---
title: Markin + Airship: deciding what deserves the notification
url: https://markin.ai/compare/markin-and-airship
kind: stack
description: Airship delivers mobile-first journeys across push, in-app, SMS and wallet. Markin decides which opportunity is worth the interruption, and proves it against a holdout.
updated: 2026-09-03
---

# Markin + Airship: deciding what deserves the notification

> Airship is a mobile-first customer experience platform: push, in-app, SMS, email, wallet and app experiences, with AI that helps design journeys. Markin sits before it and decides which commercial opportunity justifies a customer's attention, and what it is worth. Airship still delivers the experience.

## In short

- Airship: How do we design and deliver this app experience or notification? Output: Delivered notifications and in-app experiences, journey state, engagement reporting.
- Markin, the decision + execution layer: Which opportunity justifies this customer's attention, and what is it worth? Output: A ranked, sized decision per customer,  including hold, written into Airship.
- Opt-out is a permanent cost, and it is rarely priced into the decision to send.
- Markin does not deliver notifications and does not manage channel permissions.

## What each layer is

**Airship**, A mobile-first experience platform: push notifications, in-app messages, Scenes, SMS, email, wallet and app experiences, with journey design and AI-assisted creation.

Answers: How do we design and deliver this app experience or notification?

**Markin, the decision + execution layer**, A layer that generates and sizes revenue opportunities per customer, ranks them on expected incremental margin, chooses the treatment and decides when to stay silent.

Answers: Which opportunity justifies this customer's attention, and what is it worth?

## Side by side

|  | Airship | Markin, the decision + execution layer |
| --- | --- | --- |
| Question it answers | How do we design and deliver this app experience or notification? | Which opportunity justifies this customer's attention, and what is it worth? |
| Primary input | Channel and opt-in state, tags and attributes, events, journey definitions, content. | Customer context, outcomes, margins, contact history, constraints, past experiment results. |
| Primary output | Delivered notifications and in-app experiences, journey state, engagement reporting. | A ranked, sized decision per customer,  including hold, written into Airship. |
| Usual owner | Mobile product and lifecycle marketing. | Growth, data science and revenue leadership. |
| How it's measured | Delivery, open and conversion rates, opt-out rates. | Incremental revenue and ARPU against a randomised holdout. |

## What stays unsolved when Airship is running well

Push is the cheapest channel to send and the most expensive to misuse. Faster journey creation increases supply; nothing on the delivery side decides how much of a customer's finite tolerance each notification deserves.

- Opt-out is a permanent cost, and it is rarely priced into the decision to send.
- Concurrent journeys are resolved by caps and eligibility rules rather than by expected value.
- Conversion after a notification includes revenue that would have arrived anyway.
- Commercial hypotheses that are not notifications,  pricing, onboarding, a broken deeplink, have no home in the platform.

## How the two run together

1. **Context in.** Markin reads context from the warehouse, billing and product systems, plus Airship channel state and engagement history.
2. **Decision.** Opportunities are generated and sized, ranked per customer on expected incremental margin, a treatment is chosen and a control group assigned.
3. **Activation back into Airship.** The decision is written as attributes or tags, or as an event that starts a Sequence, so Airship delivers it. Opt-in state, caps and channel rules stay in Airship.

## Airship's own decisioning layer

Products: Journeys AI, AI-Generated Journeys, Airship AI features

Airship is a mobile-first experience platform with AI that helps design and personalise journeys faster. Its AI generates and improves Airship's own journey objects; the commercial question of which opportunity is worth a customer's attention sits outside that scope.

**What it optimises**

- Journeys AI is presented as helping teams design smarter paths, personalise content and accelerate conversion outcomes without manual rules. [Vendor page: Airship, Orchestrate customer journeys](https://www.airship.com/platform/orchestrate-customer-journeys/)
- AI-Generated Journeys let teams answer guided prompts to generate draft journeys, for faster campaign creation. [Vendor docs: Airship docs, AI-Generated Journeys](https://www.airship.com/docs/whats-new/2024-10-31-ai-generated-journeys/)

**Documented constraints**

- Generated output consists of Airship's own constructs,  Sequences, Scenes and In-App Automations, so the AI operates on Airship's journey model rather than on the wider commercial estate. [Vendor docs: Airship docs, AI-Generated Journeys](https://www.airship.com/docs/whats-new/2024-10-31-ai-generated-journeys/)
- Airship documents its AI as efficiency and personalisation features built on responsible-AI principles with human oversight, rather than as autonomous commercial decisioning. [Vendor docs: Airship docs, AI features](https://www.airship.com/docs/guides/features/intelligence-ai/ai/)

**Evidence**

- The published economic case is a Total Economic Impact™ study conducted by Forrester Consulting on behalf of Airship, based on interviews with a small number of enterprise customers,  a vendor-commissioned study rather than an independent benchmark. [Vendor-commissioned: Forrester TEI of Airship, commissioned by Airship](https://www.airship.com/resources/total-economic-impact/)

**Where Markin differs**

- **Prices the interruption.** Airship makes journeys faster to build. Markin decides whether a given customer's next notification is worth its opt-out risk, and often decides it is not.
- **Arbitrates beyond the app.** Push, in-app, email, service contact and a pricing change compete on one scale: expected incremental margin per customer.
- **Holdout-verified revenue.** Each decision carries a control group, so uplift is reported as incremental ARPU rather than as conversions after a send.

## What Markin does not replace

To be explicit about scope, because procurement will ask:

- Markin does not replace Airship push, in-app, SMS, wallet or Scenes delivery.
- Markin does not manage opt-in state, channel permissions or frequency caps.
- Markin does not replace your mobile SDK or app instrumentation.
- Markin does not design creative or in-app experiences.

## Where Markin fits

Markin does not deliver notifications and does not manage channel permissions. It decides which opportunity is worth a customer's attention and returns that decision so Airship can deliver the experience you already designed.

- **Attention treated as a budget.** Each customer has a finite tolerance. Markin allocates it to the action with the highest expected incremental value, and holds when nothing clears the bar.
- **One decision across channels.** A push, an in-app Scene, an email or silence are ranked together instead of being decided by whichever team owns the journey.
- **Measured on revenue, not sends.** Every decision carries a holdout, so the reported outcome is incremental ARPU, and opt-out cost is visible in the same read.

Next: [Next Best Action](https://markin.ai/solutions/next-best-action)

## When Markin is not the right answer

- Your notifications are purely operational, with no commercial choice to make.
- You cannot link app engagement to a revenue outcome.
- You want to replace mobile delivery infrastructure rather than decide better above it.

## FAQ

**Doesn't Airship's Journeys AI already optimise this?**

Journeys AI helps design and personalise journeys faster, and generates drafts made of Airship's own Sequences, Scenes and In-App Automations. That is supply. Markin decides demand: which opportunity is worth a customer's attention this week, priced in incremental margin, arbitrated against everything else and read against a holdout.

**Do we have to replace Airship?**

No. Airship remains the mobile delivery and experience layer. Markin supplies the decision it executes.

**How does the decision reach Airship?**

As attributes or tags on the channel, or as an event that triggers an existing Sequence. Opt-in state and caps continue to be enforced by Airship.

**Will this reduce our push volume?**

Usually, and deliberately. Where no action has positive expected value net of opt-out risk, the decision is hold. The objective is incremental ARPU, not send volume.

**What does the first ninety days look like?**

One revenue theme, one activation route into Airship, a real holdout, and an incremental number that survives a full measurement window.

**How is Markin different from the decisioning or AI already inside Airship?**

A decisioning engine ranks actions a human already defined, inside the campaign surface it was given. Markin forms the hypotheses itself,  marketing, product, pricing or a technical anomaly holding growth back, sizes them, executes them inside Airship and your product surfaces, and reads each one against a randomised holdout. It behaves like a data science and growth team, not like an optimiser.

**Does Markin only test messages and offers?**

No. Anything a human growth scientist would investigate is in scope: onboarding friction, feature adoption, pricing and packaging, dunning, and technical health issues such as a checkout error rate or a broken deeplink quietly killing conversion. Marketing is one of four hypothesis domains, not the boundary.

**What is the business case for adding Markin on top of Airship?**

On the assumptions preloaded above, 4.5M customers at 12 a month, a small move in ARPU is a large number in absolute terms, because it applies to the whole installed base every month rather than to a campaign. Across Markin deployments the verified range on treated cohorts is +17% to +35% ARPU against a randomised holdout. The point is not more messages: it is finding the highest-value action per customer, launching it, and proving it against control before it scales.

**How long before it pays for itself?**

First sized opportunities are in test within six weeks and the first holdout-verified result lands inside 90 days. Payback depends on your base, margin and programme cost, the calculator on this page computes it from your own numbers, after applying the 20% to 40% haircut BCG finds when next-best-action programmes are incrementality-tested.

Source: https://markin.ai/compare/markin-and-airship