Analyze Customer Journey Analytics data with Coworker Chat
Adobe CX Enterprise Coworker Chat can perform advanced data analysis that was previously possible only in Analysis Workspace. Coworker Chat accesses data from your Customer Journey Analytics data views, allowing you to explore that data and get answers to natural-language prompts.
You can use Coworker Chat in two ways, depending on how much analysis you need:
- Quick answers - Ask a direct, plain-language question and get an immediate answer. Business users often use Coworker Chat this way, and analysts use it too when they need a fast answer for a stakeholder.
- Deep thought work - Have an extended, multi-turn conversation with Coworker Chat to investigate a business problem, rule out causes, and arrive at a recommendation. Analysts typically use this approach to explore data in depth before making a recommendation.
Before you begin, learn the Coworker Chat interface and configuration options, then ensure Coworker is connected to Customer Journey Analytics and the relevant data view.
Get started with Coworker Chat
Data access and permissions
Coworker Chat inherits permissions from Customer Journey Analytics. You can access only those data views, dimensions, metrics, and segments available to you in Analysis Workspace.
Interface and configuration options
Before you use Coworker Chat with your Customer Journey Analytics data, learn how to sign in and manage configuration options for the following features:
- Chat inputs
- Conversations
- Marketplaces
- MCP servers
- Memory
- Plugins
- Skills
- And more
For more information, see the Coworker Chat UI guide.
Best practices when analyzing data with Coworker Chat
Organization-level best practices
-
Appoint an analyst from your organization as a Coworker champion.
-
Create a library of vetted prompts and skills that correlate with the data and components that are available to users.
-
Create one or more skills that direct Coworker Chat to use only those components that you want used in analyses. This helps Coworker Chat give users in your organization the most relevant data.
-
Educate users on when to ask Coworker Chat for a quick answer versus when to use it for deep thought work.
User-level best practices
-
Use plan mode.
This mode is especially useful for complex tasks, but can also yield better results for simple tasks because it allows Coworker to ask follow-up questions before acting. For more information, see Plan mode.
-
When creating a prompt, be as specific as possible:
- Name the dimensions, metrics, and date range you want analyzed.
- Reference data view components by their exact name.
- Specify any segments, audiences, channels, or devices you want included, excluded, or compared.
- State whether you want a specific visualization type, such as a funnel, trend, or cohort table.
- Ask for recommended next steps if you want Coworker Chat to suggest follow-up questions.
- Ask for a forecast horizon, such as “next 30 days,” when projecting metrics.
- Mention any hypothesis you already have, so Coworker Chat can validate or rule it out.
- Ask for the contributing dimensions if you want a breakdown of a metric change.
- Specify the audience for a summary, such as leadership or the marketing team, and request a slide deck outline if you plan to present the findings.
- Name the specific report suite and data view you want to compare when validating data.
- Complete an analysis first, then ask Coworker Chat to save it as a skill, giving it a clear, descriptive name and noting how often you plan to reuse it.
-
Add standard directions to the Coworker Chat memory. For example, if you always use data from the same data views, add that to the memory.
Verify that Coworker Chat is connected to Customer Journey Analytics
In Coworker Chat, verify that Coworker is connected to Customer Journey Analytics:
-
Select the MCP icon in the left rail, then make sure that cja-mcp is available in your list of connected MCP servers.
-
(Conditional) If cja-mcp is not yet connected, select Add MCP Server, specify cja in the Server Name field and select it when it appears, then select Add Server.
Connect to the right data view
A data view is a container in Customer Journey Analytics that determines how data is interpreted.
You might have access to various data views in Customer Journey Analytics, each containing different dimensions and metrics that Coworker can use when analyzing data.
Decide which data views you want to use
Tell Coworker the types of questions you want answered, and ask it which data views you have access to that provide that information. You can also set your data view as a preference in memory.
You:
I’m interested in learning where customers drop off in the customer journey. Which data views in Customer Journey Analytics do I have access to that would be able to answer this question for me?
Coworker Chat response:
You have access to three data views. The Customer lifecycle data view contains the following dimensions and metrics, which would be best for answering your question.
You:
Great, let’s use that data view.
Coworker Chat response:
Okay, I’ll use the Customer lifecycle data view to answer future questions in this chat session.
Add a data view preference in Memory
Coworker Chat contains a memory feature that allows you to provide it with access to information that spans across all chats. It’s a good practice to add your preferred data views as preferences in Coworker’s memory.
-
In Coworker Chat, in the left nav, select the Memory icon.
-
On the Memory page, in the Stored preferences section, specify one or more data views that you want Coworker Chat to use in your chats.
Analyze in Customer Journey Analytics
After Coworker creates a visualization, you can open it in Analysis Workspace for deeper analysis and granular control. The visualization opens in a new Analysis Workspace project in Customer Journey Analytics.
To open a visualization in a new Analysis Workspace project:
-
Select Analyze in CJA next to any visualization that is created in Coworker.
-
With the visualization open in Customer Journey Analytics, you can use the Analysis Workspace drag-and-drop browser interface to make modifications, further craft your analysis, create an audience, and much more. You can even share your Workspace project with anyone you choose.
For more information about Analysis Workspace, see Analysis Workspace overview.
Use cases for Customer Journey Analytics
You can see Customer Journey Analytics use cases and sample prompts that practitioners are using in Adobe CX Enterprise Coworker Chat, from quick answers to deep thought work investigations. Each prompt is built to be copied, adapted with your own data and context, and refined through conversation.
For more information, see Use cases.
Analytics skills
The following skills are available for analyzing Customer Journey Analytics data.
Query and analyze data
This skill (cja) lets you query Customer Journey Analytics in real time and analyze the results without building the request yourself in Analysis Workspace.
Required permissions
- View access to the data view you want to query
Key use cases
- “Show me page views for the last 30 days”
- “List top segments in the master data view”
- “Compare revenue by channel month over month”
- “How does mobile vs desktop conversion look this quarter?”
- “Walk me through the checkout funnel”
- “Show conversion funnel from PDP to purchase”
- “Forecast sessions for the next 30 days”
- “Are we on track to hit our revenue goal?”
In scope
- Real-time querying of metrics, dimensions, segments, and data views
- Side-by-side comparisons across channels, time periods, or segments
- Multi-step funnel and fallout analysis
- Metric forecasting based on historical trends
Out of scope
- Creating or editing data view components
- Data outside the data views you have access to
- Predictive modeling beyond metric forecasting
Root cause analysis
This skill (cja-root-cause-analysis) investigates why a metric changed instead of just reporting that it changed.
Required permissions
- View access to the data view being analyzed
Key use cases
- “Why did conversions drop last week?”
- “What caused the revenue spike on Jan 15?”
In scope
- Investigating a change in a known metric over a known period
- Surfacing the dimensions and segments that contributed to the change
Out of scope
- Detecting anomalies you haven’t asked about (no automated or real-time alerting)
- Root cause analysis for metrics outside a data view you have access to
Executive summaries and performance digests
This skill (cja-executive-summary) produces stakeholder-ready summaries of your Customer Journey Analytics data.
Required permissions
- View access to the data view or data views covered in the summary
Key use cases
- “Give me an executive summary of last month”
- “Create a slide deck outline from this quarter’s data”
In scope
- Summarizing performance over a specified period
- Generating prescriptive recommendations based on the data
- Outlining content for a slide deck or stakeholder readout
Out of scope
- Building the final slide deck or presentation file
- Summaries that span data views you don’t have access to
Data validation with Adobe Analytics
This skill (aa-cja-validation) compares, audits, and reconciles data between Adobe Analytics and Customer Journey Analytics.
Required permissions
- View access to the Adobe Analytics report suite and the Customer Journey Analytics data view being compared
Key use cases
Compare, audit, and reconcile data between Adobe Analytics and Customer Journey Analytics.
For more information, see Validate data with Coworker when upgrading from Adobe Analytics to Customer Journey Analytics.
- “Compare my Adobe Analytics report suite to my Customer Journey Analytics data view”
- “Validate page views between Adobe Analytics and Customer Journey Analytics”
In scope
- Comparing metric values between a report suite and a data view
- Flagging discrepancies between the two data sources
Out of scope
- Resolving the underlying cause of a data discrepancy
- Validating data sources other than Adobe Analytics and Customer Journey Analytics
Create custom skills
This skill (cja-skill-creator) turns an analysis you’ve already run into a reusable skill that persists across sessions.
Required permissions
- Manage skills
Key use cases
- “Turn this weekly revenue analysis into a reusable skill”
- “Save this as a skill for monthly funnel reporting”
In scope
- Converting a completed analysis into a named, reusable skill
- Making a saved skill available across your future chat sessions
Out of scope
- Sharing a saved skill with other users automatically (organization-level skill libraries require admin setup)
- Editing the data view components a skill references
Example: Find where customers drop off
You can ask Coworker Chat to use your data to analyze any business question.
For example, as a marketing manager, merchandiser, or growth lead, you might want to understand where customers abandon the path to purchase, why they did, and what can be done to drive higher conversion rates.
The types of questions you can ask
You don’t need to know what a “fallout visualization” is, or which metric goes in which slot. Instead, describe the outcome you care about in plain language:
- “Where are we losing people on the way to purchase?”
- “What’s my full conversion funnel, from product view to checkout?”
- “Which marketing channels leak the most between cart and purchase?”
- “Is cart abandonment worse on mobile or desktop?”
- “Do people from paid social convert differently than people from email?”
The conversation
1. Start with the business question
You:
Where are we losing people on the way to purchase?
Coworker Chat response:
Notice that Coworker Chat chooses the right visualization, sequences the touchpoints, and even adds recommended next steps.
This is a funnel question — let me run the conversion funnel to see where drop-off happens on the path to purchase.
Here’s where you’re losing people:
The big finding: cart to purchase is your biggest leak
28,463 people add items to their cart but never complete the purchase. That’s an 83.6% cart-to-purchase drop-off rate — by far the largest single point of failure.
What this tells you about priorities:
Recommended next steps:
- Break down cart abandonment by marketing channel — are certain channels bringing low-intent traffic that adds to cart but never buys?
- Look at cart abandonment by device type — is mobile checkout friction killing conversions?
- Identify which products get abandoned most — pricing issue? Shipping surprise?
Which of these would you like me to dig into next? The channel breakdown would tell you directly which marketing actions are bringing you “almost-buyers” versus actual buyers.
2. Drill into the leak
You:
Okay, let’s break down by marketing channel
Coworker Chat response:
There’s a custom Marketing Channel dimension. Let me pull the funnel by that, then calculate the per-channel conversion rates and identify where each channel leaks.
The verdict: the leak is not channel-specific
This is a really revealing finding. Here’s the per-channel funnel breakdown:
Key takeaways:
- The cart-to-purchase drop-off is uniformly bad across all channels (83-84%). This means the problem isn’t about traffic quality — every channel is converting people into the cart at similar rates. The leak is in your checkout experience itself, not your marketing.
- No channel is significantly better or worse at converting. Overall conversion rates cluster tightly between 3.8% and 4.2%, and revenue per session is nearly identical ($495-$512).
- Where the volume impact matters most: paid search loses the most people in absolute terms (9,015 cart abandonments), followed by organic search (6,449 lost). These are the biggest dollar-value opportunities because they drive the most traffic.
What this tells you strategically
The marketing channels are doing their job — they’re all getting people interested and adding to cart at similar rates. The problem is downstream of marketing: your checkout and payment flow is where the money is leaking.
Recommended next investigations:
- Break down mobile cart abandonment by product category — is it worse for high-price items (suggesting mobile payment trust issues)?
- Look at which marketing channels drive the most mobile traffic — are you spending paid media dollars to send people into a broken mobile funnel?
- Check if there’s a page or site-section signal — can we see which checkout pages mobile users hit last before abandoning?
Which angle would you like to explore next?
3. Keep investigating
You:
Interesting. Let’s try breaking down by device type.
Coworker Chat response:
Found it: mobile checkout is broken
The funnel is identical across devices until checkout — then mobile falls off a cliff:
The smoking gun
The browsing and add-to-cart experience is fine on mobile (identical ~45% browse rate, ~55% add-to-cart rate). People are equally interested and engaged. But something in the mobile checkout flow is killing conversions.
The revenue opportunity
If mobile checkout converted at the same rate as desktop, you’d gain ~1,991 additional purchases — a 35% increase in total purchase volume from fixing one experience.
What’s likely happening
The data pattern (equal engagement, catastrophic checkout drop) is a classic signal for:
- Payment form friction on small screens (autofill issues, form layout)
- Unexpected costs revealed at checkout (shipping, tax) causing sticker shock
- Guest checkout not available or a login wall on mobile
- Page load or timeout issues during payment processing on mobile
Recommended next steps:
- Break down mobile cart abandonment by product category — is it worse for high-price items (suggesting mobile payment trust issues)?
- Look at which marketing channels drive the most mobile traffic — are you spending paid media dollars to send people into a broken mobile funnel?
- Check if the “holiday_flash_sale” campaign had better checkout conversion — urgency-driven campaigns might overcome checkout friction
Which would you like to explore, or is this enough evidence to take to your UX and checkout team?
4. Turn it into a decision
You:
So what should I do?
Coworker Chat response:
Based on what the data shows, here’s where I’d focus:
- Fix the mobile checkout — it’s not a channel problem. Every channel is converting people into the cart at a similar, healthy rate, so the ad spend is working across the board. The revenue is being lost specifically in the mobile checkout flow, which is where the fix belongs.
- Start with payment and form friction. These are the most common reasons high-intent mobile shoppers abandon a cart. Offering mobile wallet options and shortening the form are usually the fastest wins.
- Re-check after the fix. Once checkout changes ship, watch whether mobile’s cart-to-purchase rate moves toward the desktop rate of 23.2%.
Want me to save this as a project so you can track it, or set an alert if mobile cart-to-purchase drops below a threshold?
What happened
In four plain-language questions, Coworker helped us:
- Build a multi-step conversion funnel and flag cart-to-purchase as the biggest leak
- Rule out marketing channel as the cause — every channel leaked at nearly the same rate
- Isolate the real problem to mobile checkout, and quantify the fix at a 35% lift in purchases
- Walk away with a specific fix to prioritize: mobile payment and form friction. This is benchmarked against desktop’s 23.2% conversion rate