Every article about Adobe Experience Cloud and Adobe Commerce says the same thing: unified customer profiles, real-time personalisation, AI-driven insight, seamless omnichannel experiences. All of it is true. None of it helps you decide.
So here is the version with the decision in it.
What the pieces do
Adobe Experience Platform, and Real-Time CDP on top of it, is the profile layer. It ingests events and records from wherever you have them — the store, the POS, email, support, offline — and resolves them into one customer profile with segments you can act on. This is the piece that makes the rest worth having.
Adobe Target does the acting. Rules and audiences decide what a given visitor sees, and it runs the A/B and multivariate testing that tells you whether the decision was right.
Adobe Experience Manager manages the content those experiences are made of, across web, email and app.
Adobe Commerce contributes catalogue, cart and order data, and consumes the segments coming back.
Adobe Sensei is the AI layer threaded through all of it. Worth noting that the branding has moved on: Adobe now talks about Sensei GenAI and product-level AI features rather than Sensei as a single umbrella, so a document that says only "Sensei" is showing its age.
What it takes to run
This is where the brochures stop.
It is not one purchase. Experience Platform, Target and AEM are separate products with separate licences, quoted per customer. The stack is priced for enterprises, and the number is a conversation with sales rather than a page you can read.
Someone has to feed it. Real-Time CDP resolves profiles from the data you send it. Getting the POS, the email platform, the support desk and the store into a consistent schema is an integration programme, not a connector. The profile quality you get out is exactly the data hygiene you put in.
Someone has to run it after launch. Audiences drift. Test results need reading and acting on. Content has to be produced for every variant you decided to serve. A personalisation stack with nobody operating it degrades into an expensive way of showing everyone the same page.
The value scales with volume. A 2% conversion lift is transformative on large revenue and a rounding error on small revenue. The arithmetic decides this, not the technology.
When it is the right answer
When you already hold real customer data across several channels and have no single view of it. When traffic is high enough that a test reaches significance in days rather than months. When running experiments is somebody's actual job and you can produce the content the winning variants need.
Meet those three and the stack earns its licence. It is a strong toolkit, and nothing in open source comes close on profile resolution at scale.
When it is not
For a single-channel store doing solid but not enormous volume, most of the benefit is available for a fraction of the money and effort. Adobe Commerce's own Product Recommendations and customer segments cover a lot of ground. Decent analytics and a lighter testing tool cover much of the rest.
Buying the full stack before you have the data to feed it or the team to operate it is the common failure mode. The tools sit configured and unused, the licence renews, and the personalisation programme takes the blame for what was really a resourcing decision.
The question worth asking first
Not "should we adopt Experience Cloud" but: what would we personalise tomorrow, for whom, using data we already hold?
If that has a concrete answer, the tooling conversation is easy. If it does not, no platform will supply one, and the cheapest way to find out is a small experiment on what you already have.
If you would rather talk it through with somebody who has no licence to sell you, get in touch.
