Changelog — Enhancement

• ✓Relationship overlays are one drill-in menu instead of a section each
• ✓SCIM adopts existing users instead of erroring, and accepts more email types
• ✓Imports skip rows whose Id is not in your workspace
• ✓Tag filtering matches whether the server or the browser does it

A month of the things that are individually too small to announce and collectively very noticeable.

On a view. Relationship overlays used to take a toolbox section each, which pushed everything else out of reach once you had a few. They are now a single list you drill into. Dropdowns inside a drawer scroll themselves into view when they open instead of appearing off-screen, and value stages have room for two lines before they clip.

Provisioning. A SCIM POST /Users for somebody already in the organisation now adopts that user rather than returning a conflict, and email types other than exactly "work" are accepted — both cases real directories produce and the old behaviour rejected.

Imports and filters. Topic import rows carrying an Id that is not in your workspace are skipped rather than failing the run. Server-side tag filtering now returns what the browser's own filtering would have shown, so the two no longer disagree on the same query.

Elsewhere. Optional policies drop their target on save and hide the completion target they were never going to use, the fact template form explains what a rich text template is for, and the session cookie has an explicit lifetime rather than an implicit one.

• ✓Renew is now Confirm facts
• ✓The modal groups facts by what confirming will do to them
• ✓Facts can take effect on a future date and say so
• ✓Untrusted is now Unconfirmed, with plain-English captions

"Renew" described what the system did to a record. "Confirm" describes what you are actually doing, which is saying yes, this is still true.

The modal behind it was rebuilt to match. Facts are grouped by what confirming will do to them rather than listed in whatever order the query returned, so a batch where half the facts expire tomorrow and half are already lapsed reads as two decisions instead of one long list.

• Where several facts share a template, the one shown is the longest-held, which is the one whose age you are actually judging.

• A coloured status dot carries the state, and the Trust centre's own vocabulary changed to match: Unconfirmed for a Fact Card missing at least one expected fact, Lapsed for one confirmed once and now out of date, Trusted when everything expected is recorded and in date.

• Facts dated ahead of today now render with a note saying when they apply, instead of looking like they are already in force or vanishing until the date arrives.

• Confirmations asked in a modal are now asked in a styled dialog rather than the browser's own grey box.

• ✓New Portfolio health panel replaces the old glance panel
• ✓See what share of expected facts is recorded, per Fact Card type
• ✓Trust and confidence sits alongside it, linking through to the Trust centre
• ✓Recent activity now sits directly on the dashboard

The old glance panel counted things. Counting is not the same as knowing whether your portfolio is in good shape.

Portfolio health answers two questions instead. How much of what you said you needed is actually there, measured against the required and recommended facts you configured. And how much of that is current enough to rely on.

• Each Fact Card type gets a row, so a portfolio that is healthy overall but hollow in one corner shows that corner rather than averaging it away.

• The panel loads in its own frame. The rest of the dashboard is usable while it works out the numbers, and there is a skeleton in the meantime rather than a blank space.

• Recent activity moved out of its box and onto the page itself, below the health panel. Top contributors is gone — it was a leaderboard nobody acted on.

• Trust percentages no longer round to a confident 0% or 100% when they are merely very small or very nearly complete.

• ✓Controls float over a live preview instead of pushing it aside
• ✓A0 to A4 page sizes are honoured properly
• ✓Legend stays inside the sheet on every side
• ✓Past exports are collected on the page you made them from
• ✓Browser printing opens the designer

Exporting a view used to mean setting options on one screen and finding out what you had actually produced on another. The designer now shows you the sheet while you change it.

The controls sit over the preview and can be minimised out of the way when you want the whole canvas, and the export region runs to the edges of the screen rather than a box in the middle of it.

• Page size finally behaves. A0, A1, A2 and A3 were being quietly rendered at A4; picking one now produces a sheet at that size, with the preview's viewport derived from the sheet's real dimensions rather than an assumption about it.

• The legend is properly contained on all four sides. Above and below were the awkward cases — a wide legend used to run past the edge of the canvas — and its entries now take the colour of the series they describe, so a printed sheet reads the same way the screen does.

• A card on the export page collects what you have generated, showing each one as it moves from Generating… to Ready, with everything else under Past exports.

• Printing a view from the browser now opens the designer instead of sending the raw page to the printer, which is almost always what was actually wanted.

• Settings passed in the URL take priority over your stored preference, so a link to a specific export produces that export.

Keeping your catalogue organised just got quicker. From the Fact Cards table you can now tag many cards at once instead of opening them one at a time.

• Select the Fact Cards you want, then choose Assign tags from the actions bar.

• Pick one or more tags in the picker and apply — the tags are added to every selected card.

• Existing tags are always kept. Assigning is additive, so nothing you've already tagged is removed.

• Rows update in place as soon as you apply, so you can see the result without reloading.

This joins the other bulk actions on the Fact Cards table — Archive, Restore, and Delete — so you can manage a whole set of cards together.

• ✓New views start as private drafts — only you can see them
• ✓Publish in one click to make a view visible to everyone
• ✓Unpublish a published view back to a draft at any time

Views now have a visibility state, so you can put a view together and refine it without cluttering everyone else's workspace until it's ready. Existing views are unaffected — they remain visible to everyone.

• A new view opens as a Draft, private to you, marked with a Draft badge in the header, view lists, and Manage Views

• Publish is the primary button in the view header; while a view is still a draft, Save steps back as the secondary action

• Use the view's ⋯ menu to Unpublish to draft when you want to take a published view back out of sight

• Manage Views shows each view's state at a glance — Draft, Published, or Disabled

We've reorganised how you set the data quality bar for each fact, so it's quicker to find, easier to read, and easier to explain to teammates.

One drawer for everything

From any Insight or template page, click Adjust on the data quality card to open the Configure data quality drawer. Every policy field - scope, target completion, missing-data priority, review period, change reason, and fact lifecycle - now lives in one place.

Fact lifecycle as visual cards

The lifecycle picker is now a 2×2 grid of cards: Starts when completed, Planned start date, Expiry date, Start and expiry dates. Each card shows an icon, title, and short description, so the choice is obvious at a glance.

Plain-English labels and help text

We've revisited the wording across the drawer:

• Applies to - choose which Fact Card types should collect this information.

• Target completion - set the percentage of in-scope records that should be completed.

• Missing data priority - choose how strongly missing data should be highlighted.

• Review period - choose how long facts remain trusted before review is required.

• Change reason - choose whether contributors need to explain updates to this information. Options are now Not required, Optional, and Mandatory (coming soon).

Where to look: Settings → Templates → any fact template → "Adjust", or the data quality card on any Insight page.

• ✓Keyboard shortcut to toggle the toolbox
• ✓Clearer button and interaction feedback in places you use often

Alongside larger releases, we’ve shipped a bundle of smaller improvements that add up: faster access to tools you use repeatedly, tighter feedback when you click, and touch-ups to previews and stepped flows.

• Toggle the toolbox with a dedicated shortcut (see Keyboard shortcuts for the current binding)

• Polished button interactions where micro-feedback matters

• Incremental improvements to previews and multistep experiences

• ✓Safer trust story for anyone cleaning up group structure

Removing a group is a bigger deal than it looks, so we spell it out: the dialog warns you clearly, and fact cards are unlinked as part of the delete so relationships stay consistent.

• Stronger warning copy before a group is removed

• Automatic unlinking of fact cards from the group being deleted

• Fewer orphaned or confusing group memberships after cleanup

Relation types (e.g. dependencies, supports, stores) are now available as columns in the Facts table.

• Added relation type columns to the Facts bootstrap table

• Columns are hidden by default — toggle visibility via the column dropdown

• Only relations for enabled fact card types in your workspace are shown

• ✓Lazy-loaded panels so views open faster
• ✓Archive or delete panels from the menu; mapping overlay when placing cards
• ✓Resize and layout fixes for more predictable panels

Views feel snappier and easier to share: panels load when you need them, presets can be captured as short links, and day-to-day management (archive, delete, mapping topics into a panel) is more explicit.

• View panels use lazy Turbo loading to defer work until a panel is shown

• Table presets can expose a short code for shareable snapshot-style links

• Archive or remove panels from the dropdown; mapping topic cards shows a dedicated overlay

• Sizing, shadows, and resize behaviour refinements for steadier layouts

• ✓Inline editing on the fact card page and drawer for faster updates
• ✓Supports inline edits for name, aliases, and definition

You can make quick corrections where you already are: on the fact card itself, and—where enabled—for groups straight from the drawer.

• Fact card page supports inline editing for a smoother update loop

• Group drawers allow inline changes to name, aliases, and definition (subject to LaunchDarkly rollout for your workspace)

• Less context switching between read and edit modes for everyday tweaks

• ✓Lightbox experience for reviewing library content
• ✓Stroger template card previews and layout
• ✓Incremental UX and styling improvements throughout the library

The Content Library is easier to scan and compare: open items in a lightbox, see cleaner previews on cards, and benefit from a round of layout and interaction polish.

• ✓ServiceNow template group and external template type for integrated content
• ✓Field and attribute mapping with sync frequency and status in the integration summary
• ✓External facts surfaced in the topic drawer and on the fact card, with a more dynamic fact card layout

You can bring ServiceNow data into Colloquial with clearer mapping controls and a more predictable experience when fact cards combine internal and external sources.

• Template group — Organise ServiceNow-backed templates alongside your existing library.

• Mapping and sync — Configure field and attribute mappings, set sync frequency, and see sync state in the integration summary so you know when data was last refreshed.

• Where facts appear — External facts show in the topic drawer and on the fact card, so the same relationship context is visible wherever you’re working.

• Licensing — If you’d like to discuss licensing for the ServiceNow integration, please reach out to us and we’ll help you with the right setup for your organisation.

Improved Topics import is more user friendly and adds capabilities to fine tune data import.

What is new

1. Create/update: Creates new or updates existing Fact Cards

2. Archive: Archives existing Fact Cards

3. Purge: Permanently deletes existing Fact Cards

4. Skip: Ignores rows during import

5. Add group only: Adds to groups without updating other data

6. Remove group only: Removes from groups without updating other data

7. All data: Updates all Fact Card data (for multi-groupable types)

Where you will see this

Data import capabilities:

• Basic Data: Name, aliases, description

• Groups: Single and multi-group assignments

• Hierarchical Groups: Support for nested group structures (up to 5 levels)

• Assessments: Scale values and numeric responses

• Tags: Multiple tag assignments with automatic creation

• Relationships: Parent-child relationships and partnerships

• Detail Answers: Fact template responses with type validation

Validation system

• Column Headers: Automatic detection and validation

• Data Integrity: Required field validation and type checking

• Import Handler Rules: Validation of handler combinations

• Reference Validation: Ensures referenced entities exist

Error handling

• Row-level Errors: Specific error messages with row numbers

• Validation Errors: Clear descriptions of data issues

• Reference Errors: Missing parent, group, or template errors

• Import Summary: Counts of created and updated records

See the Import Data User Guide (https://help.colloquial.io/article/42-import-data) for further details.

Tab groups handle cases where there are more tabs than fit in the available space.

What is new

• Smart Overflow: When you have many tabs, excess tabs move into a dropdown menu

• Flexible Layout: Tab bars adapt to different screen sizes and content lengths

• Better Navigation: Access all tabs easily, even with large numbers of options

• Consistent Experience: Tab positions stay stable when switching between tabs

Where you will see this

• Settings Groups: Navigate between configuration types with quick access through the overflow menu

• All Tab Interfaces: Screens with multiple tabs handle overflow consistently

Benefits

• Improved Usability: No cramped or cut-off tabs on smaller screens

• Better Organization: Keep the workspace clean while retaining access to all functionality

• Future-Ready: Add more tab options without cluttering the interface

This helps complex workflows that rely on many tab views, especially for power users managing extensive configurations or datasets.

Stakeholders are now visible in fact card tables for all fact card types except stakeholders themselves.

The platform is upgraded to Ruby 3.3 for performance and security improvements.

What is new

Ruby 3.3 improvements that benefit users:

• Performance boost: Up to 15% faster response times across the platform

• Enhanced security: Latest security patches and vulnerability fixes

• Memory optimisation: Reduced memory usage for better scalability

• Improved garbage collection: More efficient memory management

Benefits for users

• Faster page loads when navigating between fact cards and dashboards

• Better reliability during peak usage

• Improved search performance

• Enhanced API responsiveness for connected systems

Technical highlights

• Updated Ruby runtime from 3.2 to 3.3.1

• Compatibility testing across platform features

• Performance benchmarking

• Security audit to validate dependencies

Seamless transition

The upgrade was completed with zero downtime during a scheduled maintenance window. Existing functionality is unchanged; users should see improved performance without action on their part.