If you have used Adobe Analytics for years, Customer Journey Analytics looks superficially like the same product with a different logo on the login screen. Analysis Workspace is there, freeform tables work the way you expect, segments look familiar. That resemblance is misleading, and treating CJA as “Adobe Analytics, refreshed” is the single most common reason migrations go badly.
They share an interface. Underneath, they are architecturally different products with different data models, different limits and a fundamentally different philosophy about when decisions get made. This post covers what actually differs and what it means in practice.
The one difference everything else follows from
Adobe Analytics makes its decisions at collection time. Customer Journey Analytics makes them at report time.
In Adobe Analytics, the processing that shapes your data happens on the way in, through VISTA Rules, Processing Rules and Marketing Channel Processing Rules. Once data lands in a report suite, it is essentially set. Get an eVar’s expiry wrong and the historical data carries that mistake permanently; the only remedy is fixing the implementation and waiting for new data to accumulate.
CJA works the other way. Data lands in Adobe Experience Platform in its raw form, and the interpretation happens afterwards through report-time transformations. Change your mind about an attribution window, a dimension’s persistence, or how a field should be bucketed, and you change a setting rather than re-tagging a site and waiting two quarters.
Adobe describes these alterations as non-destructive, and that word carries the whole argument. In CJA the raw data is never modified; a data view is a lens over it. You can hold several lenses over the same dataset simultaneously, which is impossible in the Adobe Analytics model.
Data views versus virtual report suites
Data views are CJA’s central concept, and Adobe positions them as the equivalent of virtual report suites, with considerably more power. A data view defines which components are available, what they are called, how attribution behaves and how values persist. Because it is applied at report time, two teams can look at the same underlying data through different, equally valid views.
The practical consequence is worth stating plainly: finance and marketing can each have a data view matching their own definition of a conversion, from one dataset, without either being wrong or anyone re-implementing anything. In Adobe Analytics that argument gets resolved by whoever controls the processing rules.
The limits disappear
Anyone who has managed a mature Adobe Analytics implementation has felt the variable ceiling. Adobe Analytics supports a maximum of 75 props and 250 eVars, and in a large organisation those get allocated, argued over, recycled and documented in a spreadsheet that is always slightly out of date.
CJA has unlimited customer dimensions and metrics, no cardinality restriction on unique values within a dimension, and unlimited string arrays replacing listVars. The high-cardinality dimensions that were painful in Adobe Analytics, product IDs, search terms, content IDs, stop being a capacity planning exercise.
The discipline does not disappear, it moves. Instead of rationing variables you are governing an XDM schema, and a badly designed schema is just as capable of producing an unusable dataset as a badly allocated eVar plan. The constraint changed shape; it did not vanish.
Identity: the actual reason most companies move
Adobe Analytics is fundamentally device-centric, scoped to a single report suite and a device graph. It answers “what happened on this browser” extremely well and struggles with “what did this person do across their phone, laptop, the call centre and a shop”.
CJA is built around identity. It offers a seamless combination of device-specific datasets from unauthenticated and authenticated sessions, with historical backfill, and supports graph-based stitching through Adobe’s Identity Service. That backfill point matters more than it sounds: when a user finally logs in, their earlier anonymous behaviour can be retrospectively associated with them, so the pre-login research that led to a purchase stops being invisible.
Because CJA accepts any dataset in XDM rather than only web and mobile, that identity spine can also carry CRM records, call centre logs, point-of-sale transactions and subscription events. This is the capability people are really buying.
Features that exist only in CJA
- Forecasting: time-series prediction inside Workspace.
- Guided analysis: structured, self-serve workflows for common questions, aimed at people who will never build a freeform table.
- Intelligent Captions: generative summaries of what a visualisation is showing.
- BI Extension: connecting BI tools directly to CJA data.
- SQL access via Data Distiller, which for a data team is often the headline: your analytics data stops being trapped behind a reporting UI.
- HIPAA readiness with enhanced security options, relevant if you are in healthcare and previously could not use Adobe Analytics at all.
- Lift and confidence across any data source, rather than being limited to Analytics for Target.
What you give up, and the honest trade-offs
This is the part vendor material skips.
- Alerting is coarser. CJA supports daily, weekly and monthly alerts rather than hourly, a direct consequence of how data arrives. If you rely on hourly anomaly alerts to catch a broken checkout, that workflow needs rethinking.
- You are adopting Experience Platform, not just a reporting tool. CJA sits on AEP, which means schemas, datasets, identity configuration and a genuinely different operating model. Organisations that treat this as an analytics migration rather than a platform adoption underestimate it substantially.
- The skills are different. Your Adobe Analytics expert knows eVar allocation and processing rules. CJA rewards people who think in schemas, identity graphs and data modelling. There is overlap, but it is not the same job.
- Cost and complexity go up before value does. The flexibility is real, and so is the ramp.
Where CJA sits now
Adobe has been consolidating its analytics naming, and Customer Journey Analytics now sits within the Adobe CX Analytics portfolio introduced at Summit 2026, alongside Adobe Analytics itself, a CJA B2B Edition, Marketing Campaign Analytics and Content Analytics. The strategic signal is clear enough: CJA is the centre of gravity, and Adobe Analytics is the mature product being carried forward rather than the one attracting investment.
Which one do you actually need?
Adobe Analytics is still the right answer if your questions are genuinely about web and app behaviour, your implementation is mature and working, and nobody is asking for offline or CRM data in the same report. It is fast, well understood, and not going away. Migrating a well-run implementation purely because CJA is newer is a large amount of work for very little.
CJA earns its cost when the questions span channels and identities: you have meaningful offline or call-centre interactions, you need one profile across devices and logged-in states, your eVar allocation is a constraint you actively manage, or you want SQL-level access to your own behavioural data.
The test we use with clients is a single question: are the analyses you cannot currently do limited by the tool, or by the data you collect? If it is the data, CJA will not rescue you, and the AEP work will surface every gap you already had. If it is genuinely the tool, CJA removes a real ceiling.
If you are weighing that decision, or partway into a migration that feels heavier than it was sold as, talk to us. We have done this from both sides.
