Browse
On this page

Set up pods and staff so dispatch works

Add a tech, put them in a pod with the right skill tags, turn on Elise Dispatch Mode, and watch an unassigned ticket land on that pod instead of on you.

What you will have

  • Add a new tech and set their permission group so it actually sticks.
  • Create a pod, choose Owner or Specialty, and give it skill tags.
  • Add a tech to that pod as a member.
  • Turn on Elise Dispatch Mode so tickets can route on their own.
  • Watch an unassigned ticket land on a pod member through the real waterfall, not on you.

Why it works this way

New staff automatically land in this tenant's starter pod, the one carrying the System badge, the moment they're added. That is why Marco Ellis showed up in Help Desk before anyone put him there on purpose.

A client's Default Ticket Assignee outranks every pod. Bluebird Dental in this tenant already had one set from its seed data, so the first ticket in this walkthrough went straight to jordan.reyes no matter what pod or skill tag said; clearing that field on the client's own Settings let the pod actually have a say.

A ticket created from the admin UI self-assigns to whoever created it, the same as claiming it. The pod waterfall only runs on its own for a ticket that arrives another way, client portal or email, or when a person runs Dispatch with Elise on it by hand.

Tier is optional on a pod member. Leave it blank and the pod stays flat, Elise picks by availability, skill, and load; type a number in and dispatch runs a tier waterfall inside that pod instead.

Pods and Dispatch live on separate pages under Team on purpose. Pods is the page techs and managers open daily, member lists, client assignments, tier waterfalls; Dispatch holds slower-cadence config, tag estimates and preview grading, that an operations manager tunes occasionally. Putting both on one page would bury the daily view in knobs almost no one touches.

Elise Dispatch Mode lives on the AI Assistant settings, not on the Dispatch & Scheduling settings its name suggests. Settings > Dispatch & Scheduling > Dispatch Rules shows the same value read-only with an Edit link back to Settings > AI Assistant > Actions & Automation, which is the one page where the value actually saves.

Steps

  1. Open Team > Staff and click Add Staff.

    Northwind Managed Services already runs one pod, Help Desk, with jordan.reyes and renee.castillo in it. Adding your very first hire and inviting them is its own walkthrough (Add your first client and invite your team); this one adds a second tech, marco.ellis, a synthetic hire, and gives him a real place in a pod.

    Open Team > Staff and click Add Staff.
  2. Type the new tech's email address, and see that Add is still greyed out.

    [email protected] is the whole of what this dialog asks for by way of identity; there is no name field. An email on its own is not enough to add him: the Role field, labeled Role (required), opens on "Select a role..." with nothing picked for you, and Add stays greyed out until a real group is chosen. This walkthrough picks Viewer here, the read-only group, and then sets his real permission group from the Staff list in the next step, to show that door too. Whatever you pick in the dialog is the group the account is created with.

    Type the new tech's email address, and see that Add is still greyed out.
  3. Back on the Staff list, set his Group to Help Desk Tech.

    Marco already shows Pod Help Desk before you touch anything else: Help Desk carries this tenant's System badge, so every new hire joins it automatically. Group is the door that actually grants access; jordan.reyes and renee.castillo both carry Help Desk Tech, so Marco gets the same one. Group assigns one of this tenant's permission groups, built-in ones include Full Admin, Service Manager, Senior Tech, Help Desk Tech, Account Manager, Billing Admin, and Viewer, the read-only one, and each one sets a tier, None, View, Edit, or Admin, for every section of the nav; setting a parent section's tier caps every child section under it at that tier or lower. A custom group beyond those built-ins is built in Team > Staff Roles.

    Back on the Staff list, set his Group to Help Desk Tech.
  4. Open Team > Pods and click New Pod.

    Help Desk already shows 3 members, jordan.reyes, renee.castillo, and Marco, and 1 client, Bluebird Dental. New Pod is where a second pod gets built, one that does not own any client of its own but steps in on skill alone.

    Open Team > Pods and click New Pod.
  5. Name the pod, set its Type to Specialty, and fill in Skill tags.

    Escalations is named for the work it exists to catch. Owner pods own clients and are the default routing target; a Specialty pod owns no clients at all, it only receives a skill-routed handoff when a ticket's inferred skill matches its tags. network, security tells dispatch this pod steps in whenever either skill is needed.

    Note: Auto-escalate on SLA risk stays off here, which is the shipped default. Turned on, Elise can escalate a ticket on her own, a tier climb, a cross-pod handoff, or a management surface, the moment its SLA timer is at risk, with no one to approve it first. Off, she proposes the escalation as a card on the Command Center instead and waits for a person to say yes.
    Name the pod, set its Type to Specialty, and fill in Skill tags.
  6. Back on the Pods list, confirm Escalations was created.

    Escalations shows as Specialty with the network and security skill chips, 0 members and 0 clients, since a Specialty pod owns no clients by design. Help Desk stays Owner with its 3 members and 1 client, untouched.

    Back on the Pods list, confirm Escalations was created.
  7. Open Escalations > Members and pick marco.ellis to add.

    Marco is the only tech this new pod needs for now; a tech can belong to more than one pod at once, so joining Escalations does not pull him out of Help Desk.

    Open Escalations > Members and pick marco.ellis to add.
  8. Click Add and confirm marco.ellis is now a member.

    The Members count reads 1. Can triage flags a tech as eligible to clear tickets Elise routed to the triage queue because her confidence was low; Can review flags one who can sign off on Elise's proposed resolutions. Both stay off here since Marco is new to the team.

    Note: Tier is optional. Leave it blank and the pod stays flat, Elise picks by availability, skill, and load; type a number in and dispatch runs a tier waterfall inside this pod instead.
    Click Add and confirm marco.ellis is now a member.
  9. Open Settings > AI Assistant > Actions & Automation and set Elise Dispatch Mode to On (live).

    Off means no auto-dispatch at all. Preview has Elise record who and when she would schedule, with no calendar writes, for grading on the Dispatch > Preview page. On dispatches live, though only for a pod whose own Auto-Dispatch toggle is also on. Settings > Dispatch & Scheduling > Dispatch Rules shows this same value read-only with a link back here; this page is the one place it actually saves.

    Note: Leaving a pod's own Auto-Dispatch toggle off does not turn dispatch off for that pod. It still self-serves the plain waterfall pick; the toggle only decides whether Elise goes on to book the tech's calendar herself.
    Open Settings > AI Assistant > Actions & Automation and set Elise Dispatch Mode to On (live).
  10. Unassign the new ticket, then click Dispatch with Elise.

    Waiting room monitor stuck on setup screen (ticket #5, Bluebird Dental) self-assigned to Northwind Setup on creation, the same admin self-assign every ticket made from the admin UI gets. Setting Assigned To back to Unassigned and clicking Dispatch with Elise, in the Scheduling card, ran the Help Desk waterfall for real: it landed on renee.castillo and booked her calendar that same afternoon.

    Note: Dispatch with Elise is a manual, per-ticket action. It runs the pod waterfall and books the calendar whether or not that pod's own Auto-Dispatch toggle is on; the toggle only gates whether Elise dispatches new tickets for that pod automatically the moment they're created.
    Unassign the new ticket, then click Dispatch with Elise.

If it did not work

  • If Add Staff says "This user is already on staff," that email is already active staff; use the search box on the Staff list to find and edit them instead of adding again.
  • If a pod's Config tab shows no Delete pod button at all, it is this tenant's seeded System pod (Help Desk here); a System pod can never be deleted, no matter how many other pods exist. If the button shows but stays disabled, at least one client still points at this pod as its owner and has to move first.
  • If a client's tickets keep landing on the same one tech no matter which pod or skill tag is set, check that client's Default Ticket Assignee (Clients > that client > Edit > Settings); it outranks every pod until it is cleared.
  • If Elise Dispatch Mode is Off, nothing dispatches on its own; every new ticket falls back to whatever its creating door does, the same admin self-assign ticket #5 got here before it was manually dispatched.
  • If you go looking for pod setup under Settings instead of Team, two doorway pages explain rather than do: Settings > Administration > Pod Configuration and Settings > Dispatch & Scheduling > Pod Scheduling Rules both say a pod's members, boards, tiers, and hours belong to the pod itself, and link straight back to Team > Pods, the same door this walkthrough already used.

Questions this page answers

What is a pod?

A pod is a group of techs that owns clients and receives ticket dispatch. Clients map to one owning pod; when a ticket arrives, the dispatch engine picks a free tech inside that pod (or hands off to a specialty pod if the ticket needs a specific skill). Boards stay as work-types (Help Desk, Projects, etc.) - pods are orthogonal and layer on top as a view filter.

What is the difference between Owner and Specialty pods?

Owner pods own clients and are the default routing target. Specialty pods do NOT own clients - they receive skill-routed handoffs (e.g., "network", "firewall", "m365"). If a ticket needs a skill your owning pod does not carry, the dispatcher hands it off to the first specialty pod whose skillTags match.

What are skill tags used for?

Skill tags describe what the pod can handle. Empty = catch-all (any ticket). Narrow tags (e.g., "firewall, networking") make the pod specialty-shaped - tickets whose requiredSkills match will route here. Elise infers requiredSkills from the ticket subject/body during triage.

What is the "System" badge?

The seeded starter pod (Help Desk on an MSP install) carries the System badge so single-pod shops never see pod complexity. You can rename it, change its skills, and add or remove members, but it can never be deleted, no matter how many other pods you create, its Configuration tab shows no delete button at all. Creating a second pod is an explicit admin action.

Do pods have to use tiers?

No. Pods are flat by default - Elise picks by availability + skill + load. If you opt a member into a tier (T1/T2/T3 or Lead/Member - your choice), dispatch runs a Pax 8-style waterfall inside that pod: T1 tries first, escalates to T2, then T3. Cross-pod members can have independent tiers per pod.

What does "Can triage" on a member do?

It flags a tech as eligible to clear triage-queue tickets (those Elise flagged with needsHumanTriage=true because her confidence was low). When OFF, a tech is a regular dispatch candidate but will not see the triage queue. Useful if you want a Pax 8-style primary/secondary triage rotation within the pod.

What does "Auto-escalate on SLA risk" do?

When ON, Elise can escalate a ticket (tier climb, cross-pod handoff, or management surface) without waiting for a human when the SLA timer is at risk. When OFF (default), Elise proposes the escalation as a Tier 2 action card on the Command Center and waits for approval.

Why can't I delete this pod?

Three reasons a pod can't be deleted: (1) It is the seeded System pod. It is protected and can never be deleted, on its Config tab the Delete pod button does not appear at all. (2) It is the last remaining pod, and you must always keep at least one. (3) Clients still point at this pod as their owner. Here the Delete pod button is shown but disabled, with a tooltip telling you how many clients still point at it. To clear it, either reassign each client to a different pod from that client's detail page (its owning-pod field), or open this pod's Clients tab and remove the client there, removing a client from the pod drops that client to default ticket routing rather than moving it to a specific pod. Once no clients point at this pod, delete it here.

What is the Pod Configuration link for?

This doorway explains that a pod is a team object with its own members, boards, and tiers, so you configure it in the Team area. Settings only holds the tenant-wide dispatch defaults that pods build on.

How do I create or configure a pod?

Click "Open Pods" on this page, or go straight to Team → Pods. Pick the pod to configure, or create a new one. Open its Configuration. Then manage its members, board assignments, and tier coverage.

What are Pod Scheduling Rules?

This doorway explains that scheduling rules, hours, tiers, and how dispatch loads a pod belong to each pod. They live on the pod itself, not on a global Settings page. The tenant-wide dispatch defaults nearby are only the starting point each pod builds on.

How do I set a pod's hours and tier coverage?

Click "Open Pods" on this page, or go straight to Team → Pods. Pick the pod. Open its Configuration. Set its members, boards, tier coverage, and operating hours.

What does Dispatch Mode (off / preview / on) do?

Off = no auto-dispatch; you schedule manually. Preview = Elise records who + when she WOULD schedule (no calendar writes) so you can review her picks on the Preview page - the safe on-ramp. On = Elise dispatches live, per pod, only for pods whose Auto-Dispatch toggle is on. This row only shows the value. It is not editable here. Click its Edit link, or go to Settings > AI Assistant > Actions & Automation, to change it.

Why is Dispatch separate from Pods?

Different audiences and frequencies. Pods is the daily-driver surface - techs and managers check member lists, client assignments, and tier waterfalls often. Dispatch is slower-cadence config the operations manager tunes occasionally. Co-locating them would clutter the daily view with knobs that almost no one touches day-to-day.

When should I turn Auto-Dispatch off?

Auto-Dispatch routes new tickets through pod skill-tag rules and auto-assigns. It is controlled by Elise Dispatch Mode (Off / Preview / On) on Settings → AI Assistant → Actions & Automation, plus each pod's own Auto-Dispatch toggle, not on this page. Set it to Off (or leave pods' toggles off) while you're configuring or tuning pods, otherwise creation runs the matcher before your rules are ready and tickets get assigned to the wrong techs. Existing assignments and explicit reassignments are unaffected.

How does the tier-based permission system work?

Permissions use a tier system: None (no access), View (read-only), Edit (create/modify), and Admin (full control including delete). Tiers cascade - setting a parent section to "View" caps all child sections at "View" or lower. Defaults are defined per Permission Group and can be overridden per user.

How do I add a team member?

Go to Team > Staff and click "Add Staff". Enter the person's email address, choose the permission group that sets their access, and pick a pod if your instance has more than one, then click Add. The group is required: Add stays disabled until you choose one, and the group you choose is the group they are created with. They are added as staff and sent a sign-in link. You can change their access later from the Group column on the Staff list.

Was this helpful?

Last validated 2026-09-17