> ## Documentation Index > Fetch the complete documentation index at: https://docs.vortexiq.ai/llms.txt > Use this file to discover all available pages before exploring further. # Abandoned Findings (>14d no movement), Confluence > Findings sat in the backlog with no status change for two weeks, these are the ones losing money silently. How to read it, why it matters, and how to act ... **Metrics type:** [Key Metrics](/nerve-centre/overview#metrics-types-explained) • **Category:** [Project Management](/nerve-centre/connectors#connectors-by-type) > Findings sat in the backlog with no status change for two weeks, these are the ones losing money silently. ## At a glance > Open finding-pages in your Confluence *Findings* Space whose `version.when` (Confluence's last-updated timestamp) has not changed for 14+ days. The finding-page is sitting in the wiki, nobody's touching it, and the underlying issue (broken redirect, slow page, abandoned cart) keeps bleeding revenue. | | | | ------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **What it counts** | Pages in the mapped Findings Space where `vortexiq.status` is in the open set AND Confluence's `version.when` is older than 14 days. | | **What "no movement" means** | Confluence updates `version.when` on any page edit (content, page-property change, attachment add). Comments and likes do NOT bump the version, so a finding-page with a lively comment thread but no actual page edits will count as abandoned. This is intentional, comments without a status change rarely lead to resolution. | | **Confluence Space scope** | Same single Space as `Open Findings`. Other Spaces are not crawled. | | **Status filter** | Same open set as `Open Findings` (`open`, `investigating`, `in_progress`, `blocked`). Closed pages never count, regardless of how long they've sat closed. | | **The 14-day cliff** | Day 13: not abandoned. Day 14, 00:00 UTC: counted. Threshold fixed at the manifest level. | | **Why 14 days** | Two-sprint cadence. If a finding has not been opened, edited, or had its status changed across two sprints, neither sprint planning picked it up, that's the working definition of "forgotten" for Atlassian-aligned teams. | | **Atlassian-suite consequence** | If you also run Jira and use Confluence-primary routing, an abandoned Confluence finding-page implies a stale linked Jira ticket too. The two cards (`con_vortexiq_findings_abandoned` and `jir_vortexiq_findings_abandoned`) typically agree to within 1, 2 rows for these merchants. | | **Time window** | `RT` (real-time, threshold applied at query time). | | **Alert trigger** | `>5 (warn)` opens an amber banner; `>15 (critical)` triggers a red banner and a Slack ping if Slack is connected. | | **Sentiment key** | `{'type': 'threshold', 'thresholds': {'warn': 5, 'critical': 15}}` | | **Roles** | owner, operations | ## Calculation Calculated automatically from your Confluence data. See the At a glance summary above for what the metric tracks and the worked example below for a typical reading. ## Worked example Same US enterprise retailer from the *Open Findings* card. 34 open finding-pages in their VFND Space, snapshot taken on 24 Feb 26. `version.when` distribution across the 34 open pages: | Last-edited age | Open pages | Counts as abandoned? | Note | | --------------- | ---------- | -------------------- | ----------------------------------------------- | | 0, 7 days ago | 13 | no | Active work. | | 8, 13 days ago | 9 | no | Inside the 14-day window. | | 14, 30 days ago | 6 | yes | Genuinely forgotten, last touched in early Feb. | | 31, 60 days ago | 4 | yes | Old. Should be triaged. | | 61+ days ago | 2 | yes | One has been sitting since Nov 25. | ```text theme={null} Abandoned count = 6 + 4 + 2 = 12 ``` The card shows **12** with an amber border (warn at 5, critical at 15). The dashboard banner reads *"12 finding-pages haven't been touched in 2+ weeks"*. **The recommended response (Atlassian-aligned teams):** 1. **Schedule the triage in Confluence itself.** Use the page-properties report at the top of the Findings Space, sort by *Last edited* ascending. The 12 abandoned pages bubble to the top. 2. **For each abandoned finding, make one decision:** * Pick it up now: assign to a teammate via `vortexiq.assignee`, change `vortexiq.status` to `in_progress`. The act of editing resets the abandonment clock. * Defer with a date: add a `vortexiq.defer_until` page property and reset `vortexiq.status` to `open`. (Defer-aware abandonment scoring is on the roadmap.) * Decline: change `vortexiq.status` to `wont_fix` with a one-line comment explaining why. 3. **If both Jira and Confluence are connected**, this triage should also update the linked Jira issue, Vortex IQ's two-way sync handles that automatically when you change the Confluence page-property. After triage this card typically drops below 5. If it climbs back above 5 within a fortnight, the team has a deeper problem (under-staffed, wrong audit-rule priorities, or findings that aren't relevant). ## Sibling cards merchants should reference together Abandoned is the high-value subset of Open. Always read alongside: | Card | Why it matters next to Abandoned | What the combination tells you | | ------------------------------------------------------------------------------------------- | --------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------- | | [VortexIQ Findings Open](/nerve-centre/kpi-cards/confluence/vortexiq-findings-open) | The denominator. Abandoned ÷ Open = "stale ratio". | Stale ratio above 30% means the queue is mostly dead weight. Triage, don't add capacity. | | [Finding Resolution Rate (90d)](/nerve-centre/kpi-cards/confluence/finding-resolution-rate) | Active turnover. | Low Resolution + high Abandoned = team is paralysed. High Resolution + high Abandoned = team picks easy wins, leaves hard ones. | | [Avg Time-to-Fix](/nerve-centre/kpi-cards/confluence/avg-time-to-fix) | Days from creation to closed for the ones that *do* move. | Fine Time-to-Fix + rising Abandoned = team is fast on what it picks up but bad at picking up everything. Priority misalignment. | | [Tickets by Assignee](/nerve-centre/kpi-cards/confluence/tickets-by-assignee) | Whether abandonment concentrates on one person. | If one assignee owns most of the abandoned pages, that's a single-point-of-failure conversation. | | [Backlog by Priority](/nerve-centre/kpi-cards/confluence/backlog-by-priority) | What priority the abandoned pages are. | If 8 of 12 abandoned are `low`, close them with "Won't fix" and stop counting them, the high-priority abandoned are the ones that matter. | ## Reconciling against the vendor's own dashboard **Where to reproduce this in Confluence itself:** Use the page-properties report macro that the connector installs at the top of your Findings Space. By default it shows status, priority, assignee, and last-edited. Sort by *Last edited* ascending; rows older than 14 days with status not in the closed set are the abandoned set. The row count should match this card. If the page-properties macro is missing (someone deleted it), the alternative is a CQL search: ```cql theme={null} space = VFND AND label = "vortexiq-finding" AND lastModified <= now("-14d") ``` Then post-filter the result page to exclude statuses in `done, wont_fix, duplicate`. CQL doesn't query page-property values directly; the page-properties report is more accurate. **Why our number may legitimately differ from the page-properties report:** | Reason | Direction | Why | | --------------------------- | ----------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Time-zone boundary** | Edge cases | We compute "14 days ago" in UTC. Pages aged 13d 22h locally for a Sydney merchant may already be 14d 8h UTC. Drift at most 1 day. | | **Bulk page-property edit** | Ours lower | If a script or automation touches every page's properties (e.g. a nightly enrichment), `version.when` resets and nothing looks abandoned to us. Avoid bulk-touching pages you're not actively working. | | **Comments without edits** | Ours higher | A finding-page with active comment activity but no page edits looks abandoned to us. This is intentional (comments rarely lead to resolution alone), but you may want to bump status manually after substantive comment threads. | | **Restored from trash** | Ours lower | A restored page has its `version.when` updated; it looks freshly edited even if it had been abandoned for months pre-deletion. Edge case. | | **Reorganised pages** | Either | Moving a page to a new parent counts as an edit (`version.when` updates) but the underlying finding hasn't necessarily progressed. We can't distinguish move-only edits from substantive edits via the API. | **Cross-connector reconciliation. Confluence vs Jira (Atlassian-suite):** For Atlassian-suite-merchants with Confluence-primary routing, the linked Jira issue's `updated` timestamp typically tracks the Confluence page's `version.when` to within a few seconds (Vortex IQ's two-way sync handles this). So `con_vortexiq_findings_abandoned` and `jir_vortexiq_findings_abandoned` should match within 1, 2 rows. If they diverge widely (>5 rows), the most likely cause: a Jira automation that touches issue fields nightly (e.g. a "stale issue" reminder) is bumping the Jira `updated` field but not the Confluence page. Disable the Jira automation for issues labelled `vortexiq-finding` if you want the two cards to stay aligned. ## Known limitations / merchant FAQs **Why is this number going up?** Inflow continues, triage isn't happening. The fix is a fortnightly 30-minute triage (see *Worked example*), not "do more findings". You can't out-execute a queue that grows because nobody is making decisions. **Should I bulk-close everything older than 30 days?** Tempting, but no. Bulk-closing makes the number drop and the underlying culture problem ("we ignore findings") gets worse. Better: spend 30 minutes reading the abandoned pages, decide which ones are real, close the others with a one-line `wont_fix` reason. Those reasons themselves are useful, after a few weeks of "Won't fix because not relevant to subscriptions", you tune Vortex IQ's audit rules. **My team uses 4-week cycles, not 2-week sprints. Is 14 days right?** For most teams, yes, the threshold is the *early warning*, not the deadline. Even 4-week-cycle teams benefit from a fortnightly nudge that something is going stale. Per-workspace configurable thresholds are on the roadmap. **What about findings genuinely waiting on an external party (vendor, supplier, customer)?** Set `vortexiq.status = blocked` and add a comment with the unblock date. Blocked still counts as open (so it appears in *Open Findings*) but the team triages blocked-rows separately from genuinely-untouched ones. Don't close the page just because work is paused, that hides the dependency. **A finding-page has active comments but is still showing abandoned, why?** Comments don't bump `version.when` in Confluence's data model. If the page itself isn't being edited (status changed, content updated, attachments added), it counts as abandoned. After a substantive comment thread that resolves something, edit the page (even just to add a "Resolved per discussion" line) so the timestamp updates. **The card flagged abandoned but I edited the page-properties macro this morning, why is it still showing?** Macro edits *do* bump `version.when`. Either: (a) 60-second sync window, refresh in 2 minutes; (b) you edited the macro on a sub-page, not the parent finding-page (Confluence does NOT propagate sub-page edits to the parent's `version.when`); (c) the connector lost permission to read the page (check page restrictions). **Can I see *which* finding-pages are abandoned, not just the count?** Yes, click the card to drill in, or use the page-properties report at the top of the Findings Space sorted by *Last edited* ascending. **Does this work for Confluence Server / Data Center?** Cloud only today. Server / Data Center support is on the roadmap for late 2026, the abandonment logic is identical, but the API path differs (`/rest/api` vs `/wiki/api/v2`). **Why is this a Key Metrics card?** Because it's the queue-rot indicator. Open tells you queue depth; Resolution Rate tells you throughput. Abandoned tells you whether the queue is *alive*. A team can have fine numbers on the other two and still be quietly losing money on 8 forgotten high-impact findings. **Our Confluence has 50+ Spaces, are we accidentally counting other content?** No. Only pages in the mapped Findings Space with the `vortexiq-finding` label and a `vortexiq.status` page-property are counted. Your engineering wikis, runbooks, and meeting notes in other Spaces are invisible to this card. *** ### Tracked live in Vortex IQ Nerve Centre *Abandoned Findings (>14d no movement)* is one of hundreds of KPI pulses Vortex IQ tracks across Confluence and 70+ other ecommerce connectors. Nerve Centre runs the detection layer; Vortex Mind investigates the cause when something moves; Ask Viq lets you interrogate any number in plain English. [Start for free](https://app.vortexiq.ai/login) or [book a demo](https://www.vortexiq.ai/contact-us) to see this metric running on your own data.