Skip, pause, every three months: a Subscribe & Save playbook for the Alexa era

If you sell coffee pods, vitamins, pet food, diapers, detergent or anything else people run out of, Subscribe & Save is probably one of the quieter engines of your Amazon business. A shopper picks a cadence once, and then the orders keep coming until something makes them stop.

For years, the thing that most often made them stop was friction in the other direction. Overstocked shoppers had to open the app, find the subscription, and change the schedule. Many canceled outright instead, or let boxes pile up until they did.

On October 29, Amazon changed the effort on both sides. In a post titled Alexa for Shopping adds 100+ new capabilities, it listed Subscribe & Save among the things shoppers can now manage "just by saying" a sentence. Amazon's own examples: "Skip my next delivery," "Pause my subscription," and "Change my delivery to every three months." The actions are available now to all US Amazon customers, from the Alexa icon in the Amazon Shopping app or at the top of the screen on desktop.

This post is a playbook for brands that depend on repeat orders. It does not repeat our broader Alexa for Shopping playbook; it focuses on the cadence question.

What changed, in Amazon's words

The October 29 update covered five groups of actions: lists and registries, Subscribe & Save, Prime and other memberships, delivery instructions, and payments and billing. Amazon said the number of customers using Alexa for Shopping's agentic features "has grown more than eight times since the start of the year," and that over 400 million customers have used the assistant in the last 12 months.

Table: What Amazon reported about Alexa for Shopping and essentials, October 29, 2026

Measure	Figure Amazon reported	Where
Customers who used Alexa for Shopping, last 12 months	400 million+	Q3 release and Amazon blog
Customer interactions, year over year	up 570%	Q3 release
Customers using agentic features, since start of 2026	grown more than 8x	Amazon blog
Fresh groceries and everyday essentials, share of units	about half	Q3 release
Groceries and essentials growth vs. rest of store	about 3x faster	Q3 release

Sources: Amazon Q3 2026 results and Amazon's October 29 Alexa for Shopping post.

Amazon has not published how many shoppers manage subscriptions this way, or what happens to subscription retention when they do. Neither have The Verge or TechCrunch, which covered the update the same day. So what follows is our reasoning, not a measured effect.

Why it cuts both ways

Easier cadence changes can help a brand or hurt it, depending on who is asking.

The overstocked shopper. Someone with three unopened bags of dog food used to face two choices: put up with it, or cancel. Now they can say "skip my next delivery" or "change it to every two months" in a few seconds. For a brand, a slower subscription is better than a canceled one. That is the good case.

The autopilot shopper. Someone whose delivery arrived on time and got used without much thought had no reason to touch it. Now the assistant is a place where they already ask questions, and a chance to reconsider is one sentence away. Some will slow down when they did not need to and run out. Running out is the moment a shopper is most likely to grab whatever is available, or ask the assistant what to buy instead.

The new subscriber. A first-time buyer picking a cadence is usually guessing. If they guess too often, they overstock and pause. If they guess too rarely, they run out. Either way, the first choice is shaky.

The common thread is information. A shopper or an assistant can only choose a good interval if it knows how long one unit lasts. That is the gap most listings leave open.

What Amazon's assistant can read

Amazon describes Alexa for Shopping as having "access to hundreds of millions of products and their detailed information and reviews, plus customers' shopping history." It also says the assistant gets more personalized "learning from conversations, browsing, purchases, and preferences."

Two of those inputs are yours: product information and, indirectly, reviews. Shopping history belongs to Amazon and the shopper. If a shopper asks "how often should I get this delivered," the assistant can lean on their own purchase gaps, on what reviewers say, or on what your listing says. Only one of those is in your hands.

Look at a typical consumables listing with that question in mind. You will usually find net weight, a count, maybe "servings per container" in a nutrition panel image. You will rarely find a sentence that says "one bag lasts a 30-pound dog about four weeks" or "one 64-load bottle lasts a family of four about two months at one load a day." Images are a weak place to hold that fact; text is easier for both people and software to read. We made the same point about HTML tables on product pages.

The playbook

Here is what we would do over the next two weeks if Subscribe & Save is a meaningful share of your Amazon revenue.

1. Write a days-of-supply line for every subscribable item

For each item, write one sentence with three parts: the use count on your label (servings, loads, pods, tablets), the use rate you are assuming, and the result in days or weeks. For example: "Contains 60 capsules. At the label serving of two a day, one bottle lasts 30 days."

Keep it strictly tied to your label. If your label says two capsules a day, do not write "lasts up to two months." If use varies a lot (a dog's weight, a household's laundry), give a short range by size, not a single optimistic number. Tip 40 has the checklist.

2. Put it in text where the assistant can read it

Add the line to an Amazon bullet, ideally near the top, and repeat it in the description. Put the same sentence on your own product page, and in your feed to other retailers. The same fact in the same words everywhere removes a source of conflicting answers.

3. Explain pack sizes by household, not by price

If you sell a 12-pack, a 24-pack and a 48-pack, say who each one suits. "The 24-pack is about a month for two coffee drinkers having one cup each a day" is something an assistant can repeat when a shopper asks which size to subscribe to. It also nudges the shopper toward a size that matches a sensible cadence.

4. Answer the cadence question in the Q&A and reviews you already have

Look through your existing customer questions for "how long does this last" and similar. Where shoppers keep asking, that is a sign the answer belongs in your bullets. Where reviewers report a very different use rate from yours, check whether your serving guidance is clear enough.

5. Test what the assistant says

Once a month, open Alexa for Shopping and ask, in a shopper's words: "How long will one bag of [your product] last?" and "How often should I get [your product] delivered?" Try the version with your brand name and the version without it. Write down the answer and whether it uses your numbers. If it gives a wrong figure, trace it back to its likely source, usually a review or an old listing. Tip 25 covers what to do when an assistant gets your product wrong.

6. Watch your subscription reports for a shift

If your Amazon reports include Subscribe & Save data, use it. Over the next two to three months, look at whether skips and pauses rise, whether average intervals lengthen, and whether cancellations fall as a share of changes. You will not be able to attribute any change to the assistant directly. A clear shift after October 29 is still worth knowing about.

What about Prime?

Amazon does not break out Subscribe & Save in its results. Prime and its other memberships sit inside the subscription services line, which has grown steadily through 2026.

Show the data

Quarter	Revenue, $ millions
Q2 2025	12208
Q3 2025	12574
Q4 2025	13122
Q1 2026	13427
Q2 2026	13730
Q3 2026	14006

Source: Amazon, Q3 2026 results, segment information. Subscription services includes Prime and other subscriptions; Amazon does not break out Subscribe & Save.

The steady climb matters because the same update lets shoppers ask "How much have I saved with Prime this year?" and manage memberships by conversation. Amazon is making the whole account easier to talk to, and subscriptions are part of it.

Beyond Amazon

The cadence problem is not unique to Amazon. If you run subscriptions on your own site, the same logic applies: a shopper who can see how long a unit lasts picks a better interval. Shopping agents that buy on a shopper's behalf, like the ones Walmart is preparing to admit with spending limits, will need the same fact to reorder at the right time.

That is why we would write the days-of-supply line once and use it everywhere: Amazon, your site, Walmart, Target and any feed you send out. It costs an afternoon. It also gives you a clean way to check what assistants say: if an answer about your product's lifespan does not match your line, you know something upstream is wrong. Teams that want to track those answers across assistants can do it by hand or with tools such as XENA Intelligence and its SearchPanel, which show how each product is described.

The bottom line

Amazon just made subscription changes a one-sentence task. That will not cost you subscribers who love your product. It will expose the ones who were never sure how much they needed. Give them, and the assistant they are talking to, the number that answers that question.