Writing push notifications and mobile messages
Why a correct text gives CTR 1-2%, not 10%
The average user receives 50 to 100+ push notifications per day. Reaction to each is formed within 0,5–1 seconds — in this time, the gaze scans the app icon and the first 2-3 words of the title, the rest is usually left unread. This state is called notification blindness: the finger swipes the notification before the brain has time to consciously perceive it.
According to Pushwoosh data from 2025 year, the CTR spread across industries is huge — from 0,5% to 7,4%, and the benchmark for an effective campaign is 5-10%, with top results reaching 10%. If the text is made "by the rules," but CTR is stuck at 1,5-2%, the issue is not with the call-to-action wording. The problem is that copywriting competes not with other text, but with the reflex of "swiping away without looking." This is not fixed by a more vivid verb, but by work at the level of the first two words of the title and targeted segmentation — send less frequently, but more relevant to a specific person.
Fogg's Model: Why motivation in the text is insufficient
BJ Fogg's B=MAP behavioral model describes an action as the product of three factors: motivation, ability (ease of execution), and trigger cue. Copywriters of push notifications almost always work only with motivation — they write about discounts, scarcity, and benefits. But if the ability to perform an action is low (e.g., it takes a long time to load an app, enter a password, or find a section), even the most motivating text will not work: the user will click and leave.
From this, a practical conclusion follows: the highest ROI is not provided by boosting motivation in the text, but by reducing friction after the click. If a push leads directly to a product card with an already applied discount, rather than to the main page of the app, the conversion from click to target action will increase more than that from any rephrasing of the headline.
Tone of voice for push: why you cannot copy from email
Brands often simply truncate the email tone of voice to the size of a push — and get either a stilted text or excessive familiarity, which in the context of the lock screen appears intrusive. Push is a channel that invades the personal space of the phone without consent, so the tone here should be more restrained in emotion but more direct in content than in an email or social media.
A working approach is not to rewrite the existing brand book, but to create a separate application for push with three parameters:
- the degree of formality in the first three words of the headline — these are the ones that are read;
- the acceptable set of emojis and symbols (they are displayed differently on different operating systems);
- the boundaries of personalization — addressing by name works, but excessive details about user behavior («we saw that you viewed...») more often scare than attract.
Urgency that converts, and manipulation that kills subscriptions: case study
The difference between «There are only a few units left» and «There are 3 left in your size» is the difference between manipulation and genuine urgency. The first formulation is a template that the user subconsciously reads as a mass mailing and a standard response. The second contains a specific number and a personal data reference (size), and is therefore perceived as a signal that the message was made specifically for this person, not sent out to everyone.
A telling antitype is the frequency without relevance. Studies show: 46% of users unsubscribe if they receive from 2 to 5 messages per week without an explicit personal reason. That is, urgency works only when it is substantiated by specifics (a number, size, remaining quantity), not when it is used as a hook in and of itself. A manipulative pattern is "Act now within 24 hours!" without explaining what exactly is burning; a converting pattern is "Only 3 pairs of 42 size left in stock, we're holding the reserve until evening." The difference is in one word — in the second case, the user can verify the truth of the statement by clicking.
iOS and Android: different text due to different screen behavior
Technical differences between platforms directly affect how the message should be structured.
| Parameter | iOS | Android |
|---|---|---|
| Visible text volume | 40-50 characters visible on the lock screen for the title (the rest is truncated) | slightly more, less rigid truncation |
| Behavior on the lock screen | notification disappears after phone unlock | notification remains visible in the strip until explicitly swiped away |
| Impact on CTR | first phone access — the only chance to grab attention | the user has several 'approaches' to one notification |
| Rich push with media | requires Notification Service Extension — without it only text will be sent | media content is easier to attach, without mandatory extension |
From this follows a practical rule: for iOS, the key idea should be squeezed into the first words of the title as tightly as possible — there may be no second chance. For Android, a more elaborate structure is acceptable, because the notification physically stays visible for longer.
Web push, SMS and in-app: different text logic
Three channels often write the same way, although the context of message delivery is fundamentally different for each.
| Channel | Context of delivery | Copywriting feature |
|---|---|---|
| Web push | browser is open, user is busy with another task | a maximally fast hook is needed, competition with website tabs |
| SMS | a separate messenger app, perceived as a more personal channel | text without emojis and caps, closer to a business message, link is mandatory and short |
| In-app push | user is already inside the app, there is context of their current action | can refer to what the person is doing right now, tone is more conversational |
The path from click to goal: where copywriting hasn't finished its work yet
It's a mistake to think that the text's job ends at the click. In fact, push copywriting continues to work afterwards — if the text promised specificity, but the landing page doesn't confirm it instantly, the user closes the app before the target action.
- The headline and first words — form the decision to open within 0,5-1 seconds of scanning.
- Click — transfers the expectation set by the text to the app screen; if the offer in the push was "20% discount on sneakers," but the main page opened — trust is lost instantly.
- The first screen after the click — must visually confirm the push text literally: the same product, the same figure, the same context.
- The target action — here, the push text no longer works, but the factor of "ability" from the Fogg model: the number of steps to purchase or subscribe.
Therefore, diagnosing low CTR needs to be separated from low conversion after the click — these are two different problems, which are treated with different tools: the headline text in the first case and simplifying the in-app scenario in the second.
Why a correct text gives CTR 1-2%, not 10%
The average user receives 50 to 100+ push notifications per day. Reaction to each is formed within 0,5–1 seconds — in this time, the gaze scans the app icon and the first 2-3 words of the title, the rest is usually left unread. This state is called notification blindness: the finger swipes the notification before the brain has time to consciously perceive it.
According to Pushwoosh data from 2025 year, the CTR spread across industries is huge — from 0,5% to 7,4%, and the benchmark for an effective campaign is 5-10%, with top results reaching 10%. If the text is made "by the rules," but CTR is stuck at 1,5-2%, the issue is not with the call-to-action wording. The problem is that copywriting competes not with other text, but with the reflex of "swiping away without looking." This is not fixed by a more vivid verb, but by work at the level of the first two words of the title and targeted segmentation — send less frequently, but more relevant to a specific person.
Fogg's Model: Why motivation in the text is insufficient
BJ Fogg's B=MAP behavioral model describes an action as the product of three factors: motivation, ability (ease of execution), and trigger cue. Copywriters of push notifications almost always work only with motivation — they write about discounts, scarcity, and benefits. But if the ability to perform an action is low (e.g., it takes a long time to load an app, enter a password, or find a section), even the most motivating text will not work: the user will click and leave.
From this, a practical conclusion follows: the highest ROI is not provided by boosting motivation in the text, but by reducing friction after the click. If a push leads directly to a product card with an already applied discount, rather than to the main page of the app, the conversion from click to target action will increase more than that from any rephrasing of the headline.
Tone of voice for push: why you cannot copy from email
Brands often simply truncate the email tone of voice to the size of a push — and get either a stilted text or excessive familiarity, which in the context of the lock screen appears intrusive. Push is a channel that invades the personal space of the phone without consent, so the tone here should be more restrained in emotion but more direct in content than in an email or social media.
A working approach is not to rewrite the existing brand book, but to create a separate application for push with three parameters:
- the degree of formality in the first three words of the headline — these are the ones that are read;
- the acceptable set of emojis and symbols (they are displayed differently on different operating systems);
- the boundaries of personalization — addressing by name works, but excessive details about user behavior («we saw that you viewed...») more often scare than attract.
Urgency that converts, and manipulation that kills subscriptions: case study
The difference between «There are only a few units left» and «There are 3 left in your size» is the difference between manipulation and genuine urgency. The first formulation is a template that the user subconsciously reads as a mass mailing and a standard response. The second contains a specific number and a personal data reference (size), and is therefore perceived as a signal that the message was made specifically for this person, not sent out to everyone.
A telling antitype is the frequency without relevance. Studies show: 46% of users unsubscribe if they receive from 2 to 5 messages per week without an explicit personal reason. That is, urgency works only when it is substantiated by specifics (a number, size, remaining quantity), not when it is used as a hook in and of itself. A manipulative pattern is "Act now within 24 hours!" without explaining what exactly is burning; a converting pattern is "Only 3 pairs of 42 size left in stock, we're holding the reserve until evening." The difference is in one word — in the second case, the user can verify the truth of the statement by clicking.
iOS and Android: different text due to different screen behavior
Technical differences between platforms directly affect how the message should be structured.
| Parameter | iOS | Android |
|---|---|---|
| Visible text volume | 40-50 characters visible on the lock screen for the title (the rest is truncated) | slightly more, less rigid truncation |
| Behavior on the lock screen | notification disappears after phone unlock | notification remains visible in the strip until explicitly swiped away |
| Impact on CTR | first phone access — the only chance to grab attention | the user has several 'approaches' to one notification |
| Rich push with media | requires Notification Service Extension — without it only text will be sent | media content is easier to attach, without mandatory extension |
From this follows a practical rule: for iOS, the key idea should be squeezed into the first words of the title as tightly as possible — there may be no second chance. For Android, a more elaborate structure is acceptable, because the notification physically stays visible for longer.
Web push, SMS and in-app: different text logic
Three channels often write the same way, although the context of message delivery is fundamentally different for each.
| Channel | Context of delivery | Copywriting feature |
|---|---|---|
| Web push | browser is open, user is busy with another task | a maximally fast hook is needed, competition with website tabs |
| SMS | a separate messenger app, perceived as a more personal channel | text without emojis and caps, closer to a business message, link is mandatory and short |
| In-app push | user is already inside the app, there is context of their current action | can refer to what the person is doing right now, tone is more conversational |
The path from click to goal: where copywriting hasn't finished its work yet
It's a mistake to think that the text's job ends at the click. In fact, push copywriting continues to work afterwards — if the text promised specificity, but the landing page doesn't confirm it instantly, the user closes the app before the target action.
- The headline and first words — form the decision to open within 0,5-1 seconds of scanning.
- Click — transfers the expectation set by the text to the app screen; if the offer in the push was "20% discount on sneakers," but the main page opened — trust is lost instantly.
- The first screen after the click — must visually confirm the push text literally: the same product, the same figure, the same context.
- The target action — here, the push text no longer works, but the factor of "ability" from the Fogg model: the number of steps to purchase or subscribe.
Therefore, diagnosing low CTR needs to be separated from low conversion after the click — these are two different problems, which are treated with different tools: the headline text in the first case and simplifying the in-app scenario in the second.
Recent articles
MegaFon, BMW, VTB and more 240+ companies. See how it looks in the showcase reel:
Our branding agency is approached to order:
Branding agency services
