At-risk franchisee interventions

What the intervention queue is

Every network has franchisees who are quietly drifting: margin sliding for two quarters running, a certification about to lapse, a safety record ticking up, a franchisee gone dark on data. The intervention queue is where Verinode HQ turns that drift into a tracked, working list. It lives inside Decisions, alongside the other three kinds of network-level work HQ manages there: decision plans (playbooks), consent requests, and program audits and violations. An intervention is simply the risk-domain row type: a flagged at-risk location, the reason it was flagged, and a running record of what HQ is doing about it.

This is not a franchisee performance dashboard and it is not a case-management tool bolted onto a CRM. It is the same Decisions surface every other network-level item runs through, filtered down to the rows that read as risk. Verinode surfaces the pattern and gives HQ a place to track the follow-up; HQ decides what to do and does it.

Where to find it

Open Decisions from the HQ sidebar, in the unlabeled top group above Feed, Action Plans, Playbooks, Broadcast, and Vault. The URL is hq.verinode.ai/decisions.

Decisions runs as a gallery, not a Gantt or calendar (those views belong to Action Plans). Every row across the network, decision plans, interventions, consent requests, program audits, and program violations, lands in the same tile grid, filtered by status and business area. Interventions are the rows tagged with the Risk business area.

The subtitle under the page header sums up the whole surface in one line: every network-level decision across plans, interventions, consent requests, and program audits, all in one franchisor inbox.

How a franchisee lands in the queue

Interventions are not auto-generated from a scoring model. An HQ admin flags a franchisee deliberately, from that franchisee's own record on the Network page (hq.verinode.ai/network, the unified home for the franchisee directory and network health rollups; the older /franchisees URL still works as a deep link).

1. 1Open the franchisee's record on the Network page and click Flag.
2. 2Pick a Reason from the dropdown: "Engagement dropped, no recent data," "Margin trending below network," "Compliance gap, cert expiring or expired," "Brand-protection concern (reputation / safety)," or "Other (note inline)."
3. 3Optionally add an Intervention type, a free-text label such as "coaching call," "on-site visit," or "broker check-in." It shows up on the intervention's tile in Decisions.
4. 4Click Flag franchisee. The confirmation reads "Flagged [franchisee]. The intervention queue updates next refresh," and the row appears in Decisions as a new, queued intervention.

Only group admins can flag a franchisee. Verinode will not let the same franchisee stack two open interventions at once, if one is already queued, contacted, or in progress, flagging blocks with "An open intervention already exists for this franchisee" until the open one is closed out. Every other admin in the group gets a notification, "Franchisee flagged for intervention," so the whole leadership team sees the queue update without the person who flagged it needing to announce it separately.

What an intervention tile shows

Each intervention tile in the gallery carries:

• Title: the flag reason itself, rendered large since interventions carry no dollar estimate. Verinode does not try to price what a franchisee relationship going sideways is worth, so there is no impact figure and no cost-of-inaction framing on this tile type, just the reason, plainly stated.
• A colored status pill, top right, in place of a company logo (franchisees aren't the kind of counterparty Verinode resolves a logo for). The color also runs as a thin accent bar across the top of the card: red for an urgent pending item, amber for a routine pending one, green once it's been acted on or resolved, grey once it's been closed with no action.
• The franchisee's name, or its de-identified stand-in, depending on your network's privacy setting (see below).
• A domain glyph watermark identifying it as a Risk-domain row, distinguishing it at a glance from the Operations, Governance, and Compliance rows mixed into the same gallery.

Status and domain facets

Above the gallery, a row of status pills filters the whole board: All, Pending, Acted, Parked, Ignored, Resolved, each with a live count. For interventions specifically:

• Pending covers a freshly queued intervention that hasn't been touched yet.
• Acted covers one that's been contacted or is in progress.
• Parked never applies to interventions, an intervention has no "come back later" state, only open, acted-on, resolved, or closed.
• Ignored covers one closed out with no action taken.
• Resolved covers one closed out as fixed.

Below the status pills, a set of business-area chips narrows further within whatever status is active; interventions carry the Risk chip. A search box (placeholder "Search decisions…") filters by keyword, and a sort control offers Newest first (default) or Biggest impact, the latter isn't very discriminating for interventions specifically since they carry no dollar value, but it still matters on a mixed board with plans, audits, and violations that do.

Working an intervention: Act and Ignore

Open a queued intervention's tile and two buttons are live: Act and Ignore. There's no "Not now" for interventions (it's greyed out where it would normally sit, interventions don't have a parked state to defer into).

• Act advances the intervention one step through its real lifecycle: queued moves to contacted, contacted moves to in progress, and in progress moves to resolved (stamping a resolved date). The first time you click it, the tile tints green, picks up a small Acted stamp in the corner, and the button row clears, that's Verinode's signal that this one is now being worked and off the "needs a decision" pile. From there, keep the intervention moving and keep a record of what's happening by working it from the franchisee's own record on the Network page (the Add notes / Edit notes button covered below), and by opening the tile again to loop the agent in.
• Ignore closes the intervention out immediately as no further action, from any open status. It's the right call when the flag turns out to be a false alarm or something the franchisee has already resolved on their own.

Once an intervention is resolved or ignored, it's terminal, it sits in the Resolved or Ignored facet and no longer takes actions from the tile.

Opening an intervention: the workspace

Click any intervention tile (not one of its buttons) to open the full-screen workspace overlay. It carries:

• The flag reason as the title, and the franchisee's name (real or de-identified, per your privacy setting).
• The Risk domain, the severity, and the dates: when it was flagged, when it was actioned, when it was resolved.
• The notes body, HQ's own running record of the outreach, calls, and next steps, decrypted the same way described below.
• Left/right arrows (and swipe) to page through the same filtered list you had open in the gallery, so you can work a batch of interventions without closing back out to the grid each time. Escape, the backdrop, or the close button return you to the gallery.

Opening the workspace also hands the intervention's context to the AI agent panel, so you can ask the Franchisee Intervention Coach for a recommended next step or a check-in script without leaving the queue. That specialist reads only HQ's own network aggregates (margin position, cash runway band, cert status, the open intervention itself), never a franchisee's underlying business records, and it will say so plainly if the data behind a recommendation is thin rather than making something up.

Adding and editing notes

The notes you (and other HQ admins) keep on an intervention are managed from the franchisee's own record on the Network page, via an Add notes (or Edit notes, once notes exist) button. It opens a modal titled "Notes · [franchisee]" with a single field:

Intervention notes: "Encrypted at rest under the group's admin-only key. Only HQ admins with active sessions can read these notes."

The placeholder prompts you toward what to write: what's the latest with this franchisee, what did the broker say, what's the next step. Saving writes the note (and revalidates the queue) without changing the intervention's status, so you can log progress mid-lifecycle without accidentally advancing it.

How the notes are actually protected

This is the detail the frontend copy hints at but doesn't spell out, so here it is precisely: intervention notes are encrypted under the group's own admin-only DEK (data encryption key), which is a completely different key from anything tied to a franchisee's private business data.

• Every HQ group has its own admin-only DEK. It's wrapped once per admin, under that individual admin's personal Vault Key, the same personal key each HQ admin sets up for their own account. There is no master copy of it sitting anywhere in Verinode's infrastructure: unwrapping it requires an authorized admin's active, signed-in session. Verinode's own servers cannot decrypt these notes on their own.
• When you save a note, Verinode writes it twice: the plaintext column (for the transition period while this protection rolls out) and, whenever your session already has the group's admin-only key cached, a ciphertext column alongside it.
• When the queue or the workspace needs to show you a note, it prefers the ciphertext: if your session has the admin-only key cached, it decrypts the ciphertext. If that key isn't in scope for any reason (session just started, key not yet provisioned for this admin), it quietly falls back to showing the plaintext column instead, so a note is never simply missing.
• This key has nothing to do with any franchisee's own operator-side vault. A franchisee's private business data is protected under keys HQ never holds and never sees the plaintext of. The admin-only DEK exists solely to protect what HQ writes about a franchisee inside HQ's own workspace, HQ's private notes on its own intervention, kept away from anyone without an authorized HQ admin session, including Verinode's own infrastructure.

Privacy boundary: named or de-identified

Whether an intervention shows the franchisee's real name or a de-identified stand-in depends on your network's entity model, a setting on the group itself:

• Most networks (the default, independent operators): franchisee names are never shown as-is on this surface. Instead you'll see a stable label like Franchisee #A1B2, generated from the last four characters of that franchisee's operator ID so the same franchisee always shows the same label across every intervention, without ever exposing the real business name.
• Same-entity networks (one legal entity operating multiple company-owned locations): the real location name passes through unchanged, since there's no separate business to anonymize from.

This same rule applies everywhere an intervention's franchisee identity could surface, the tile, the workspace, and even the text handed to the AI agent, so a network running in the default, anonymized mode never has its franchisee's identity leak through a chat answer either.

Empty states

• No interventions at all, viewing "All": the gallery reads, "No decisions for you yet. They'll land here as Verinode spots cost savings, risk, and growth opportunities in your data." Since interventions are HQ-initiated rather than auto-generated, this is genuinely the state for a network where nobody has been flagged, not a sign anything is broken.
• Filtered to a status with nothing in it (say, Resolved with nothing resolved yet): the gallery reads, "Nothing resolved for you right now" (the same pattern for whichever status pill is active).
• On the franchisee's Network page, the read-only interventions list there (a sibling view, for context rather than action) shows: "No franchisees flagged for intervention. Use the directory to flag at-risk operators."

Best-practice example

Say the margin cohort report flags a franchisee whose net margin has dropped from 47% to 38% over two straight quarters. From that franchisee's record on the Network page, click Flag, pick "Margin trending below network," and leave a note on the intervention type, "coaching call." The intervention lands in Decisions, Pending, Risk. Open it, ask the Franchisee Intervention Coach for a check-in script grounded in the actual margin gap, and use it to open the conversation. After the call, log what came out of it in Add notes, "labor mix and pricing reviewed, follow-up scheduled." Click Act to move it from queued to contacted. As the coaching plays out over the following weeks, keep the notes current; when margin recovers or the franchisee has a credible plan in motion, either resolve it by working it through to completion, or Ignore it if it turns out to have been a one-quarter blip rather than a trend.

Related reading

• HQ overview, for how Decisions fits alongside Feed, Action Plans, Playbooks, and the rest of the HQ shell.
• Network health, for the franchisee directory, the Flag button, and the quartile and cohort views an intervention often starts from.
• HQ compliance, for cert-lapse and audit-driven flags that share the same Decisions gallery under the Compliance business area.
• Broadcasting to your network, for reaching the whole network at once, as distinct from the one-to-one follow-up an intervention represents.