Most ecommerce push programmes fail before the first message is sent. The team treats the app as another broadcast channel, then wonders why opt-outs rise and revenue stays flat.

That is the wrong job for push. A good ecommerce mobile app notification brings a known customer back to a relevant product, basket, order, or offer in one tap. The message matches the customer’s situation. The app opens on the right screen. The result is measured against what would have happened without the message.

Here is the framework I would use with an ecommerce team starting from scratch. It focuses on the campaigns that deserve automation, the segments that matter, and the mistakes that quietly destroy the channel.

What an ecommerce push notifications app needs to handle

An ecommerce push notifications app should connect four things:

  • A customer signal, such as a product view, basket, purchase, or stock change.
  • A useful message that matches that signal.
  • A deep link to the next action.
  • Reporting that connects the notification to sessions, orders, margin, and opt-outs.

If the team still exports lists, builds every journey by hand, or sends every customer to the app home screen, the channel will not scale. Check permission states, audience rules, deep links, frequency caps, transactional updates, suppression after conversion, and campaign reporting before choosing a provider or app service.

Start with the customer action, not the message

“Send a push notification” is not a strategy. Decide what the customer should do next:

  • Finish a purchase.
  • Return to a product they viewed or saved.
  • Reorder something they buy regularly.
  • Use a relevant offer.
  • Discover a related product.
  • Check an order or delivery update.

Each action needs its own trigger, audience, message, destination, and success metric. This is how you stop a mobile app push notification programme becoming a calendar of random promotions.

The six campaigns to build first

1. Abandoned basket recovery

Basket recovery is the first automation I would ship. The customer has already shown purchase intent, so you do not need to manufacture interest.

Send one reminder while the basket is still relevant. Add a second message only when the margin and customer behaviour justify it. Stop the flow as soon as the order is placed. Do not lead with a discount by default. Fix the friction first and take the customer back to the basket, not the home screen.

For the wider funnel, read how a native ecommerce app can reduce cart abandonment.

2. Back-in-stock alerts

Back-in-stock alerts work because the customer has stated an intention. Send the alert for the exact product or variant. Deep-link to that product. Include a useful detail such as size, colour, or quantity.

Do not replace a specific alert with a generic category promotion. “The blue medium is back” gives the customer a reason to tap. “New products are waiting” does not.

3. Replenishment reminders

Replenishment suits skincare, pet food, supplements, coffee, and household products. Use purchase history to estimate the next buying window. Then let the customer reorder with as few taps as possible.

There is no perfect send time. Test a small window around the predicted date. Review results by product type. A single schedule across the full catalogue usually hides useful differences.

4. Price-drop alerts

Use price alerts for saved products, wish-list items, and repeated views. “The trainers you saved are now £20 less” is stronger than “Do not miss out today.” Add an expiry rule so the notification cannot outlive the offer.

5. New collection launches

Start with customers who bought, browsed, or saved a related category. Someone who buys running shoes may care about a new performance collection. They do not need every new product in the catalogue.

Give the campaign one job. That might be discovery, early access, or a specific launch. A clear reason to open beats a catalogue dump.

6. Post-purchase updates

Order confirmation, dispatch, delivery, and return updates are service messages, not promotions. They build trust and give customers useful reasons to open the app.

Keep service messages separate from promotional campaigns. Customers should be able to control those notification types independently.

Segment before you personalise

Adding a first name to a generic message is not personalisation. The useful question is: what does this customer need next?

  • New app users: installed but have not ordered.
  • First-time buyers: placed one order and need a reason to return.
  • Repeat customers: have a known purchase pattern.
  • High-value customers: may respond to relevant early access or better service.
  • Inactive users: stopped opening or buying through the app.
  • Category buyers: show a clear interest in one product area.

Exclusions matter just as much. Suppress customers who already bought the promoted product. Exclude people who received the same campaign recently. Remove customers who changed their permission status. A relevant message sent too often becomes a reason to uninstall.

Deep links decide whether the campaign earns money

The click experience is part of the notification. Send basket reminders to the basket, stock alerts to the exact variant, reorder reminders to the repeat product, order updates to tracking, and launch campaigns to the relevant collection.

A slow webview, expired product page, or generic home screen wastes the intent created by the message. Deep links also make reporting cleaner. You can connect the notification, destination, session, product, basket, and order.

The same rule applies to email and SMS. See why deep links stop dropping mobile buyers into Safari.

Use a sending framework before you increase volume

Set frequency caps

Set a maximum number of promotional messages per customer over a rolling period. Start conservatively. Then adjust using opt-outs, conversions, revenue per recipient, and margin. Keep transactional updates under separate rules.

Use behaviour to choose timing

There is no universal best time for ecommerce push notifications. Use the customer’s local time, previous engagement, and likely purchase cycle. A replenishment reminder should not go to every customer at the same hour if the data supports a better window.

Write for the lock screen

Use a clear product or action. A simple structure is context + value + action.

“Your daily moisturiser is nearly due for a refill. Reorder in two taps.”

“The jacket you saved is back in stock in your size. View it in the app.”

The copy should make sense before the customer opens anything.

Measure incremental revenue, not just open rate

Open rate helps diagnose the message and timing. It does not prove that the campaign created a sale.

Track:

  • Eligible customers.
  • Delivery rate.
  • Open rate and app sessions.
  • Products viewed and baskets created.
  • Orders and revenue.
  • Margin after discounts.
  • Unsubscribes and opt-outs.
  • Revenue per recipient.

Use a holdout group when possible. Compare customers who received the campaign with similar customers who did not. This separates incremental orders from purchases that would have happened anyway.

Also check channel shift. A customer opening a push and buying what they would have bought through email is not the same as a new order.

A 30-day rollout plan

Week 1: Fix the foundations

  • Confirm permission and opt-out handling.
  • Map every campaign to a destination screen.
  • Define conversion, order, margin, and revenue events.
  • Set frequency and suppression rules.

Week 2: Launch high-intent automations

Start with basket recovery, back-in-stock alerts, and order updates. Keep the first versions simple. You need a clean baseline before adding complexity.

Week 3: Add repeat-purchase triggers

Identify products with a repeatable cycle. Build reorder audiences. Test timing around the predicted repurchase window.

Week 4: Review incrementality

Compare campaign revenue with a holdout group. Review margin after discounts. Check opt-outs by segment. Keep only campaigns with a clear job.

Common mistakes that kill ecommerce app push performance

  • Sending to everyone: broad reach creates broad fatigue.
  • Using discounts as the default: customers learn to wait for a coupon.
  • Ignoring app permissions: a large install base is useless if few users can receive the message.
  • Opening the home screen: every extra tap costs intent.
  • Counting attributed revenue as incremental: last-click reporting overstates the value of push.
  • Mixing service and promotion: customers lose control over the notifications they actually need.

FAQ

What is the best first ecommerce mobile app push notification?

Start with a high-intent trigger, usually abandoned basket recovery, back-in-stock, or an order update. These messages have a clear reason to exist and a clear destination.

How often should an ecommerce app send push notifications?

There is no safe universal number. Set a conservative promotional cap, separate service messages, and adjust it using opt-outs, incremental revenue, and margin by segment.

Should ecommerce push notifications include discounts?

Not by default. Test the message without a discount first. Use an incentive only when it solves a real conversion problem and the margin supports it.

What good looks like

A strong ecommerce mobile app push programme is not the one that sends the most messages. It gives customers useful reasons to return to the app.

Start with high-intent events. Segment around real behaviour. Deep-link to the next action. Measure completed orders and incremental revenue. Expand only where the data supports it.

ConvertNative provides a fully managed native iOS and Android app with targeted push notifications, platform integrations, and 360° analytics. If your mobile site attracts traffic but not enough repeat revenue, see how ConvertNative can help you build and grow the app.