What Push Notification Buyers Wish They'd Known Before They Signed

Six lessons from verified interviews with the people who send mobile and web push: push bundled with email covers the basics but some formats and setups need checking, on Shopify the app builder often sends push while segments and journeys live elsewhere, push depends on an opted-in audience, mobile channels need engineering, fatigue is a coordination problem, and browser permissions limit web push.

Based on verified interviews with the CRM, lifecycle, retention and app teams who send push and in-app messages, at DTC and enterprise brands. Buyers are anonymized before publication; vendor names and views are reported as given. No vendor paid to appear or could edit this page.

7.7/10
Klaviyo's overall buyer rating. Its mobile push sits beside email and SMS; buyers describe friction with deep-link setup and richer formats, and native web push is not supported.
7.5/10
Braze, praised here for app engagement and for letting urgent product alerts take priority over marketing messages.
Opt-ins
The limit buyers name on push. Mobile push needs an installed app, an eligible device and notification permission; web push needs browser permission. The reachable audience is smaller than the customer base.

What do push notification buyers wish they had known before buying? In Alium Research's verified interviews, the main lesson is that the push tool matters less than what surrounds it: the app and its opt-ins, the platform that holds the segments, the engineers who set it up and the system that decides which message goes first. Buyers describe six recurring lessons: push bundled with an email platform covers the basics, but advanced formats, deep-link setup and web push need checking; on Shopify the app builder often sends push while segments and journeys live elsewhere; push depends on an opted-in audience, and high-value app users were often valuable already; mobile channels need engineering; fatigue is a coordination problem; and browser permissions limit how far web push reaches. For what push is and who runs it, see what is push notification software?; for the messages shown inside the app, what is in-app messaging software?; and for the channel beside it, what SMS-marketing buyers wish they'd known.

The six lessons

01

Push bundled with your email platform covers the basics, but advanced formats, deep-link setup and web push need checking

Companies sending push from a separate tool often want it in the platform that already sends their email and SMS. A media subscription company that uses its push tool only for blasts says “My recommendation is not that we invest heavily in OneSignal, it's that we just get a platform that does all of the things we need it to do.” A fashion brand with “all these different PushOwls, like a one-off app through Shopify” saw Iterable as a platform that “could house all of them,” and an online learning company that had just renewed OneSignal for web push says “In an ideal world, I will be able to do that via Braze, which is another selling point for Braze.” Those three are plans, not completed moves. A health company that is moving its messaging to Customer.io, a migration under way, chose it for “the ability to connect our SDK and also send push notifications and link up the email and SMS together.”

Bundled push has limits that buyers found once they relied on it. An apparel brand runs app notifications from its app builder and web push from PushOwl beside Klaviyo: “The main reason for using PushOwl is that Klaviyo doesn't support web push notifications.” It also wants richer notifications, saying “you cannot add any countdown timer or live coupons that cannot be added via notifications.” A jewelry brand that now schedules push in Klaviyo finds deep links harder than in the app builder it used before (its story is below). These are three different kinds of gap. Klaviyo's help center, updated November 6, 2026, says it does not support web push, although web notifications can be triggered through API calls. Mobile deep links are supported, but the destinations and the routing are set up in the app by its developer, so the jewelry brand's complaint is about the work of using them, not their absence. Richer formats depend on the vendor, the phone's operating system and how the app was built. What these buyers describe is friction in their own setups, not a limit every Klaviyo customer meets.

Named in this context: Klaviyo · OneSignal · Braze · Iterable · Customer.io · PushOwl

02

For Shopify apps, the builder often sends push, while segmentation and journeys may live elsewhere

Several Shopify brands here send mobile push through the app builder that runs their app, such as Tapcart or Fuego, and like how quickly marketers can launch. Of Fuego, one says “Anyone can come on day one, and then they can start sending app notifications”; of Tapcart, another says “Super, super easy to use, great UX, good analytics, easy push notifications.” But the system that delivers a push is not always the one that holds the audience, the segment or a multi-step journey. Some brands keep those in Klaviyo and use the builder to deliver. An apparel company with several brands finds “there's very little personalization or ability to segment in Tapcart,” so its abandoned-cart and welcome journeys for the app still run from Klaviyo. An apparel brand on Fuego says “The segmentation is very poor” and “For advanced segmentations, we have to push the segments from Klaviyo to Fuego, and then we need to send them across,” and it builds multi-step flows in Klaviyo for the same reason. Others run their campaigns and push entirely from the builder. The complaints describe each company's setup at the time. Tapcart's help center, in an article dated August 11, 2026, describes a Klaviyo integration for Enterprise customers only that sends automated push based on user behaviour and push segmented by user location, and Fuego's Shopify App Store listing offers personalized push to segments, scheduled or triggered automatically for abandoned carts, restocks and shipped orders. A feature on a vendor's site does not show that the companies interviewed had it or used it; our recommendation is to test your own contract tier, event data, journey branching and the handoff between systems.

When an app builder controls both the app and its push delivery, replacing it can mean rebuilding the push configuration, the integrations and the app's journeys, even if the email platform and its segments stay. Several brands here have completed a move off Tapcart: to Fuego, to Appbrew or to another app builder, in one case to an agency-built custom app, and in another to no app at all; two more ended it without naming a destination, and others are still evaluating. Each company is counted once, and this is not a churn rate. The reasons they give are app customization, faster updates and the work of maintaining the app, not push. “We were able to do a lot more in terms of customization on the app versus what we had been able to do in the past,” says an accessories brand, and a beauty retailer says its new builder “was able to provide to us a much quicker turnaround in terms of tech updates.” Other brands rate Tapcart well for its services, and the ratings on this page come from different groups of buyers, so they are not a before-and-after comparison.

Named in this context: Tapcart · Fuego · Appbrew · Klaviyo

03

Push depends on an opted-in audience, and high-value app users were often valuable already

Mobile push needs an installed app, a device that can receive it and a customer who allows notifications, so the reachable audience is smaller than the customer base and has to be built through app adoption and permission requests. It can grow slowly, especially for retailers whose customers have little reason to install an app, though the interviews do not establish a typical time. A home furnishings retailer that sees little need for push today still worries that “if we haven't built up an opt-in strategy up until that point, you're basically starting from scratch at that point.” A merchandise company decided against an app entirely “because we know that people are not necessarily going to download the app anyways,” and others list app downloads and push opt-ins among their growth goals.

The customers who do opt in look like the best ones. A fashion retailer with stores finds “the customer who has an app is much more valuable and is much more loyal, and their frequency is about five times more than a customer who just shops at the store,” and another fashion retailer says app customers “tend to buy with a higher conversion rate and have a higher average order value.” The brands that tried to measure what the app and its push add are less sure. A direct-to-consumer brand says of its app “I would love to shut it down, and see what happens. I just haven't.” A men's accessories brand is “still not convinced that it's incremental,” an apparel brand says “proving incrementality on a mobile app is something we're always challenged with,” and a bag brand dropped its app because it was too much work to maintain and “we weren't feeling like we were driving incremental value through the app users.” App users and everyone else are two self-selected groups, and so are customers who allowed notifications and those who declined. Even if app users buy more, that does not show what one push campaign added: whether an app attracts valuable customers and whether its notifications change what they do are separate questions.

Named in this context: Tapcart · Fuego

This is the average buyer's picture. What does your stack need?A free, peer-powered checkup returns a shortlist for your stack and your next steps.
Start my checkup →
04

Push and in-app messages need engineering, even on platforms marketers run

Platforms that marketers run day to day still need developers for the mobile channels. A healthcare company's lifecycle director rates Salesforce Marketing Cloud highly for email and lower for mobile, saying “anything related to mobile channels, so SMS, push, in-app, and also our message center, does require a bit of additional engineering and setup, and it's not as straightforward as email.” A retailer using Optimove's push found it “was requiring my team, who are frontend business users, they're not technologists or engineers, to basically understand JSON in order to send out targeted push notifications.” A Braze user describes its Content Cards, which personalize the app's home screen and deliver offers: “You just need developers in order to place that content.” Who controls the app matters as well. At an app company, the product and engineering teams pushed back when marketing wanted to send notifications and in-app messages from its email platform, because they wanted to own those channels. The work is the SDK, deep links, message payloads and the app releases that carry them, and it comes before the first campaign. For how often buyers say they need developers, and who owns push, see what is push notification software?

Named in this context: Salesforce Marketing Cloud · Optimove · Braze · Iterable

05

Fatigue is a coordination problem: caps and priorities need one view of what each customer was sent

Buyers worry less about a single push than about the pile-up across channels and tools. A large retailer's CRM director says “I have yet to find a great way for us to say, this customer, the best message for them today is this in this channel,” and that suppressing sends to protect frequency can drop the one push that mattered that week. A software company sending transactional email and push from one platform, in-app guides from another and marketing messages from a third says “it remains a challenge knowing exactly what is being fired out to each customer, managing fatigue, and ensuring that there are no duplicative customer journeys from the lack of visibility.” The problem is sharpest where product alerts and marketing share a channel. A company now integrating Braze, whose product sends time-sensitive alerts as push, says “I don't want a customer to get three messages at the same time. Then they're going to turn off notifications. But Braze allows us to do that prioritization.” Even a team happy with Braze for push notes the cost of splitting channels between tools: “It's always good to have one tool that can address multiple channels versus having too many different tools.” These accounts point to a coordination requirement, not proof that one platform solves fatigue. Campaigns, journeys and product alerts can start in different systems, and a frequency cap in one tool cannot count sends from another unless delivery and suppression data are shared and current. The implication, ours rather than a buyer's, is that brands need precedence rules so an urgent service notice is not blocked by a promotional cap.

Named in this context: Braze · Iterable · Amplitude · SendGrid

06

Browser permissions limit web-push reach, and declining engagement is hard to explain

Web push can work for publishers, and buyers say its reach has become harder to hold. A media publisher that sends several web pushes a day on Pushly saw delivery and clicks level off and then fall: “It feels a lot like SEO where there's just kind of an unknown of what Google's doing with what we're sending out.” In a later interview it rates the channel lower again while sending the same number of messages, saying “it seems like it was a lot better when we started than it is now.” Google announced in October 2025 that Chrome on Android and desktop would automatically remove notification permission from sites with very low engagement that send a high volume of notifications, and said less than 1% of notifications get any interaction; the change does not apply to installed web apps, and users can restore the permission. That adds a limit the browser controls, but the interviews alone cannot show it caused the publisher's decline: lost subscribers, eligibility, delivery, content and engagement are separate factors. For retailers web push is a small channel. A beauty company sending it through Batch says “it makes more money than it costs, but it's not anything significant in terms of revenue,” and an electronics retailer that stepped back from texts says browser push is the same story: “Some people just feel like you're hitting them over the head at a certain point.”

Named in this context: Pushly · Batch · PushOwl · OneSignal

How buyers rate the tools

These are overall product ratings for tools that include push, not scores for push alone. Klaviyo rates 7.7/10 and Braze 7.5; Insider rates around 8 and MoEngage around 7.5, rounded because fewer buyers rated them; Airship, the dedicated push provider named most, rates 6.1. Each figure comes from a different group of buyers, some of them rating email, SMS or broader engagement, so the table does not show which tool sends the best push. Tools with too few ratings receive no number.

Buyer ratings for push notification tools, cross-channel messaging platforms and app builders, Alium verified interviews through October 2026. A tool rated by fewer than about twenty-five buyers is rounded rather than given a decimal; fewer than eight is not given a number.
Tool Rating What buyers describe
Insider around 8 Too few buyers for a decimal. A cross-channel platform that sends app and web push beside email and SMS.
Klaviyo 7.7 Mobile push bundled with email and SMS; buyers describe deep-link implementation effort and richer-format limitations, and native web push is not supported.
Braze 7.5 Push, in-app messaging and Content Cards, with buyers praising app engagement and message prioritization. Salesforce announced an agreement to acquire Braze on October 20, 2026; as of this update, the transaction remains pending, with closing expected in Salesforce's fiscal 2028.
MoEngage around 7.5 Too few buyers for a decimal. A mobile-first engagement platform with push and in-app messaging.
Iterable 7.0 A cross-channel platform where push sits in the same journeys as email and SMS.
Tapcart around 7 Too few buyers for a decimal. A Shopify app builder whose push comes with the app; buyers praise its ease and services, and some find its push targeting thin.
Salesforce Marketing Cloud 6.8 Push and in-app beside its email tools; buyers find the mobile channels need more engineering than email.
Customer.io around 6.5 Too few buyers for a decimal. Event-triggered messaging, with push beside email and SMS in one workflow.
Airship 6.1 The dedicated push tool named most here; buyers who rate it highly cite easy setup, analytics and responsive account managers.
OneSignal not rated Too few buyer ratings to publish. A push tool companies here use for app or web push.
Fuego not rated Too few buyer ratings to publish. A Shopify app builder with push built in; buyers praise its ease and service.

Want this checked against your own stack?

Start my checkup →

The stories behind the lessons

An apparel company: push from the app builder, journeys from the email platform. An apparel company with several brands sends push from the app builder behind one brand's app, while its abandoned-cart and welcome journeys still come from Klaviyo. Some messages do not arrive as built: “But then there's breakage to where sometimes somebody won't get the push notification with the code, the promo code, and we still haven't quite figured out what's going on with that.” It would like to move those journeys into the app builder but hesitates, because the builder cannot branch a journey on what the customer does: “So my concern was would there be potentially a loss of revenue if we migrated over to something that was a little bit more basic?” CRM director · multi-brand apparel company
A direct-to-consumer brand: free push, unproven lift. The original case for a direct-to-consumer brand's app was reaching customers without paying for texts: “the thought was initially with Tapcart, is we can send push notifications to everyone that downloads our app for free, avoiding the per send SMS fees, and the platform fees.” The results did not show up where it expected: “You'd think when we launch a new collection, and we shoot a push notification, I'd see a noticeable bump in revenue at that period, because I do for email, I do for SMS.” Ecommerce and marketing director · direct-to-consumer brand
A jewelry brand: a custom app, and push that moved into the email platform. A jewelry brand left its app builder for an agency-built app so it could design custom features, including loyalty inside the app. Push moved to Klaviyo, which it rates a nine for email and lower for push: “We now have a custom-developed app, and through doing that, we can schedule through Klaviyo.” Deep links are the trade-off: “You have to actually build a custom URL for every single push notification you put together. I find that to be a little tedious, which is not really a fault of Klaviyo's, but I find that it's not the most user-friendly way of doing it.” Retention and client experience director · jewelry brand

The counter-current: where the app is the product, push earns its place

Not every buyer is fighting the channel. An app-first company chose Braze because it “is especially powerful when it comes to mobile app engagement and push notification marketing,” and another team says of Braze “For push notifications, it's been pretty good. The teams have been happy with it.” One buyer that picked Airship for push says it is a tool “that we've been quite happy with,” and another rates Airship an eight because “The ease of the platform is great.” A brand that moved its app to Fuego reports “a huge lift in app conversions and app downloads.” Web push can work too: a media publisher says “we have a lot more push subscribers than email subscribers and we get a lot more clicks from push subscribers than email subscribers.” The satisfied buyers tend to be companies whose customers already live in the app or on the site, with someone who owns the push program.

What this means for your push program

Across these interviews, the decisions that matter most are the app and its opted-in audience, where the segments and journeys live, the engineering that makes delivery reliable, and the rules that govern every channel. Our recommendations, not practices buyers describe, start with an inventory: list the push capabilities you actually use, such as scheduled campaigns, automated journeys, deep links to specific screens, rich formats, in-app messages and web push, and test each on the proposed platform with your real segments, destinations and workflows before you cancel a separate tool. Where an app builder delivers the push and another system runs the journey, verify the event sync, the segmentation, the branching and what happens when a send fails. Budget developer time for the SDK, deep links, payloads, permission prompts and app releases separately from the licence. Write separate rules for marketing, transactional and urgent alerts, and share send, suppression and priority information across every tool that messages the same customer.

To find out what a push campaign adds, randomly assign eligible opted-in users to receive it or to a holdout, keep every other message comparable for both groups, and measure an outcome chosen in advance across everyone assigned, including people who never opened a notification, over the same period. Account for the uncertainty, and report an inconclusive result as inconclusive. Keep that result apart from opens, clicks and the revenue the platform attributes to push, which count only people who responded. A campaign holdout estimates what that campaign did for that audience, not the value of the app or of the whole channel. For web push, track permitted subscribers, permission changes, eligible recipients, attempted sends, deliveries, clicks and conversions separately, evaluate installed web-app audiences on their own, and do not put a change down to one browser policy without evidence. For the basics of the channel, see what is push notification software?

Common questions

What do push notification buyers wish they had known before buying?

Six things recur in these interviews. Push bundled with an email platform covers the basics, but advanced formats, deep-link setup and web push need checking. On Shopify the app builder often sends push, while segments and journeys may live in the email platform. Push reaches only an opted-in audience, and the app users who opt in were often valuable customers already. Mobile channels need engineering even on platforms marketers run. Fatigue is a coordination problem across tools. And browser permissions limit how far web push reaches.

Is the push notification feature in Klaviyo enough, or do you need a separate push tool?

For scheduled and automated mobile push to app users, buyers here run it from Klaviyo. The friction they describe is the work of setting up deep links, which one brand builds as a custom URL for every notification, richer formats such as countdown timers, and web push. Klaviyo's help center, updated November 6, 2026, says it does not support web push, although web notifications can be triggered through API calls, which is why one apparel brand runs PushOwl beside it.

Do push notifications drive incremental revenue?

The interviews do not establish it. Brands report that app users buy more often and spend more, but several that tried to measure push or the app say they cannot tell whether it adds sales beyond email and SMS, and one dropped its app partly for that reason. App users comparing well with other customers shows who chose the app, not what a push campaign added; a randomized holdout of opted-in users is our recommendation for finding out.

Why is web push performance declining?

A media publisher that sends several web pushes a day says delivery and clicks fell over time while its volume and content stayed the same, and it puts that down to how the browser shows prompts and notifications rather than to its vendor. Google announced in October 2025 that Chrome on Android and desktop would automatically remove notification permission from sites with very low engagement that send a high volume of notifications. The change does not apply to installed web apps, and the interviews alone cannot show that it caused the publisher's decline.

Which push notification tools do buyers rate highest?

Insider rates around 8/10, Klaviyo 7.7, Braze 7.5, MoEngage around 7.5, Iterable 7.0, Tapcart around 7, Salesforce Marketing Cloud 6.8, Customer.io around 6.5 and Airship 6.1. These are overall product ratings for tools that include push, not push-only scores; for the messaging platforms they cover email and SMS as well, so they do not show which tool sends the best push. OneSignal and Fuego have too few ratings to publish a number.

Get a checkup on your tech stack

A checkup for your marketing stack, built from hundreds of verified buyer interviews: what matters most to you, a shortlist for your stack, and your next steps.

Start my checkup

Free · no password · your answers are anonymized before they inform research like this.

Want to come back to this?We'll email you a link to this page, so it's easy to find later or forward to your team.
Methodology. Alium conducts verified interviews with software buyers — the CRM, lifecycle, retention, ecommerce and product leaders who select and use these tools. The lessons on this page come from interviews in that corpus since 2024, through October 2026, in which buyers discuss mobile push, web push, in-app messages, a push tool or the app builder that sends their push. Statements about what buyers did are confirmed in the buyer's own words; moves between tools are counted by company and classified as completed, under way or planned; advice is marked as ours. Vendor capabilities are checked against each vendor's own documentation, and Google's announcement for Chrome, in November 2026 and dated. The ratings table is each tool's average across every buyer in the corpus who rates it, with duplicate product rows combined; for the messaging platforms it covers email and SMS as well as push. A tool rated by fewer than about twenty-five buyers is rounded rather than given a decimal, and one rated by fewer than eight is not given a number. Buyer identities are verified at interview time and anonymized before publication; vendor names and ratings are reported as given. No vendor paid to appear or was able to edit this page.