How Mission Control works. You can read this without signing in. If you are waiting for access, an admin has to approve you first, and you will not get an email when they do. Just sign in again.
This is the user-facing guide to Mission Control (MC). It describes what you see on screen and how to get things done. It is also the knowledge source the in-app assistant uses to answer questions, so it deliberately contains no internal implementation detail.
Written for the people who use MC: service-delivery managers, leads, and team members.
Mission Control is a single pane of glass over ServiceNow and Azure DevOps. It does not replace either system. It reads your tickets and work items, puts the things that need attention in one place, and lets you act on them (reassign, nudge, approve, add a work note) without opening ServiceNow.
Two facts worth knowing up front:
The main areas are Pulse (the view across the queues you look after), My Day (your personal view), Report (the monthly report generator), Settings, and Admin.
Use Sign in with Microsoft with your work account. There is no separate MC password.
Every new person lands on an Access Pending page first: your account is authenticated, but an admin has to grant access. The page shows your email and a Pending badge, plus two buttons:
Gotcha: you are not notified when you are approved. No email, no message. Just sign in again and you are through.
After approval you go through a three-step wizard: ServiceNow → Queues → Azure DevOps.
Gotcha: there is no Skip button. You skip step 3 by leaving it blank.
Gotcha: the wizard does not save as you go. Close the tab midway and your queue picks are lost. Your ServiceNow connection does survive.
Gotcha: you cannot re-run onboarding yourself. Later changes are made in Settings. Only an admin can push someone back through the wizard.
There are four roles. Each level includes everything below it.
| Role | Lands on | Sees in the sidebar |
|---|---|---|
| Team Member | My Day | My Day |
| Lead | Pulse | Pulse, My Day |
| Manager | Pulse | Pulse, My Day, Report, Settings |
| Admin | Pulse | Pulse, My Day, Report, Admin, Settings |
Roles control which MC views you see, not what you can do in ServiceNow or Azure DevOps. Reassigning, approving, and commenting are always governed by your ServiceNow permissions. If ServiceNow allows it, MC allows it.
Gotcha: there is no "access denied" message. Open a page your role does not have and you simply land on My Day with no explanation.
My Day is in the sidebar for every role. It was once hidden from Leads and Managers on the reasoning that their own items also appear on Pulse. That was wrong in practice: Pulse answers "how is the team doing" and My Day answers "what is on my plate", and a newly onboarded manager had no personal view at all.
Everyone starts as a Team Member, however they arrived: signing in with Microsoft on your own, or being created through Admin → Invite user. An admin raises your role deliberately after that. Self sign-in still waits for approval before it gives you anything at all.
Gotcha: a Team Member or Lead who sees "ServiceNow: not connected" is told to connect from Settings, but they cannot open Settings. They need a Manager or Admin to raise their role.
Pulse is the dashboard for the work you look after: a row of action cards, a row of health cards, and panels for workload, changes, requests, and Azure DevOps.
It is not limited to one team. Pulse covers every queue on your desk, so if you look after several assignment groups (your own team plus a queue you are covering, or two teams you manage), Pulse shows all of them together. When more than one queue is in play, lists gain a queue filter and an Assignment Group column, and the health cards break the numbers down per queue, so you can always see which queue a number belongs to.
Two chips at the top right report whether Pulse can read your data. Click either for details and the time it was last verified.
Gotcha: "not configured" for Azure DevOps means no project has been added to this desk. It is a setup state, not a broken connection. "error" is the state that means configured but not reachable.
Gotcha: the "session expired" pop-up message appears only once, at the moment the connection breaks. If you reload the page while already disconnected, the chip is quietly amber and no message appears.
If a Partial data chip appears in the topbar, ServiceNow answered for some things and not others on the last refresh. It names what was missed, for example "Partial data: Enhancement tasks".
What it means for the numbers: the counts on the page are lower than reality, because the tickets MC could not fetch are simply absent. It is the difference between "there are three" and "we could only see three", and without the chip those look identical.
It usually clears on the next refresh. If it keeps naming the same source, that source is refusing MC's query rather than being briefly unavailable, and it is worth reporting.
Gotcha: the chip is about data, not connection. ServiceNow can be connected and green while part of an answer is still missing.
The ↻ Refresh button re-reads everything from ServiceNow, ignoring anything saved from a moment ago, so it is your "show me the truth right now" button. Next to it, a countdown lets you pick how often the page refreshes itself: 30 seconds, 1 minute (default), 2 minutes, 5 minutes, Off, or your own value.
The countdown pauses while a list, ticket panel, or nudge is open, and picks up where it left off when you close the panel.
It keeps running when you switch to another browser tab, so Pulse stays current in the background and is already up to date when you come back. Nothing extra happens on your return; there is nothing to catch up on.
Gotcha: browsers slow timers down in background tabs, so a refresh that comes due while you are on another tab can land up to about a minute late, and a tab the browser froze entirely refreshes the moment you look at it again. The countdown measures real elapsed time rather than counting seconds, so the gap is a short delay, never a stretched interval.
Gotcha: the automatic refresh reuses recently saved data for the radars, Advisor, and trend cards. Only clicking ↻ Refresh forces those to re-read live.
Pulse hides every card and shows "One step left - connect ServiceNow". Nothing on the page lights up with numbers while disconnected, on purpose, so you never read stale counts behind the banner.
Click any card to open a list of exactly those tickets.
| Card | Shows |
|---|---|
| P1 / P2 Incidents | Active P1 and P2 incidents, plus their incident tasks. Enhancement tasks are not counted here: they carry no SLA, and this card is for SLA-bearing triage. They still appear in Unassigned, Stale, and On Hold |
| Unassigned | Incidents, requests, and child tasks with nobody assigned |
| Stale Tickets | Assigned tickets with no work note or comment for N or more business days |
| My Approvals | Approvals waiting on your personal sign-off |
| Enhancements | Active enhancements in your queues, all states |
| Reopened | Active incidents reopened at least once |
| On Hold | Tickets pending or on hold for more than 24 business hours |
| Closed Today | Tickets closed today |
| SLA at Risk | Live resolution SLAs breaching in under 5 hours |
| Hours Charged | Open incidents and requests where the person the ticket is assigned to has booked more hours against it than your threshold |
Gotcha: My Approvals is the one card not filtered to your team's queues. It shows every approval assigned to you personally, even on another team's tickets.
SLA Compliance, Breached, Azure Advisor, Backlog Trend, Reopen Rate, and Change Failure Rate. On desks with more than one queue, each card also shows small per-queue chips so you can see which queue is moving the number.
Stale. A ticket that is assigned to someone and has had no work note and no comment for N or more business days (N is the number on the Stale Tickets card, default 2). Weekends do not count. Activity deliberately excludes reassignment and re-categorisation, so a ticket someone touched an hour ago without doing real work still counts as stale. Tickets that arrived in your group in the last 24 hours are excluded, so new arrivals get a triage grace period.
Critical / P1-P2. Any active incident or child task whose priority starts with 1 or 2. No age or assignment condition.
Unassigned. Any active incident, request, or child task with an empty assignee.
Reopened. An active incident that has been reopened at least once.
Pending / On Hold. A ticket in Pending, On Hold, Awaiting caller, or Awaiting evidence, untouched for more than 24 business hours.
Closed today. Incidents counted by when they were actually resolved (not by ServiceNow's later auto-close), plus requests and child tasks by last update.
SLA at risk. A live resolution SLA that has not breached, on a still-open ticket, whose deadline falls within the next 5 hours.
SLA breached. A resolution SLA that breached within the last 30 days, attributed to the queue that owned the ticket at the moment it breached. Inheriting an already-breached ticket does not count against you.
SLA compliance. Of the SLA outcomes decided in the last 30 days, the share that were met. Once a ticket breaches, that is the outcome even if the ticket is still open. Still-running SLAs are excluded from the percentage.
Change failure rate. Of the non-standard changes closed in the last 30 days, the share that closed Unsuccessful. A change that closed successfully but with a noted issue does not count as a failure here, so writing up a problem honestly never makes your number look worse.
Gotcha: Change Radar's Failed bucket is deliberately wider than this. It also lists changes that closed successfully with issues, because that bucket is a list of things worth looking at rather than a score. So the two can disagree, and that is intended.
The Stale Tickets subtitle is a live control: "no comment for − 2 + business days". Click − or + to change it; the count recomputes immediately. Range is 1 to 30.
Gotcha: this is your own setting. Your desk belongs to you, so changing the number changes what you see on Pulse, not what a teammate sees. It is also the one desk setting a Lead can change.
Gotcha: clicking − or + does not open the list, but clicking the words or the number does.
The Hours Charged subtitle works the same way: "more than − 1 hour + charged". Click − or + to change it in half-hour steps. Range is 0 to 24 hours.
Set it to 0 and the card switches off and disappears from Pulse. Add it back from the card library when you want it.
This card answers one question: is the person working a ticket spending longer on it than the work should take. So it counts only the hours booked by the ticket's assignee, not everyone who touched it. Two people logging an hour each on the same request is not the same thing as one person logging two, and only the second is what this card is looking for.
It covers incidents and requests, which have a single assignee. Child tasks are excluded, because several people can book time against one parent and the reading would not hold.
The Hours charged column in the list shows exactly what ServiceNow shows, to the quarter hour. If ServiceNow says 2.25, this says 2.25.
Nudging from this card asks openly what is taking the time, and names the hours once as a fact. It is deliberately not worded as a complaint: the usual answer is hidden complexity or a dependency someone has not raised yet, and the point of asking is to surface that early enough to help. Leave the snooze checkbox ticked and the row goes quiet for four hours, so you are not chased about it again while the answer is still coming.
Gotcha: like the stale threshold, this is your own setting, not your team's.
Gotcha: the card lists open tickets only, so acting on one removes it from the list. It is not a monthly total.
Every card, workload bar segment, and radar bucket opens the same list, which slides in from the side of the screen.
Click the ticket number to open the ticket detail panel. Esc closes the list.
Gotcha: Nudge is only enabled when everything selected belongs to one person. Reassign refuses a selection spanning two or more assignment groups, because it picks from one group's members. Move has no such restriction and works on a mixed selection, since where the tickets go does not depend on where they are now.
Gotcha: Move takes the tickets out of your queues. They stop appearing on Pulse, your card counts drop, and it clears whoever was assigned, exactly as changing the group in ServiceNow does. You cannot undo it from Mission Control, because once the tickets belong to another group you can no longer see them. MC asks you to confirm before it opens the group picker.
Gotcha: you will not see all three buttons on every card. A button only appears where it could actually work: Reassign is ServiceNow only, so it is absent from an Azure DevOps list; Hide is absent from cards that have no hiding (Enhancements, for one); Nudge is absent from a list where nobody is assigned.
The two states mean different things, and a button never moves between them as you select. Missing means not possible on this card at all. Greyed means not with what you have selected right now, and hovering it says why.
Gotcha: if a list looks empty, check the empty message. "No rows match the current filters." comes with a Clear filters button. "Nothing here right now. The team's clear." means there genuinely is nothing.
Hiding a row (per-row eye icon, or the bulk Hide) removes it from that card's list for a chosen duration: 30 minutes, 1 hour, 4 hours, until you unhide, or a custom period.
Gotcha: hiding also reduces the number on the card. If the card says 4 and ServiceNow says 6, you have two rows hidden. A "N hidden" chip with a Show all button appears above the list.
Gotcha: hides are per card. Hiding a ticket on the Unassigned list does not hide it on the P1 / P2 list.
Gotcha: hiding is MC-only. It changes nothing in ServiceNow.
Adding. The + Add button opens a menu listing every card, grouped into "Available to add" and "On Pulse". Click to toggle. The menu stays open so you can add several.
Hiding. Most cards have an X. This is a soft hide: the card comes back on its own when its number rises above what it was when you hid it. A short confirmation appears at the bottom of the screen with an Undo button, which lasts about 8 seconds.
Gotcha (important): the X and the "On Pulse" menu row are not the same action. The X is a soft hide that self-restores. Clicking the card in the "On Pulse" list is a hard remove that stays gone until you add it back.
Cards that vanish at zero. Most cards only exist while they have something to say. These disappear when their count hits 0: P1 / P2 Incidents, Unassigned, My Approvals, Enhancements, Closed Today, Reopened, On Hold, SLA at Risk, Breached, Azure Advisor, Change Radar, My Requests, Team Workload.
These never vanish: Stale Tickets, SLA Compliance, Backlog Trend, Reopen Rate, Change Failure Rate, and Azure DevOps project cards.
To bring one back: wait (it returns by itself when the count rises above 0), add it from + Add (which pins it on permanently, even at zero), or reset the layout.
Gotcha: the disappearance happens on the next refresh, not the instant the number changes.
Gotcha: on your very first Pulse session a starter set shows even at zero. From the second session those cards follow the normal hide-at-zero rule. So "it was there yesterday and is gone today" is expected, not a bug.
Reset Pulse layout clears every card you hid or pinned plus any saved order, and reloads. There is no undo. Your queues, Azure DevOps projects, drafts, and team data are untouched.
Gotcha: drag-reorder and card resizing are no longer available.
One row per person: a stacked bar of ServiceNow, P1/P2, and Azure DevOps counts, with the number printed in each segment. Click any segment to open that person's tickets.
Gotcha: bar widths are not strictly proportional (a very loaded person is capped so they do not squash everyone else). Read the numbers, not the lengths.
Gotcha: people are matched between ServiceNow and Azure DevOps by email. If someone appears twice, their email is missing on one side.
Change Radar has four clickable counts: Failed (24h), In flight (now), Overdue (past planned end), Tomorrow.
My Requests is scoped to requests you personally raised, not your team: In flight, Done (today), Followup (7d), Aging (14d+).
Gotcha: a change can be both In flight and Overdue. Aging is a slice of In flight. The buckets deliberately overlap.
Gotcha: Standard changes (the pre-approved templates that skip CAB) appear in Overdue only. Failed, In flight and Tomorrow leave them out on purpose, because a queue full of routine templates buries the changes actually worth your attention. Overdue is the exception because it asks a different question, "what has passed its window and stopped moving", and standard changes are the most likely to sit parked. They get raised and run in minutes, so nothing in their lifecycle reminds anyone to close them out.
So if a change is missing from the radar, check its Type first. The ticket window shows it as a pill next to the state, beside P4 - Low and similar.
One panel per project and team pair, with a sprint donut (To Do, In Progress, Done / Ready), a Stale card, and Closed Today.
Gotcha: "stale" means something different on each side. ServiceNow measures the last work note or comment (default 2 days). Azure DevOps measures the last change to the work item (default 3 days).
Gotcha: the Azure DevOps stale threshold and type are per card and per browser. The ServiceNow stale threshold is your own setting: your desk belongs to you, so changing the number changes what you see on Pulse, not what a teammate sees.
Gotcha: the X on an Azure DevOps card is destructive. It deletes the card configuration, not just hides it. You have to add it back from + Add → Add Azure DevOps card.
Managers and Admins only. Settings is one scrolling page with two parts: My Account and My Desk.
Everything saves by itself. There is no Save button. A small pill flashes "Saving…" then "Saved", or "Save failed" with the reason.
Gotcha: a name cannot contain ^ , = < > & or |. Those are ServiceNow query separators, so a name carrying one would change the search rather than be searched for. Saving one is refused with a message.
Gotcha: your display name only affects MC's screens. Work notes you post are attributed to your ServiceNow account.
Gotcha: a Teams nudge is not sent from your own Microsoft account. MC holds one Microsoft connection for the whole workspace, set up by a manager or admin, and every nudge is delivered from that account. So the person receiving it sees the message as coming from whoever connected Teams, not from you. MC records who really sent it in the audit log. Sending is limited to Lead and above for this reason.
Your desk is your permanent workspace. Queues are what is currently on it. Your desk is yours: adding a queue changes what you see, not what a teammate sees.
The topbar pill reads My Desk, or My Desk · N queues when you have more than one. Click it for your queue list, and to add or remove one.
Adding a queue: click the pill, then + Add a queue, and type at least two characters to search your ServiceNow assignment groups. Click a result to add it, or press Enter once the search has narrowed to a single match.
You can only add a group the search actually found. A name you typed yourself is not accepted, and the server confirms the group exists before saving, so a typo cannot become a queue that then silently shows nothing.
Removing a queue: hover the queue in that list and click the small x. You must keep at least one, because a desk with no queues has nothing to show.
Working across teams: add every team whose work you handle. You will then see what is assigned to you in each of them, plus their unassigned work you could pick up. Anyone can manage their own queues. It is not a manager-only setting.
What multiple queues change: the topbar pill gains a count, lists gain a queue filter row and an Assignment Group column, the health cards gain per-queue breakdowns, and workload and the reassign picker widen to cover every queue.
Reordering queues still happens in Settings (Managers and Admins), because order drives some sort priorities. Adding and removing do not need Settings.
Gotcha: adding a queue needs a live ServiceNow connection, because the group has to be found and then verified. Without one the picker cannot add anything.
Under My Desk, ServiceNow sign-in shows Linked or Not linked. Connect once and the connection auto-refreshes for about 100 days. You can Disconnect or Reconnect (refresh credentials) any time.
Gotcha: "Linked" only means MC has your saved sign-in. It is not proof that ServiceNow is reachable right now. The chip on Pulse is live status.
While disconnected, everything that reads or writes ServiceNow stops: ticket lists and counts, search, workload, approvals, reassignment, work notes, nudges, time cards, report data, and queue search. Azure DevOps cards keep working; the two connections are independent.
| Setting | Where it actually is |
|---|---|
| Theme (light/dark) | Topbar theme button |
| Refresh interval | The countdown chip in the Pulse header |
| Stale threshold | The stepper on the Pulse Stale Tickets card |
| Azure DevOps projects | Pulse → + Add → Add Azure DevOps card |
| ServiceNow instance URL | Admin → App settings (Admins only) |
| Pulse layout | Reset Pulse layout on Pulse |
| Hidden rows | The eye icon on the row itself |
| Team roster | Nothing to maintain, derived from your queues |
Gotcha: there are no notification settings. No email digests, no alert sounds, no per-card alerting. Cards attract attention by appearing and by their counts.
Gotcha: apart from your display name, personal preferences (theme, refresh interval, card layout, hidden rows) are remembered by your browser. A different computer or browser starts from the defaults.
Admins only. Five sections: Users & access, App settings, Health, Audit log, Feedback.
Approve sign-in requests, assign roles, and manage access. Filter chips: All, Pending, Approved, Revoked, Admins. There is an Export CSV and an Invite user button.
Row actions depend on status: - Pending: Approve, Deny - Revoked: Reinstate - Approved: Reset onboarding, Force sign-out, Revoke access
What they do: - Revoke access locks the person out and removes every saved sign-in MC held for them (ServiceNow, Azure DevOps, Microsoft). Reversible: they appear under Revoked with a Reinstate button, and sign in from scratch when they return. - Reset onboarding clears their queue setup, memberships, and Pulse preferences, and pushes them through the wizard on next sign-in. ServiceNow and Azure DevOps data are untouched. - Force sign-out invalidates their sessions. Approval status and tokens are unaffected.
Gotcha: Deny and Revoke access are the same action underneath.
Gotcha: Invite user sends no email. It pre-approves the person; you have to share the MC URL with them yourself.
Gotcha: invited users default to Team Member, while people who sign themselves in become Manager. Set the role deliberately.
Gotcha: you cannot change your own role or revoke your own access. Promote someone else to admin first if you need to step down.
Two settings: the ServiceNow URL for the whole MC environment (change only when migrating tenants), and the Audit log buffer size.
Gotcha: the buffer size does nothing until the app restarts.
Read-only snapshot: deploy identifier, database ping, integration status (ServiceNow, Azure DevOps, Microsoft Graph for Teams nudges, Azure OpenAI for report summaries), user counts, and the ServiceNow URL.
Gotcha: "configured" means the settings are present, not that the service answered just now. For live status use the chips on Pulse.
A live tail of what people have done in Mission Control: sign-ins, admin changes, nudges, settings edits. Email addresses are hashed, and ticket text and assistant prompts are never written here.
It is not the record, and it says so. The list lives in memory, so it empties on every restart and deploy, and once it is full the oldest events drop off. The footer tells you what period it actually covers, and warns you when the buffer is full and has started discarding.
That matters because of how the page gets used. If you look for something, do not find it, and conclude it did not happen, you may instead be looking at a window that starts after it happened.
For anything older, the durable record on Azure is Log Analytics
(az-3sh-law-mc-dev), which keeps 30 days and survives restarts.
Buffer capacity is set under Settings, and only changes how much of the live tail you can see. It takes effect on the next restart.
Everything people submit through the in-app assistant, newest first, with When, Type (idea / bug / question), Message, From, and Page.
Gotcha: the submitter never picks a type. MC classifies the message from its wording, so expect the occasional mislabel.
Gotcha: what you see depends on your role. An admin sees every submission, because triaging the queue is an admin job. A manager sees submissions from people in their own workspace only.
Gotcha: it is a read-only list. No mark-resolved, no reply, no filters, no export. Triage happens outside MC.
My Day is your own work: what is assigned to you, what has gone quiet, what your team has left unclaimed, and what you logged time against this week. It covers ServiceNow only, and it is scoped to your desk's queues.
| Card | What it counts |
|---|---|
| My Open Tickets | Every active ticket assigned to you across your desk's queues: incidents, requests, and tasks. Not just the flagged ones. |
| My Stale Tasks | Those with no work note or comment for the desk's staleness window |
| My P1 / P2 | Those at priority 1 or 2 |
| Closed Today | Tickets assigned to you that closed or resolved today |
Who counts as "you": the ServiceNow account you connected with, matched on your ServiceNow identity, not on your name. Two people with the same display name can never see each other's tickets.
Click any card to open the list of exactly those tickets, with the same search, filters, and Export CSV as anywhere else.
Gotcha: changes (ticket numbers starting CHG) and enhancement records (starting ENHC) are not counted on My Day. It covers incidents (INC), requests (RITM), and tasks (TASK and ENHT) only.
Two tables under the cards. My active tickets is sorted urgent-first: P1 and P2 at the top, then anything stale, then most recently updated. That is the order to work down.
Gotcha: these panels disappear entirely when empty rather than showing an empty message. A quiet day looks like four zero cards and My week, nothing else.
Active tickets in your desk's queues with nobody assigned. Assign to me claims one: it writes the assignment to ServiceNow, then the page reloads so the ticket moves out of the pool and into your list, with the cards bumped.
Whether you are allowed to take a ticket is decided by ServiceNow, not MC. You must be a member of the ticket's assignment group.
A read-only mirror of your ServiceNow time cards. Time is entered in ServiceNow; there is nothing here to type into and no submit. Its job is recall.
Weeks run Sunday to Saturday. Navigate with ← and →, and This week appears once you have moved away. The grid shows a Work column, seven day columns, and a Total, with a totals row at the bottom. Each row shows the ticket number as a link, the time card's comment (what people actually type to describe the work), and the category.
Gap rows ("Worked on this, no time logged") are greyed rows for tickets MC saw you working on this week that have no time card. They appear on the current week only, and are capped at 8.
Gotcha: no target hours are shown. MC has no configured weekly expectation, so it deliberately shows only what was logged.
Focus bullets written from your own numbers, not by AI. They only appear when there is something to say, so on a quiet day the card is absent. The same bullets appear in Houston's "Today's focus".
| My Day | Pulse | |
|---|---|---|
| Audience | You | A lead or manager, looking across the queues they cover |
| Scope | Assigned to you, plus the unassigned pool | Every queue on the desk, which may be more than one team |
| Cards | Fixed set | Add and remove cards |
| Systems | ServiceNow only | ServiceNow and Azure DevOps |
| Refresh | Manual | Auto-refresh with a selectable interval |
Gotcha: My Day has no auto-refresh timer. It does update itself about a second after you change a ticket from inside MC, so a ticket you just took appears without a manual Refresh.
Writes to ServiceNow, immediately, as you, with no undo: posting a work note or comment; assigning or reassigning; changing the assignment group; approving or rejecting; and the nudge audit note (group nudges only). Each of these is recorded under your own ServiceNow account, so your name appears on it in ServiceNow's history.
Mission Control only, never touches ServiceNow: hiding or snoozing a row; hiding a card; selection, filtering, sorting, and Export CSV; the nudge message itself (that is a Teams chat); anything typed into Houston; and unposted drafts.
Nothing in MC ever writes to Azure DevOps.
Click the ticket number in any list. (The arrow icon next to it is a different control: it opens the record in ServiceNow.)
There are no tabs. It is two panes: context on the left (details grid, description, additional information, planning and risk for changes, approvers) and Activity on the right. Drag the divider to resize; your setting is remembered in that browser.
Reading the Activity feed. Notes keep the shape they were written in. Numbered and bulleted lists, indentation and tables stay laid out instead of being flattened into a single paragraph. A long entry is collapsed behind Show more, so one 40 line note does not bury everything below it, and Show less folds it back. Links that ServiceNow put in a note (the "you can find the chat record here" kind) open the ServiceNow record in a new tab. If MC cannot work out where a link was meant to go, it shows the text without making it a link, rather than sending you somewhere wrong.
Previous / Next arrows (and the ← → keys) walk the list you came from, exactly as you filtered and sorted it.
Gotcha: the arrows are hidden when you opened the ticket any way other than from a list, for example from global search.
Gotcha: there is no SLA, related-items, or attachments section here. SLA lives on the Pulse health cards.
Two buttons at the top of the composer:
Post with the Post button or Ctrl+Enter. Adding a note also stops the ticket counting as stale on the next refresh.
Gotcha: the only thing that changes when you switch is the small "Internal" / "Customer-facing" label under the box. Read it before posting; a customer comment cannot be unsent.
Drafts are kept automatically per ticket and per kind, surviving a reload, and cleared once you post. There is no "draft saved" indicator, and drafts stay in the one browser.
Pasting a table. Paste a table from ServiceNow, Excel, or a web page and it becomes a tidy aligned text table. It needs at least 2 rows and 2 columns; ordinary prose pastes untouched.
Opens from the chat-bubble icon on a ticket row, the same icon in the ticket panel, or Nudge N on the bulk bar. It sends a Teams chat. Email and Both are visibly disabled: only Teams is wired.
Who can send one: Lead, Manager and Admin. This matches where the buttons already are, since Pulse and its lists are Lead and above. See the gotcha under "Your profile" for why sending is restricted.
The assignee is filled in for you. Add more people in "Add a teammate…". With two or more recipients, Send as one group chat is on by default; turn it off for private 1:1s. A summary line always states exactly what will happen.
Gotcha: if a name cannot be matched to exactly one person the chip stays amber with no email and Send stays disabled. This is deliberate: a nudge once went to the wrong person of the same name. Type the address or remove the chip.
The draft is written for you, tailored to the situation. Try another rewrites it. With Azure OpenAI enabled that is a genuinely AI-written alternative; otherwise you get a template variant. The badge tells you which: Draft means AI, Suggested draft means template. Once you edit the message, Try another will not overwrite your text.
The two footer checkboxes: - Snooze this row for 4 hours after sending. Only visible when you opened the composer from an action card. Ticked by default on most cards, because chasing something and then wanting quiet is the usual pattern. Left unticked on the cards where four hours of silence could cost you: SLA at Risk, P1 / P2 Incidents, My Approvals, and every Change Radar bucket except Tomorrow. A ticket breaching in three hours is too important to hide, and an approval only you can sign is your own bottleneck, so on those the tick is yours to make rather than ours to assume. - Save nudge to SNOW work notes. On by default. Only appears, and only writes, for a group chat with 2 or more recipients. Private 1:1 messages are deliberately kept off customer-facing tickets.
Gotcha: so an ordinary nudge (one person, one ticket) leaves no trace in ServiceNow at all.
Gotcha: the "you nudged this recently" bar is advice, not a limit. There is no enforced cooldown, and the bar is hidden for multi-ticket or multi-recipient nudges.
The bar counts every nudge you have sent on that ticket to that person, from anywhere in MC. It is one shared history, not one per card, so a nudge sent when you assigned the ticket still shows up later in the SLA at Risk panel.
Gotcha: if the time cannot be worked out, the bar says you nudged them "before" rather than guessing. Until September 2026 it could show a recent nudge as "1 min ago" when it was in fact hours old.
Bulk nudge only works for one assignee. A selection spanning several people is refused.
From the pencil next to the assignee, by clicking the Assigned to value in the ticket panel, or Reassign N on the bulk bar.
The picker searches the ticket's group members. Names starting with what you typed are listed first, then names that merely contain it.
You can only pick active members of the ticket's own assignment group. This is enforced in the list, in the search, and again on the server. There is no override. To assign someone else, change the group first.
The group picker is two steps: clicking a group only stages it; nothing is written until you press Save.
Gotcha: changing the group clears the current assignee. Expect to assign someone straight after.
Gotcha: a single assign happens on one click, with no confirmation and no undo. (The group picker, by contrast, makes you press Save.)
Gotcha: Azure DevOps items cannot be assigned from MC at all.
The ticket panel shows a green Close note card in Activity only when the ticket is actually closed, resolved, completed or cancelled.
Gotcha: ServiceNow keeps close notes in a field, not in the activity trail, and people often write them before the work is signed off. A ticket that is still open can therefore already hold closure text. MC does not show it, because presenting it would read as though the ticket had been closed.
Gotcha: if ServiceNow has not recorded who closed the ticket, the card is headed just Close note with no name. MC will not fill that gap by guessing. Until September 2026 it fell back to whoever last edited the record, which could attribute a note to somebody who only approved or touched the ticket.
Requests (RITM) carry the answers given on the catalog form. The ticket panel shows them under Additional information, in the order the form asked them, and skips any the requester left blank.
Gotcha: answers that only repeat something already on screen are left out. ServiceNow copies the main question into the ticket Description when the form is submitted, and stores Requester both as a field and as a form answer, so those two would otherwise appear twice in the same panel. Both are still shown in their own places higher up.
Gotcha: for drop-down answers MC shows the value ServiceNow stored, which is usually the same as the label on the form but is not guaranteed to be. Where MC is allowed to read the form's choice list it shows the label instead.
The My Approvals card opens "Awaiting your approval", the only list with a Your approval column and inline Approve / Reject buttons.
From the list you get a confirmation first. From inside the ticket panel there is none: one click sends it.
Gotcha: you cannot add a reason. There is no comment box on approve or reject. If a rejection needs an explanation, add it in ServiceNow.
Gotcha: approving or rejecting cannot be undone from MC, and there is no bulk approve.
After you approve or reject from inside the ticket panel, the panel re-reads the ticket from ServiceNow. While it does, the state pill and the approver rows turn grey for a moment and then settle. That is the only thing that happens: the panel stays open and nothing else on the page reloads.
Gotcha: ServiceNow does not apply an approval instantly. If it has not caught up by the time MC looks, the panel keeps showing what you did rather than appearing to undo it, and checks once more. Closing the panel straight after clicking is fine; so is paging to another ticket, which cancels the check.
MC does not gate approvals by role; ServiceNow's permissions decide. The buttons only appear if MC can match your approver record on that ticket.
The eye icon just after the ticket number (it appears on row hover). Choose 30 minutes, 1 hour, 4 hours, a custom period up to 7 days, or Until I unhide.
The row leaves that list and the card's count drops until the timer expires, then it returns by itself.
Gotcha: a hide applies to one card only. Hiding a ticket on Stale does not hide it from Pending/On Hold.
Gotcha: hiding is MC-only and writes nothing to ServiceNow. Nobody else's view changes.
Recover with the "N hidden" chip and Show all, then click the eye on any revealed row. There is no "clear all hides" button.
Gotcha: the X you may be thinking of is on whole Pulse cards, not ticket rows.
The round button in the bottom-right corner. It has two parts: Today's focus (the same bullets as the dashboard, refreshed with it, which you cannot edit or dismiss) and a box where you can ask a question, share an idea, or flag a bug.
Houston can answer questions about how Mission Control works ("can I export this?", "what counts as stale?", "how do I add a queue?"). It answers from this guide, and it takes your role into account, so it will not walk you through a page your role cannot open.
What it deliberately will not do:
When it does not know, it offers you two choices, and nothing is recorded until you pick one:
This box is where ideas and bugs are submitted. Type them here, and Houston offers to pass them to the team. There is nowhere else to submit them. (Admins read what has been submitted under Admin → Feedback, which is a read-only list and is not open to everyone.)
Gotcha: Houston is only available once you are signed in. If you cannot get in yet, the same information is on the help page, which needs no account: there is a link on the sign-in page and on the "awaiting approval" page.
Gotcha: Houston hides while a nudge panel is open, so it does not cover the Send button. It returns when you close the nudge.
Managers and admins only.
Pick a Month, Year, and optionally a ServiceNow group, then click Generate Draft. MC pulls the month's data from ServiceNow and renders six sections, editable in place:
Then Download produces a Word document (.docx) with a title page, a real table of contents, and those six sections.
Gotcha: there is no separate preview. The draft on screen is the document, and only what is on screen when you click Download goes into the file.
Gotcha: the table of contents is empty until Word fills it. Word usually offers to update fields on open; if not, press F9.
Incidents and Requests each show two blocks, answering different questions. They are separate on purpose and should not be added together.
Raised in
Activity in
Gotcha: "Open now" in the activity block means right now, not at month end. It is the only figure in the summary that moves after the month closes.
Enhancement Requests works the same way it always has: three populations (new, closed, active backlog), each with its own breakdown that adds up.
A fresh draft starts with an empty project table. MC never carries your portfolio forward by itself. Three ways to fill it:
Timeline is a status word (In Progress, Completed, Blocked, Scoping, Ready, Recurring), not a date. The prose most people think of as status belongs in Key Updates / Next Steps, which is what prints in the final column.
Projects linked to an enhancement get a Suggest button that pulls recent ServiceNow work notes, and the section header has Suggest All from ServiceNow. Suggestions never overwrite silently: you get Use this or Dismiss.
Gotcha: removing a project has no confirmation.
Section 5 bullets are AI-written from each project's ServiceNow notes. The Generate from Projects button rewrites them.
Gotcha: Generate from Projects replaces every bullet in section 5. Any hand-written edits there are lost. Add your projects first, then generate.
Section 6 always starts empty. Fill it with + Add Bullet or from the Suggested Pipeline list.
Gotcha: section 6 has no regenerate button. Only section 5 does.
Section 5 bullets carry no ticket number. They are written for senior managers and read as prose about what improved. Section 6 bullets do keep the ENHC number, because each one commits to a named enhancement.
In the downloaded document both sections are numbered lists, and each starts again at 1.
Deleting all the bullets in a section is permanent for that draft; they will not reappear in the download.
Drop a .docx (max 2 MB) into the carry-forward zone. It fills sections 4, 5, and 6 only: projects, improvement bullets, and focus bullets. The metrics in sections 1 to 3 always come fresh from ServiceNow.
Clear import appears after a successful import and restores what those sections held before. There is no confirmation.
Section 3 has three upload zones for ServiceNow Performance Analytics screenshots (open workload, incident trend, request trend). Click, drag, or press Ctrl+V anywhere on the page.
Gotcha: the whole submission is capped at 2 MB, so an oversized screenshot surfaces as a failure at Download time, not at upload. Crop or compress.
Your draft saves itself. A second or so after you stop typing, everything on the page is written to your browser: the bullets, the project rows, the breached reasons, and the screenshots. The toolbar shows the time it last saved.
Come back later, pick the same month, and the draft is restored with a note saying when it was from. If you meant to start over, Start fresh on that note clears it, and Clear draft in the toolbar does the same at any time. Clear draft is greyed out when there is nothing saved, which is how you can tell at a glance whether a draft is waiting.
The draft survives reloading, closing the tab, signing out, and the app being updated underneath you.
The draft follows you to another computer. Sign in elsewhere, pick the same month, and the writing is there. It is still a safety net for coming back to unfinished work rather than a place to store a report long term, so download the DOCX when you are done.
Gotcha: screenshots stay on the computer that took them. Pictures are far larger than text, and sending them would crowd out every other setting you have. So the words travel and the images do not: open the draft elsewhere and you will find your writing intact with the screenshots missing. Add them on the machine you finish on.
Gotcha: clearing your browsing data still removes the local copy, and a very large draft stays on the device that made it rather than syncing. If that happens the browser console says so by name; nothing is lost, it just does not travel.
What else follows you between computers. The cards you pin and hide on Pulse, their layout, your column order, your auto-refresh interval, hidden Azure DevOps cards, the stale task thresholds, and your theme. Sign in anywhere and Mission Control looks the way you left it.
Gotcha: until October 2026 several of those settings were saved only in the browser you set them in, with nothing on screen saying so. If a setting you had changed months ago appears to reset itself once, that is this being corrected, and it will hold from then on.
Gotcha: each month has its own draft. Switching the month picker switches which draft you are looking at, so work on April is never overwritten by work on May.
Gotcha: the chip in the panel header looks like a save indicator. It is not. It only mirrors the status line. The saved-at time next to Clear draft is the real one.