard's instinct still holds. Prime members spent $1,170 a year at Amazon in 2024 against $570 for non-members, per CIRP, using calendar 2024 data that is now closer to two years old. A more recent CIRP read found that gap comes almost entirely from how often members buy, not how much they spend per visit. That is the profile of a habit.
A single point of retention is worth more than it sounds, though the membership base makes it awkward to calculate. Amazon's last officially disclosed figure is more than 200 million worldwide, from Bezos's shareholder letter published in April 2021, and the company has not updated it in more than five years. Outside estimates since then do not agree and do not measure the same thing: CIRP puts 201 million people using a US Prime membership in the April 2026 quarter, counting individuals, while J.P. Morgan models 370 million memberships globally by the end of 2026, split 139 million US and 231 million international, counting accounts.
J.P. Morgan's 139 million US memberships keep the fee math on one basis. One percentage point of retention there is 1.39 million members. At $139 a year that is $193 million in membership fees. J.P. Morgan also estimated a $20 US price increase would drive around $3 billion in incremental annualized net sales, so a retention point on fees alone is worth about 6.4 percent of a price hike.
The fees are the small part. The spending gap is a per-person figure, so switch to CIRP's basis to use it: one percentage point of 201 million US Prime users is 2.01 million people, and at $600 each, that is $1.2 billion in annual retail purchases. Two caveats on that. It is merchandise volume rather than subscription revenue, so it does not sit next to the $3 billion. The $600 is also a gap between two groups rather than a measured drop in what somebody spends after they cancel.
Neither figure is likely to change behavior at a company Amazon's size.
Why Is Amazon Pushing Prime Benefits Harder Now?
Because the membership base has stopped growing, which leaves retention and spend as the available levers.
CIRP put US Prime membership at 201 million in the April 2026 quarter, flat against the December 2025 quarter after several quarters of slowing growth, under a headline asking whether the program has saturated. The share of Amazon customers using Prime on their most recent purchase sat at 73 percent, down slightly from a 75 percent high a year earlier. J.P. Morgan's June 2026 note said the same thing in plainer terms: the base is heavily saturated, which makes it harder to drive meaningful growth from new signups alone.
Amazon disputed a similar CIRP plateau finding back in 2023 and has never conceded saturation. The behavior speaks for itself. Alexa+, RxPass, One Medical, Grubhub+, fuel savings, and free same-day grocery delivery all arrived in the last few years, and the headline price has not moved since 2022. Amazon has been adding to the bundle instead of raising the number on it.
The company is also measuring what those additions do. On the Q2 2026 earnings call, Andy Jassy said in his prepared remarks that in the US, "customers who use Alexa for Shopping spend an average of over 40% more per order than those who don't," and that "customers who've tried Alexa+ are signing up for Prime at nearly 25% higher rates." Those are two different products, and Jassy was careful about the distinction. Both are engagement-to-outcome claims, and a company puts numbers like that in prepared remarks when it wants the market to price them in.
The traffic has not been one-way. Amazon ended the Prime Invitee program on October 1, 2025, which had let members share free shipping with an adult at a different address since 2009. The replacement, Amazon Family, requires a shared primary residence. That is a benefit reduction dressed as a rename.
What Does This Mean If You Are Not Amazon?
A benefit nobody activates does almost nothing for you, and Amazon operates at a scale where that stops being true.
Amazon can stack a dozen benefits behind a dozen separate doors and keep renewing at whatever its real rate is, because the membership solved shipping and shipping turned into a habit. Awareness probably carries a lot of the weight on renewal day. With a membership base Amazon last put north of 200 million, and a decade of purchase behavior underneath it, the bundle can sit mostly unopened and the renewal still happens.
Almost nobody else gets that deal. At most memberships, the value you add outside the core product only counts if the member touches it. An unopened benefit is a line item on a marketing page that a member is paying for and never sees. Closing that gap is mostly a design problem. Every extra step you put between a member and a benefit is a step where some percentage of them stop.
That is what I keep coming back to in the viral thread. Amazon served that guy well for nine years. Nobody ever put the right benefit in front of him at a moment when he could act on it, and it took a coworker at lunch to finally do it.
At BattlBox, we are constantly working on ways to provide more value outside of the physical box. That matters a lot to us, and watching how a company like Amazon handles the same problem at a scale where the rules are different is genuinely useful, as long as you stay honest about which parts transfer and which do not.
There is a version of this that goes wrong in the other direction too. Peloton spent years bundling everything into one expensive proposition before a strategic unbundling helped save the business. A fuller bundle is only better when people use what is in it.
Frequently asked questions
How much does Amazon Prime cost in 2026?
A US Amazon Prime membership costs $139 per year or $14.99 per month. Amazon last raised the price in February 2022, up from $119 per year. Discounted tiers exist for qualifying government assistance recipients and for young adults aged 18 to 24.
How much are Amazon Prime benefits worth?
J.P. Morgan valued a US Prime membership at $1,437 per year in a June 2026 sum-of-the-benefits analysis, more than ten times the $139 price. The largest single component was $550 in delivery savings, a figure Amazon itself published in February 2026. That valuation assumes a member uses the benefits, so it represents the retail price of the bundle rather than the value any individual member extracts.
Which Amazon Prime benefits cost extra?
RxPass and One Medical are the two most commonly misreported. RxPass is a separate $5 per month subscription available only to Prime members, covering more than 50 generic medications, and it is not available in California or Washington. One Medical is discounted for Prime members at $9 per month or $99 per year against $199 per year for non-members, and scheduled visits are billed to you or your insurance on top of the membership fee. Viral posts that count either as value already included in a $139 membership are wrong.
How many Amazon Prime benefits does the average member use?
There is no credible public data on this. Amazon has never disclosed it, and no reputable survey firm publishes it. The frequently repeated claim that the average member uses two to three benefits out of fifteen or more has no identifiable source. The closest usable figure comes from the FTC's September 2025 settlement, where using no more than three Prime benefits in any twelve-month period is one of three cumulative requirements to qualify for a refund. That is an eligibility test, not a measurement of how many benefits members use.
What is Amazon Prime's retention rate?
Amazon has never publicly disclosed a Prime renewal rate. The widely cited figures of 93 percent after one year and 98 percent after two come from a CIRP survey of 500 US consumers published in April 2019. Amazon reports subscription services revenue in its filings, but that line bundles Prime with other subscriptions and does not break out renewals.
Is Amazon Prime still growing?
Not much in the US. CIRP estimated 201 million people using a US Prime membership in the April 2026 quarter, flat against the December 2025 quarter after several quarters of slowing growth. J.P. Morgan described the base as heavily saturated in June 2026. Its own model puts 139 million US memberships and 231 million international by the end of 2026, and the two US figures differ because CIRP counts individuals using a membership while J.P. Morgan counts the memberships themselves. International is where the growth projections are concentrated.
Does Amazon hide its Prime benefits on purpose?
There is no evidence that it does. Amazon publishes posts on lesser-known Prime benefits, runs an Explore Prime Benefits storefront, and maintains a full benefits hub. The more accurate framing is that most benefits require a separate activation step, and Amazon has little commercial reason to push members through those steps, because renewal does not depend on activation.
Final Thoughts
The number that stuck with me is three.
Three is where the refund threshold landed. It is an eligibility screen, and I would not build a theory of the membership base on it.
Three is still a small number to see written into a federal settlement, next to a bundle a major bank prices at $1,437 a year. Whatever else that document is, it is the only place on this subject where a specific usage level had to be written down and defended.
Amazon built a bundle valued at roughly ten times its price, put a door in front of nearly every piece of it, and the membership held up anyway. That is what a decade of shipping habit buys you.
The rest of us are running a different business. Learn from how much value Amazon packs into Prime. Copy nothing about how it gets people to use it.
Interested in learning more?
Here are a few reads to check out:
