Skip to content

[No QA] [Hold for feature] Update help docs for Releases 1-4: Reactive Concierge expense analysis#92468

Open
MelvinBot wants to merge 1 commit into
mainfrom
claude-resourceUpdatesReactiveAnalysisR1to4
Open

[No QA] [Hold for feature] Update help docs for Releases 1-4: Reactive Concierge expense analysis#92468
MelvinBot wants to merge 1 commit into
mainfrom
claude-resourceUpdatesReactiveAnalysisR1to4

Conversation

@MelvinBot
Copy link
Copy Markdown
Contributor

Explanation of Change

This updates the ExpensifyHelp docs to cover the new reactive Concierge expense analysis capability shipping across Releases 1–4 of the Upgrade VirtualCFO to support reactive analysis project. Previously the docs only described Concierge's proactive spend summaries; users can now also ask Concierge questions about their expense data and get data-driven answers in chat.

Changes:

  • How-Concierge-Analyzes-Spend.md
    • Updated the intro, description, and keywords to cover asking Concierge questions on demand (not just proactive insights).
    • Added a new ## How to ask Concierge about your expenses section with example questions and a note that complex questions are redirected to the Search page.
    • Noted that Concierge suggests follow-up questions on both proactive summaries and reactive answers (Release 3).
    • Updated the "Where" section and FAQ to document that questions can be asked in the #admins room (scoped to that workspace) and in your Concierge chat (cross-workspace, Release 4).
    • Updated the Insights comparison to note Concierge now answers natural-language questions while Insights provides visual dashboards.
    • Added two FAQ entries: what kinds of questions you can ask, and where you can ask them.
  • What-Concierge-Can-Do.md
    • Added a new ## What questions you can ask Concierge about your expense data section with examples, linking to the spend analysis article.

These changes bundle the help-site updates for Releases 1–4 into one PR per the request on the tracking issue.

Fixed Issues

$ https://github.com/Expensify/Expensify/issues/634606

Tests

// TODO: The human co-author must fill out the tests you ran before marking this PR as "ready for review"
// Please describe what tests you performed that validates your changed worked.

This is a documentation-only change. Review the rendered articles to confirm the content is accurate and matches the current product behavior.

Offline tests

N/A — documentation-only change.

QA Steps

// TODO: The human co-author must fill out the QA tests you ran before marking this PR as "ready for review".
// Please describe what QA needs to do to validate your changes and what areas do they need to test for regressions.

PR Author Checklist

  • I linked the correct issue in the ### Fixed Issues section above
  • I verified any copy / text that was added is grammatically correct and adheres to capitalization guidelines

Screenshots/Videos

N/A — documentation-only change.

Document that users can ask Concierge questions about expense data in
#admins rooms and Concierge chats (including cross-workspace questions),
and that Concierge suggests follow-up questions on spend analysis messages.

Co-authored-by: Stephanie Elliott <stephanieelliott@users.noreply.github.com>
@MelvinBot MelvinBot requested a review from a team June 3, 2026 03:34
@github-actions github-actions Bot changed the title Update help docs for Releases 1-4: Reactive Concierge expense analysis [No QA] Update help docs for Releases 1-4: Reactive Concierge expense analysis Jun 3, 2026
@OSBotify
Copy link
Copy Markdown
Contributor

OSBotify commented Jun 3, 2026

A preview of your ExpensifyHelp changes have been deployed to https://0c712a2b.helpdot.pages.dev ⚡️

Updated articles:

@stephanieelliott stephanieelliott marked this pull request as ready for review June 6, 2026 01:27
@stephanieelliott stephanieelliott self-requested a review as a code owner June 6, 2026 01:27
@melvin-bot
Copy link
Copy Markdown

melvin-bot Bot commented Jun 6, 2026

Concierge reviewer checklist:

  • I have verified the accuracy of the article
    • The article is within a hub that makes sense, and the navigation is correct
    • All processes, screenshots, and examples align with current product behavior.
    • All links within the doc have been verified for correct destination and functionality.
  • I have verified the readability of the article
    • The article's language is clear, concise, and free of jargon.
    • The grammar, spelling, and punctuation are correct.
    • The article contains at least one image, or that an image is not necessary
  • I have verified the formatting of the article
    • The article has consistent formatting (headings, bullet points, etc.) with other HelpDot articles and that aligns with the HelpDot formatting standards.
    • The article has proper text spacing and breaks for readability.
  • I have verified the article has the appropriate tone and style
    • The article's tone is professional, friendly, and suitable for the target audience.
    • The article's tone, terminology, and voice are consistent throughout.
  • I have verified the overall quality of the article
    • The article is not missing any information, nor does it contain redundant information.
    • The article fully addresses user needs.
  • I have verified that all requested improvements have been addressed

For more detailed instructions on completing this checklist, see How do I review a HelpDot PR as a Concierge Team member?

cc @kadiealexander

@melvin-bot melvin-bot Bot requested review from kadiealexander and removed request for a team June 6, 2026 01:27
@melvin-bot
Copy link
Copy Markdown

melvin-bot Bot commented Jun 6, 2026

@kadiealexander Please copy/paste the Reviewer Checklist from here into a new comment on this PR and complete it. If you have the K2 extension, you can simply click: [this button]

@stephanieelliott stephanieelliott changed the title [No QA] Update help docs for Releases 1-4: Reactive Concierge expense analysis [No QA] [Hold for feature] Update help docs for Releases 1-4: Reactive Concierge expense analysis Jun 6, 2026
@stephanieelliott stephanieelliott removed the request for review from kadiealexander June 6, 2026 01:27
@github-actions
Copy link
Copy Markdown
Contributor

github-actions Bot commented Jun 6, 2026

HelpDot Documentation Review

Overall Assessment

This PR expands the Concierge spend analysis article to cover a new capability: asking Concierge direct questions about expense data. It also adds a corresponding section to the "What Concierge Can Do" overview article. The changes are well-written, clearly structured, and largely compliant with HelpDot governance standards.

Scores Summary

  • Readability: 9/10 - Clear, scannable prose. Steps and examples are well-formatted. The examples list in the new "How to ask Concierge about your expenses" section is effective. Minor note: the intro paragraph now covers two workflows (proactive analysis + on-demand questions), which slightly increases cognitive load for the opening.
  • AI Readiness: 9/10 - Good task-based headings with feature names ("How to ask Concierge about your expenses", "What questions you can ask Concierge about your expense data"). YAML metadata is properly updated with new keywords and an expanded internalScope. Heading hierarchy uses only # and ## as required.
  • Style Compliance: 9/10 - UI terms like Insights are consistently bolded. Cross-links use descriptive anchor text and relative paths. Navigation references to #admins room are clear. No prohibited language detected.

Key Findings

  • The new section heading "How to ask Concierge about your expenses" is properly task-based and includes the feature name, aligning well with AI retrieval rules.
  • Cross-links are well-placed: they appear after explanatory sentences rather than inside procedural steps, per the cross-linking standards.
  • The YAML metadata updates (keywords, internalScope, description) accurately reflect the expanded scope and will aid search retrieval.
  • The FAQ additions ("What kinds of questions can I ask Concierge about my expenses?" and "Where can I ask Concierge about my expenses?") follow the question-word heading pattern correctly.
  • One consideration: the article now covers both proactive spend analysis and on-demand question-asking, which are arguably two distinct workflows. Per Section 1 of the authoring guidelines ("Solve one primary workflow only" and "If multiple workflows are detected, split into multiple articles"), this could warrant a separate article. However, the two capabilities are tightly related and the current structure handles them cohesively, so this is a judgment call rather than a strict violation.

Recommendations

  • Consider whether the on-demand question-asking workflow is distinct enough to warrant its own dedicated article, per the single-workflow-per-article rule. The current combined approach is reasonable but could be revisited if the feature grows.
  • The phrase "in a supported chat" in the new section ("send your question to Concierge in a supported chat") is somewhat vague -- it is clarified later in the "Where" section, but consider making it more specific inline or linking directly to the explanation to avoid ambiguity.

Files Reviewed

  • docs/articles/new-expensify/concierge-ai/How-Concierge-Analyzes-Spend.md - Expanded with new "How to ask Concierge about your expenses" section, updated metadata, revised comparison section, and two new FAQ entries. Well-structured.
  • docs/articles/new-expensify/concierge-ai/What-Concierge-Can-Do.md - New section added ("What questions you can ask Concierge about your expense data") with examples and cross-link. Clean addition that fits the existing article structure.

Note: Detailed line-by-line feedback has been provided as inline comments where applicable.


You can ask Concierge questions about your expense data and it will look up the answer for you, so you don't have to build a search or stitch results together yourself.

To ask, send your question to Concierge in a supported chat. Concierge reads the context of where you're chatting to scope the answer, so you don't need to repeat details like the workspace name.
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Naming and style: "a supported chat" is vague and does not reference the exact UI location. Per HELPSITE_NAMING_CONVENTIONS.md, always reference the exact UI label — avoid phrases like "the area" or "the section." Consider specifying the exact locations (e.g., "in the #admins room for a workspace or in your Concierge chat"), which is already described later in the article under "Where Concierge shares spend analysis insights." Repeating the specific locations here would be clearer and more consistent.

description: Learn how Concierge may proactively analyze company spend to share highlights and help you stay informed.
keywords: [New Expensify, Concierge, spend analysis, Concierge management report, insights, AI, expense trends, proactive insights, spend monitoring, spend analysis, flux analysis]
internalScope: Audience is Workspace Admins and Members. Covers how Concierge performs spend analysis and what insights to expect. Does not cover how to use Insights or generate custom reports.
description: Learn how Concierge proactively analyzes company spend to share highlights, and how to ask Concierge questions about your expense data on demand.
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

AI readiness / structure: The article title and description now cover two distinct workflows: proactive spend analysis and asking Concierge questions on demand. Per HELP_AUTHORING_GUIDELINES.md Section 1: "Solve one primary workflow only. If multiple workflows are detected, split into multiple articles." Consider whether the on-demand Q&A workflow should live in its own article (e.g., "How to Ask Concierge About Your Expenses") rather than being merged into this proactive-analysis article.


After Concierge answers, it may suggest a few follow-up questions to help you dig deeper.

Concierge currently answers direct lookup questions about your expense data. If a question is too complex to answer this way, Concierge will let you know and point you to the Search page so you can explore it yourself. [Learn how to use search operators to filter and analyze spend](/articles/new-expensify/reports-and-expenses/Use-Search-Operators-to-Filter-and-Analyze).
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Naming and style: "the Search page" references a UI destination but is not bolded. Per HELPSITE_NAMING_CONVENTIONS.md, UI element names (pages, tabs, sections) should be bolded — e.g., "point you to the Search page." Apply the same fix on the similar reference in the new FAQ answer near the end of the file.


Insights are shared periodically and may vary in timing, format, and level of detail.

You can ask Concierge your own expense questions in the #admins room for a workspace, or in your Concierge chat. In the #admins room, answers are scoped to that workspace. In your Concierge chat, questions can span all of the workspaces you have access to (for example, "Which workspace spent the most last month?" or "Show all my Travel expenses across all workspaces").
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Readability / structure: This paragraph is dense and covers multiple concepts (where to ask, scoping behavior per location, and multi-workspace examples). Consider breaking it into a short list for scannability, e.g.:

  • In the #admins room: answers are scoped to that workspace.
  • In your Concierge chat: questions can span all workspaces you have access to.

Also, this information largely duplicates what is already stated in the new "How to ask Concierge about your expenses" section (line 38). Consider consolidating to avoid redundancy, which can reduce retrieval precision per HELP_AUTHORING_GUIDELINES.md Section 6.


## What kinds of questions can I ask Concierge about my expenses?

You can ask direct lookup questions about your expense data, such as who spent what, spend by category, merchant, employee, or tag, comparisons across time periods, and the status of expenses or reports. If a question is too complex to answer directly, Concierge will point you to the Search page.
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Naming and style: Same as earlier — "the Search page" should bold the UI element name: "Concierge will point you to the Search page."

---

## What questions you can ask Concierge about your expense data

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

AI readiness / structure: This section largely duplicates the content in the new "How to ask Concierge about your expenses" section of How-Concierge-Analyzes-Spend.md (same examples, same explanation of where to ask). Per HELP_AUTHORING_GUIDELINES.md Section 7 (Cross-Linking Standards): "If a workflow is important enough to link repeatedly, it likely requires its own dedicated article." Since the detailed version already lives in How-Concierge-Analyzes-Spend.md, consider trimming this to a brief summary (1-2 sentences) plus a cross-link, rather than repeating the full example list. This avoids content drift between the two files and improves retrieval precision.

@stephanieelliott stephanieelliott added the HelpDot Apply this label if the issue relates to ExpensifyHelp label Jun 6, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

HelpDot Apply this label if the issue relates to ExpensifyHelp

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants