The Findings tab: acting on process decisions in place
Processes is where your fractional COO watches how work actually gets done: the SOPs you've documented, the standards your crews should be following, and the patterns Verinode has noticed in your j…
On this page
What the Findings tab is
Processes is where your fractional COO watches how work actually gets done: the SOPs you've documented, the standards your crews should be following, and the patterns Verinode has noticed in your job history. The Findings tab is the action layer on top of all of that. It's the short list of open process decisions Verinode has surfaced for you, things like a step your cohort runs that you don't, a standard you haven't adopted, or a pattern your Business Analyst flagged in a scan, each one framed as a call for you to make rather than a problem for Verinode to solve on its own.
What makes Findings different from a plain list is that you never have to leave Processes to work one. Click a tile and it flips in place into the full decision workspace, the same Act / Not now / Why? card you'd see on /decisions, with the agent panel already scoped to that exact decision. Only when you commit with Act does Verinode hand you off to the standalone workspace page, because that's where the real plan and its progress live going forward. Everything short of that, reviewing, parking, asking why, happens without a page change.
Where to find it
Open Processes from the Operations group in the sidebar, at iq.verinode.ai/processes. Across the top of the Processes card is a row of six tabs: Findings, Flow, All Processes, By Standard, Coverage, and Benchmarks. Findings is first, marked with a red dot in the tab strip, and it's the one this article covers.
Note
Processes is a Premier-tier surface. On Contributor or Executive, the section shows a locked preview (a count of how many processes Verinode is already tracking for you, with an upgrade prompt) instead of the live Findings gallery. The fractional-COO process coach, SOP library, and lean scoring underneath it are part of what Premier unlocks.
The gallery: what each tile means
Findings renders as a grid of square tiles, one per open decision, with a small eyebrow above it reading "N decisions to review" (singular "1 decision to review" when there's only one). Each tile carries:
- A dollar figure, when Verinode has one. This is the hero number on the tile: a monthly figure with /mo, or an annual figure with a unit suffix (/yr, /wk, /job, or one-time for a pool that clears once rather than a recurring drain, so a one-off catch-up is never mislabeled as a run-rate). Underneath it, the title line, the plain-language description of what Verinode found.
- No dollar, when the risk can't be cleanly priced. Some findings, an insurance gap, a missing cert, aren't easily reduced to a number. Those tiles instead lead with a small risk-and-urgency chip (for example Compliance risk · Act soon) and a one-line consequence statement, with the title underneath in smaller text.
- A status marker, top right. Open findings carry a soft Pending marker. If you've already committed to one (see "Once you've acted" below), it swaps to a green Acted or Done rubber-stamp instead.
- "From a scan," when relevant. A small copper tag appears on findings your Business Analyst surfaced from a pattern scan across your process data, rather than ones tied to a specific standard or peer gap.
Opening a decision: the drilled-in workspace
Click a pending tile and it flips in place into the full decision. A breadcrumb at the top, ← All decisions, takes you back to the gallery, and a line at the bottom, "← / → or swipe to navigate decisions · Esc to go back," tells you how to move to the next one without closing the drill: arrow keys, a trackpad swipe, or the floating side arrows all page between the decisions in your list.
Inside, the full decision card shows:
- A header row: a colored trajectory dot (declining decisions pulse), a label (Declining, Needs context, or Decision), the entity the finding is about when there is one, and a confidence pill (high, medium, or low).
- The dollar-at-risk line, when there's a number: the annual or monthly figure at stake, tagged Calculated (green, when it's built directly from your own data) or Estimated (amber, when it's a working assumption), sometimes refined further with Peer cohort or Industry baseline to say where the estimate is anchored. A short sentence beneath explains the math in plain language.
- The no-dollar version, when there isn't one: the same risk-and-urgency chip as the tile, with the full consequence statement below it.
- The finding itself as a headline sentence, and a 4-stage lifecycle bar, Flagged → Planned → Acting → Resolved, showing exactly where this decision sits right now.
- One recommendation tile. Either a Recommended action block (the concrete step Verinode suggests, plus "If you don't: …" spelling out the cost of leaving it alone), or, when confidence isn't there yet, a Needs your context block explaining that clicking Act will have the agent ask a couple of quick questions before it commits to a specific recommendation. When confidence is less than high, the recommendation tile also carries a "Refine this estimate:" line naming the one piece of data (an invoice, a document) that would sharpen the number.
- What peers did, when a comparable cohort exists: how many operators like you faced the same signal in the last 90 days, how many acted, and, once outcomes have had time to verify, how many of those saw the situation improve. This block simply disappears when there's no cohort to compare against yet, it never shows an "insufficient data" message.
Act, park, or dig into why, without leaving Processes
Three controls sit at the bottom of the drilled-in card:
- 1Act. Commits to the decision and hands you to the standalone workspace at
/decisions/[id], the same page the decision workspace article covers in full, where the agent drafts a step-by-step plan and you track it to done. This is the one control that takes you out of Processes, because a live plan deserves its own page. - 2Not now. Opens a short reason picker right there in the card, Too busy, Need more info, Not convinced, or Other. Pick one and the decision parks in place: the card collapses to a quiet strip reading "Parked, this will resurface later," and you're still on Processes, still on this tab, ready to page to the next one.
- 3Why? Only shown when Verinode has evidence to back the finding. Toggles a drawer open in place with the proof points behind the number and, where relevant, the root-cause chain, each link in "→ this happened, which caused this" order. Click Hide to collapse it again.
For everything short of committing, Not now and Why both resolve without ever navigating away from Processes. See acting on decisions for how Act, Park, and the rest of this vocabulary behave the same way across every section of the platform, not just here.
Talking it through. There's no separate "Discuss" button on the card, because the moment you drill into a decision, the agent panel on the side of the screen becomes scoped to it automatically, the same entity, the same metric, the same title you're looking at. Open the panel and ask your question in chat; you don't need to re-explain which decision you mean.
Once you've acted
A decision you've already committed to (its status is Acted or Resolved) looks different back on the gallery: the green rubber-stamp replaces the Pending marker, and the tile's single working button reads Edit plan (or Review outcome, once it's closed out). Clicking that tile, or its button, skips the in-place drill entirely and routes you straight to /decisions/[id], because there's already a real plan running there and that's the one canonical place to keep working it.
When there's nothing to review
If Verinode hasn't surfaced any open process decisions, the tab reads:
No decisions for processes right now. As your agent finds patterns, they'll appear here.
This isn't a broken screen, it means your SOPs, standards adoption, and process history haven't produced a decision worth calling out yet. Add SOPs on All Processes, adopt a standard on By Standard, or forward the documents that feed your job history (see forwarding documents and connecting your data), and findings will start appearing as that data flows in.
Best-practice example
Say Findings shows a decision reading "$1,900/mo estimated, Vendor pre-inspection step: 68% of peers run it, you don't," tagged Estimated · Peer cohort. Click the tile. Inside, the recommendation tile explains the step and its cost of skipping it, and What peers did shows most operators in your cohort who adopted it saw their cycle time improve. If you're not ready to commit yet, hit Why? to see the peer adoption numbers behind the estimate, then Not now with "Need more info" so it parks instead of nagging you, and swipe to the next decision in the list. When you are ready, Act hands you to the workspace, where the agent turns the recommendation into a dated plan.
Data sources
Data sources
- 1.Your SOPs, standards adoption, and observed job patterns. Your business.
- 2.Business Analyst scans of your process data. Verinode IQ.
- 3.Peer-cohort adoption and outcome data. Verinode intelligence layer (anonymized).