In this blog
Analysts open Customer Journey Analytics and feel at home — it is Analysis Workspace, the tool they already know. That familiarity hides how much changed underneath. This comparison explains what actually differs between Adobe Analytics and CJA, what you gain and lose in the move, and how to time it like an engineering decision instead of a leap of faith.
The architectural difference that drives everything
Adobe Analytics is a processing pipeline: hits arrive, processing rules fire, values land in report suites under a fixed vocabulary of eVars, props and events. CJA is a query layer: data lands as datasets on Adobe Experience Platform under XDM schemas, and CJA reads whatever the schema holds, joined across datasets on stitched identities.
Every practical difference follows from that swap — the flexibility, the cross-channel power, the different cost model, and the migration work.
Capability comparison
| Dimension | Adobe Analytics | Customer Journey Analytics |
|---|---|---|
| Data scope | Digital (web, app) into report suites | Any dataset on AEP — digital, call center, POS, CRM |
| Identity | Cookie/ECID within digital | Stitched cross-channel identity (field or graph) |
| Variables | eVars, props, events — allocated slots | Any schema field; no slot rationing |
| Cardinality | Limits, "(low traffic)" buckets | Effectively unlimited |
| Retroactivity | Processing decisions largely fixed at collection | Many decisions changeable at query time (sessions, attribution) |
| Real-time | Mature real-time reports | Near-real-time; batch-friendly, improving |
| Marketing channels | Built-in channel processing | Rebuilt via schema and data views |
| Analysis UI | Analysis Workspace | Analysis Workspace (same skills) |
| Cost basis | Server-call volume | Rows ingested/retained + platform entitlements |
What you genuinely gain in CJA
Cross-channel truth. Fallout from web visit → support call → branch purchase as one report. Attribution crediting digital for offline outcomes. This is the headline, and for multi-channel enterprises it is transformative.
Freedom from variable rationing. No more allocation meetings for eVar slots, no cardinality collapse on your biggest dimensions. Schema fields are the vocabulary, and the vocabulary is extensible.
Query-time flexibility. Session definitions, attribution models and even some data corrections apply retroactively at analysis time — decisions that were permanent in Adobe Analytics become adjustable lenses in data views.
A platform trajectory. New Adobe analytics investment lands in CJA first. Adopting it aligns you with where the roadmap is going rather than where it has been.
What you give up or must rebuild
Honesty ranks: Adobe Analytics' real-time reporting remains more mature for launch-day monitoring. Marketing channel processing — years of tuned rules — must be redesigned in CJA's model rather than copied. Segments and calculated metrics built on eVars need curated remapping. And numbers will not match one-to-one: sessionization and attribution differences guarantee variance that must be explained to stakeholders once, properly, with a reconciliation document — or it will be relitigated forever.
When to move: a readiness test
Move when three things are true:
- Cross-channel questions carry business value — you can name decisions that joined data would change
- Identity keys exist — some common identifier (customer ID, hashed email) already spans your systems
- You can fund a real transition — schema design, implementation, parallel-run reconciliation and analyst enablement
If none are true, staying on Adobe Analytics for now is a defensible decision, not a failure — we advise it regularly. What we advise against is drift: no decision, aging implementation, and a migration eventually forced under deadline instead of chosen under control.
What analysts notice in the first month
The lived transition is smoother than the architecture diagram suggests, with three recurring moments. First, relief: Workspace muscle memory works on day one — tables, flows, fallout all behave. Second, disorientation: the component list shows schema fields with unfamiliar names instead of the eVars they memorized, which is why data-view curation and naming matter so much before analysts arrive. Third, the unlock: the first time a fallout report crosses from web behavior into call-center outcomes, the room goes quiet — that report was impossible last quarter. Plan enablement around those three moments: reassure with the familiar, smooth the renaming with curated views, and stage the cross-channel demo early enough to convert skeptics while goodwill is high.
A transition pattern that works
Run a phased coexistence: keep Adobe Analytics as the system of record while CJA stands up on the same Web SDK collection plus one or two offline datasets. Reconcile core metrics and document deltas. Move one analyst team's workflows entirely into CJA and let them harden the data views. Then declare the switch — CJA becomes record, Adobe Analytics enters read-only wind-down. Enterprises that skip the declaration step end up running both indefinitely, paying twice and trusting neither.
The deeper platform fundamentals are in What is Customer Journey Analytics, and the build sequence in the implementation guide. For help pressure-testing your timing — including honest reasons to wait — talk to a consultant or start with a readiness audit.
Frequently asked questions
What is the main difference between Adobe Analytics and CJA?
The data model. Adobe Analytics processes hits into report suites with eVars and props; CJA reads Experience Platform datasets under XDM schemas, joined on stitched identities. That change unlocks cross-channel analysis and removes cardinality limits.
Will my Adobe Analytics numbers match CJA?
Not exactly, and they should not be expected to. Session definitions, attribution defaults and processing differ. Plan a reconciliation phase: document the deltas, agree tolerances, and declare CJA the system of record once variance is understood.
Do my Workspace projects and segments transfer to CJA?
Workspace skills transfer almost entirely; artifacts partially. Adobe provides migration tooling for projects, but segments and calculated metrics built on eVars need remapping to schema fields — treat it as a curated rebuild, not a copy.
Is CJA more expensive than Adobe Analytics?
The models differ: Adobe Analytics prices on server calls, CJA on rows ingested and retained plus platform entitlements. Data curation controls CJA cost meaningfully. Bundled Experience Platform deals often change the comparison — model your three-year total.
Can we run Adobe Analytics and CJA at the same time?
Yes, and most enterprises do during transition — often feeding both from the same Web SDK collection. Coexistence with a declared end state avoids the two-sources-of-truth trap.