The four initiative kinds: playbook, directive, recommendation, experiment
An initiative is a piece of guidance an HQ admin writes once and pushes out across the network: a process to standardize, a mandatory action with a deadline, a suggestion worth trying, or a limited…
On this page
What an initiative is
An initiative is a piece of guidance an HQ admin writes once and pushes out across the network: a process to standardize, a mandatory action with a deadline, a suggestion worth trying, or a limited test. Every initiative is tagged with one of four kinds, set once at creation and never changed afterward: Playbook, Directive, Recommendation, or Experiment. The kind is not decoration. It tells every franchisee, and every other admin looking at the same board, what level of commitment this piece of guidance is asking for, before they open it.
This article covers what each kind means, when to reach for it, and exactly how you pick it in the plan editor. For the full authoring flow (templates, push-to-franchisees, sponsorship, status lifecycle) see Broadcasting to your network.
Note
Verinode HQ works from network-level records only. Authoring an initiative, pushing it, and watching franchisees acknowledge or complete it never involves Verinode reading a franchisee's own jobs, invoices, or client data. Franchisees own their business data; HQ sees the initiative's own text, its status, and the adoption record each franchisee's response leaves behind.
Where to find it
Open Broadcast from the HQ sidebar (hq.verinode.ai/broadcast). Broadcast has five tabs across the top in a rounded capsule strip: Announcements, Initiatives, Surveys, Polls, Consent. Click Initiatives. The same underlying table also feeds the Decisions surface elsewhere in HQ, so you may see the identical row show up there too once it's active; this article is about the authoring side, on Broadcast.
If you're a group admin, the page header shows a + New initiative button. Viewers (non-admin HQ users) don't see this button; they can still open and read every initiative, they just can't author or push one.
Authoring one: the plan editor
Click + New initiative and a modal opens (its title reads "New decision," the same modal is shared with the Decisions surface's own "+ New decision" button). It has three fields:
- Plan kind (required). A dropdown with the four options below. Under the dropdown, a one-line hint updates live as you change the selection, so you can read what each kind means without leaving the form.
- Title (required). Placeholder example: "e.g. Quarterly cycle-time review with bottom-quartile franchisees." Keep it short and specific; this is what shows up on every tile and in the notification bell.
- Detail (optional). A multi-line text box for the body of the initiative. Whatever you write here is what a franchisee actually sees once you push the plan to them, so write it the way you'd want it read, not as an internal note to yourself.
Below the fields, a line of copy sets expectations: "New decisions land in draft. Move them to Active from the detail view when they're ready to deploy." Nothing you save here goes out to a single franchisee immediately. Click Create draft (it reads "Saving…" while the write is in flight) and the initiative lands in Draft status. From there, an admin opens its detail view, reviews it, and clicks Mark Active when it's ready, then uses Push to franchisees to actually choose who receives it. See Broadcasting to your network for the full activate-and-push flow and Network health for how adoption then reports back.
If you'd rather not start from a blank form, click Browse templates above the Plan kind field. It expands an inline list of Verinode-curated (and, where a network has contributed one, community) templates, filterable by category chip (Operations, Safety, Reputation, Compliance, Recruiting, Finance, General), each row showing its kind, category, and how many times it's been used before ("used 6×"). Click Use this on a template and it pre-fills the Plan kind, Title, and Detail fields, plus shows a "From template: [name]" marker with a clear link if you want to detach and author the rest as your own. Every field stays fully editable after picking a template, it's a starting point, not a locked-in choice. If no templates have been published yet, the picker reads "No templates published yet. Start from scratch below."
Heads up
Plan kind is set once, at creation, and there is no field to change it afterward. If you picked the wrong kind, the fastest fix is to archive the mistaken draft and author a fresh one with the correct kind; there's no in-place conversion from a Recommendation to a Directive, for instance.
The four kinds
Playbook
What it means. The canonical, repeatable, multi-step process for the network. A playbook is the standard way of doing something, meant to stay current and get reused, not a one-time push.
When to use it. Use Playbook when you're documenting a process every franchisee should follow the same way, indefinitely: a storm-response protocol, a new-client intake checklist, a monthly close procedure. If the answer to "will we still be pointing franchisees at this in six months" is yes, it's a playbook.
Example. "Storm surge intake protocol": a numbered sequence covering first contact, documentation requirements, and carrier notification timing, written once and referenced by every location whenever a storm event hits.
Directive
What it means. A one-shot, mandatory action with a deadline. A directive says do this, by this date, full stop; it isn't a process to internalize, it's a single task to close out.
When to use it. Use Directive when there's a specific compliance, safety, or operational action every (or a specific set of) franchisee must complete by a real date: update a certificate of insurance before it lapses, complete a mandated safety training, migrate to a new vendor by a cutover date.
Example. "Update COI documentation by July 31": a short instruction plus the exact deadline, pushed to the franchisees whose insurance record is out of date.
Recommendation
What it means. A soft nudge with no enforcement behind it. A recommendation surfaces as guidance a franchisee can act on, adapt, or set aside; there's no expectation attached that everyone complies.
When to use it. Use Recommendation when you've spotted a pattern in the network's own benchmark data worth surfacing, but you're not mandating a response, for example suggesting a pricing adjustment on a cost category where peer data shows room, or flagging a vendor worth evaluating. It's guidance, not a mandate.
Example. "Consider revisiting labor mix on water mitigation jobs": a note pointing at what the benchmark showed, left for each franchisee to weigh against their own situation.
Experiment
What it means. A limited-scope test at the network level, something you want to try with a subset of the network and evaluate before deciding whether it becomes a playbook or a directive for everyone. Today, choosing Experiment doesn't change how the push modal behaves: you still hand-pick which franchisees receive it from the same franchisee list used for every other kind, and the adoption tracking that comes back (acknowledged, in progress, completed, declined) is the same tracking every kind gets. What Experiment buys you is honest labeling: it tells you and every other admin looking at the board that this one is provisional, meant to be judged before it's scaled, rather than something already decided.
When to use it. Use Experiment when you're testing something with a deliberately narrow group before committing the whole network to it: a new PSA vendor with three pilot locations, a revised estimate-turnaround target with the bottom-quartile group, a new referral-fee structure with a handful of volunteers.
Example. "Pilot new drying-equipment vendor with 4 Southeast locations": pushed only to those four, with the plan to compare their results against the rest of the network before deciding whether to make it a directive.
How kind shows up after you save
Once an initiative is saved, its kind travels with it everywhere it's shown:
- On tiles (Active, In Draft, Recently Completed, Recent Activity rows on the Initiatives page), the kind label sits at the top of the tile, and the tile's accent color is kind-specific so a scan down the row tells you the mix at a glance.
- In the detail view, kind renders as a pill at the top, next to a second pill for the current status (Draft, Active, Paused, Completed, or Archived).
- In the notification bell fired to every other HQ admin the moment a new initiative is created, the title reads "New [kind]: [title]," for example "New directive: Update COI documentation by July 31."
Kind and status are independent
Kind never gates which status transitions are available. Every initiative, regardless of kind, moves through the same lifecycle: Draft can go to Active or Archived; Active can go to Paused or Completed; Paused can go back to Active or to Archived; Completed can only go to Archived; Archived is terminal. A Recommendation and a Directive both start in Draft and both need an admin to click Mark Active before they push anywhere. Kind is about what the initiative is asking of a franchisee; status is about where it sits in its own lifecycle. They're separate questions the platform tracks separately.
The Initiatives page at a glance
At the top of the page, the hero panel reads the count of currently Active initiatives as its headline number, with a pill reading "[N] live," "[N] in draft," or "No decisions yet" depending on what's on the board. Below it, three secondary figures:
- Drafts, how many initiatives are sitting in Draft status, not yet pushed to anyone.
- Completed 90d, how many initiatives were marked Completed within the last 90 days, a rolling window, not a calendar quarter.
- Plan kinds, the count of distinct kinds currently represented across every initiative on the board, regardless of status. If your network only ever authors playbooks, this reads 1; if you've used all four, it reads 4.
Empty states
- No initiatives at all, admin view: "No decisions yet. Click + New decision to author a playbook, directive, or recommendation for the network." (The button itself reads + New initiative on this page; the copy names the underlying action.)
- No initiatives at all, viewer view: "Decisions appear here as the franchisor authors playbooks, directives, and recommendations for the network."
- Active row, nothing active: "No active decisions. Activate a draft from its detail view to deploy it across the network."
- In Draft row, nothing in draft: "No drafts in flight. Start a new decision to author a playbook or directive for the network."
- Recently Completed row, nothing closed in the last 90 days: "No decisions have closed inside the last 90 days. Completed playbooks surface here as proof of network-level action."
- Recent Activity row, nothing yet: "Decisions surface here as you author them."
None of these mean anything is broken. A network with no experiments authored simply won't have one show up in the Plan kinds count or on any row; that's an accurate read of what's been written, not a gap in the data.
Related reading
- Broadcasting to your network, the full authoring, activation, push-to-franchisees, and sponsorship flow this article's plan editor feeds into.
- HQ overview, how Broadcast and Initiatives fit into the rest of the HQ shell.
- Network health, the franchisee directory and adoption reporting an activated initiative pushes into.
- HQ programs, the audit-and-violation side of network compliance, a separate source from initiatives.
- HQ standards, the brand-standard side of network compliance.
- HQ compliance, the rolled-up compliance view across the network.
Data sources
Data sources
- 1.Initiative records: plan kind, title, body, status, timestamps. the network data.
- 2.Template library: kind, category, description, adoption count. the network data.
- 3.Per-franchisee adoption status once pushed. the network data.