Thread partner change management guide

Updated by Jake Gipson

Rolling out Thread to your team and customers

Thread works best when techs trust it before customers see it, and when customers understand it as an upgrade rather than a replacement. This page covers the rollout framework, the email and script templates, and the reference tables you'll need along the way.

The framework has four stages — start internal, roll out in waves, define AI vs. human ownership, and share proof quickly. The same four stages apply whether you're launching Thread for the first time or expanding into a new product like Voice AI or Triage Agent.

Product suite

Product

What it does

Who it affects

Thread Messenger

White-labeled chat widget embedded in customer portal or desktop. End users submit tickets via real-time chat.

End users

Thread Inbox

Unified service team workspace — manage all tickets, custom views, and real-time team collaboration.

Service technicians

Triage Agent

AI auto-responds, triages, and resolves incoming chat and email 24/7. Handles 10–25% of issues with zero tech involvement.

End users + techs

Voice AI

Automated phone call intake; answers, transcribes, creates PSA tickets. Three modes: Minimal (passthrough), Overflow, Full IVR.

End users + techs

Reminder Agent

Auto follows up with unresponsive customers. Sends up to 5 personalized reminders and logs activity internally.

End users + techs

Magic AI Automations

Auto Prioritization, Auto Categorization, Auto Title, Recap Templates, Sentiment Analysis.

Service techs (back end)

Client Intelligence

AI-generated knowledge base from resolved tickets. Surfaces resolution suggestions. Includes Ask Magic for QBR reporting.

Service techs

Contact Intelligence

AI memory per contact — history, devices, past resolutions surfaced on every new ticket.

Service techs

Companion Apps

Microsoft Teams and Slack integrations — techs can manage and reply to tickets without leaving their chat apps.

Service techs

Supported PSA integrations: ConnectWise PSA, Autotask PSA, HaloPSA.

Part 1: Internal rollout

If engineers don't trust the AI, they'll bypass it — and that creates friction and degrades service. The internal rollout exists to prevent that.

Stage 1: Start internal

Techs need to see the AI work before customers do. Run internal tests, let the team see Contact Intelligence surfacing context, and have them experience Triage Agent handoffs firsthand. Do not activate the Triage Agent for all customers on day one — pilot with one or two low-complexity clients first.

New customer setup

  • Navigate to the Thread workspace setup link from your Customer Success Manager
  • Enter email, verify the one-time passcode, and create your workspace
  • Upload company logo and set workspace name
  • Connect your PSA (ConnectWise, Autotask, or HaloPSA) — follow the integration guides
  • Set default Service Board and ticket sources in Thread settings
  • In Thread Inbox, create your first Views ("All Tickets" and "Triage")
  • Configure Status Mapping for workflow automations
  • Enable Magic AI Automations: Auto Title, Auto Categorization, Auto Prioritization
  • Customize bot name, avatar, and primary brand color in Messenger settings
  • Add service team members and set permissions
  • Set up the Teams or Slack Companion App
  • Run internal test tickets to verify PSA sync
  • Review AI auto-categorizations for accuracy — target is 96%+

Existing customer upgrades

When adding a new product to an existing deployment:

  • Identify which products are being activated (Voice AI, Triage Agent, Reminder Agent, Intelligence)
  • Review release notes with your Thread CSM
  • Update Views in Thread Inbox to surface new ticket sources (e.g., a "Voice AI" source filter)
  • Brief the service team on new AI feature behavior before activation
  • Confirm PSA source mapping for new channels
  • Run internal tests before customer-facing activation

Stage 2: Roll out in waves

A controlled customer set is safer than a big-bang launch. Expand as you refine processes and build confidence.

Wave

Who

Products added

Timing

1 — Pilot

1–3 tech-savvy, low-complexity customers

Thread Messenger + Inbox

Week 5–6

2 — Expansion

20–40% of customer base

+ Triage Agent (chat & email)

Week 7–8

3 — Scale

Remaining customers

+ Voice AI (Overflow mode) + Reminder Agent

Week 9–10

4 — Full AI

All customers

+ Voice AI Full IVR + Client/Contact Intelligence

Week 11+

For each wave:

  • Select the customer(s)
  • Send the customer announcement email (templates below)
  • Configure Thread Messenger branding for that customer — bot name, colors
  • Set customer-specific PSA routing in Thread Client Settings
  • Enable Triage Agent and configure intents and responses
  • Monitor AI responses closely for the first 48–72 hours
  • Collect tech feedback in daily standup
  • Adjust before the next wave

When upselling a new Thread product to an existing customer, treat it like a mini Wave 1 — brief the team, pilot internally, communicate to the customer, then activate.

Stage 3: Define AI vs. human ownership

Explicit ownership prevents confusion. Use the matrix below to clarify what the AI handles, what the tech handles, and what triggers a handoff.

Component

AI's role

Human tech's role

Escalation trigger

Triage Agent

First response, clarifying questions, categorization, simple resolutions

Complex troubleshooting, emotional customers, multi-system issues

Sentiment is angry, issue unresolved after 2 rounds, customer requests human

Voice AI

After-hours greeting, message taking, ticket creation, transcription

Review transcribed tickets, call back VIP clients

Keywords "Emergency," "Critical," "Down" — immediate human callback

Reminder Agent

Send up to 5 AI-personalized follow-up messages, log activity internally

Review tickets with no response after 5 attempts, decide to close or escalate

After 5th reminder, ticket auto-updates to "Awaiting Close"

Magic AI Automations

Auto-assign Title, Category, Priority, Type/Subtype to all tickets

Review and override AI assignments when incorrect

Repeated miscategorization — submit feedback via Thread

Contact Intelligence

Surface contact history, devices, past resolutions on every thread

Use context to personalize communication and jump to the fix

Contact not found in PSA — tech manually adds context

Client Intelligence

Generate KB articles from resolved threads, surface suggested resolutions

Review generated articles, add static client notes, use Ask Magic for QBRs

When to escalate

Send the ticket to a human tech if any of the following are true:

  • The issue is complex or multi-system
  • The customer is frustrated (negative sentiment)
  • Triage Agent failed to resolve after 2 exchanges
  • The caller uses panic words ("Emergency," "Critical," "Down") — flag priority

Otherwise, the AI continues. When the AI hits its limit, use the Flag for Human Review view in Thread Inbox and add an internal note so the next tech has context.

Stage 4: Share proof quickly

Demonstrate value early to build stakeholder buy-in. These are the metrics worth pulling and where to pull them from.

Metric

Industry avg

Thread target

Where to measure

Time to initial response

45 min – 4 hrs

< 1 minute (AI instant reply)

Thread Inbox reporting

Triage accuracy

50–65%

96%+

Compare PSA ticket data before/after

Issues auto-resolved

0%

10–25%

Triage Agent resolution logs

Tech dispatch time saved

Baseline

30% reduction

Track hours in PSA before/after

After-hours tickets created

0 (lost to voicemail)

100% captured

Voice AI source filter in Inbox

Customer response rate

Track baseline

+30% improvement

Reminder Agent follow-up logs

CSAT score

Track baseline

Improvement trend

CSAT via Recap Templates

Using Ask Magic for reporting

Ask Magic generates real-time metrics from inside Thread. Useful prompts:

  • "Act as a senior MSP analyst. Analyze ticketing data from [Month] and generate a QBR covering top issues, time saved, and trend analysis."
  • "What are the most common issues for [Client Name] in the last 30 days?"
  • "How many tickets were auto-resolved by the Triage Agent this week?"
  • "Show me all tickets where Sentiment was negative last month."

Sharing wins internally

Post weekly AI win summaries in your team Slack or Teams channel. Example:

AI Win This Week: Triage Agent auto-resolved 47 tickets, saving the team 12 hours of dispatch work. Voice AI captured 8 after-hours calls that would have been missed.

Tech daily workflow

  1. Open Thread Inbox at the start of your day.
  2. Review the Triage view — check AI-prepped tickets waiting for tech action.
  3. Check the Contact Card for Contact Intelligence data before formulating your response.
  4. Use Ask Magic for historical context on complex client environments.
  5. Use Recap Templates for rapid resolution notes and CSAT dispatch when closing.
  6. Flag errors in the team feedback channel so management can tune the AI.

A few quick rules:

  • Voice AI tickets have source = "Voice AI" — review transcript before calling back.
  • If Triage Agent is looping, escalate manually and notify your manager.
  • Check Sentiment score on the thread — red or orange means handle with extra care.
  • Use @AskMagic inside a ticket to get suggested resolutions instantly.

Internal email templates

Use these to communicate Thread changes to your service delivery team. Customize the bracketed fields.

Initial announcement (new customer)

Subject: Exciting News – We're Upgrading Our Service Desk with Thread

Team,

We are thrilled to announce that we are partnering with Thread to upgrade our service desk operations. As we scale, we want to ensure you have the best tools available to do your jobs effectively, without being bogged down by manual triage and data entry.

What is Thread? Thread is an AI Service Desk platform that integrates directly with our PSA. It includes a unified Inbox for you to work from, and AI Agents that will handle initial triage, information gathering, and follow-ups.

What this means for you:

  • Less time spent on manual ticket categorization. Magic AI handles it automatically.
  • More context provided automatically (devices, past issues) before you even talk to the client via Contact Intelligence.
  • Faster resolutions, with the Triage Agent handling routine information gathering so you can focus on complex fixes.

We will be rolling this out in phased waves so we can learn and adjust. Next week, look out for an invitation to join our new workspace and accept the calendar invite for training.

Best, [Manager Name]

Training kickoff (new customer)

Subject: Thread Training Starts [DATE] – Here's What to Expect

Hi Team,

Our Thread enablement training officially kicks off on [DATE]. It is critical that everyone attends so we can provide a unified, exceptional experience to our clients.

Training schedule:

  • Session 1: Navigating Thread Inbox, Custom Views, and the Companion Apps (Teams/Slack).
  • Session 2: Understanding Magic AI Automations and Contact Intelligence.
  • Session 3: Workflow deep dive — handling handoffs from the Triage Agent and Voice AI workflows.

Action required: Please accept the calendar invites for these sessions today. You can also preview the platform documentation at https://docs.getthread.com.

Thanks, [Manager Name]

Go-live announcement (new customer)

Subject: We're Live! Thread is Now Active for [Customer Name]

Team,

Great news! We have officially gone live with Thread for our Wave 1 pilot customer, [Customer Name].

Starting today, tickets from this client will flow into the Thread Inbox. The Triage Agent is actively greeting their users, collecting information, and summarizing issues before routing them to your queue.

Reminders for Week 1:

  • Monitor the Triage view in Thread.
  • Review the AI-generated summaries and Contact Intelligence data before replying.
  • If you notice the AI looping or needing adjustment, drop a note in our internal feedback channel immediately so we can tweak the intent parameters.

Let's show them an amazing experience!

Best, [Manager Name]

Feature upgrade (existing customer)

Subject: [Customer Name] Has Been Upgraded – New Thread AI Features Active

Hi Team,

We've just activated new Thread AI capabilities for [Customer Name]. They are now live on [Voice AI / Reminder Agent / Triage Agent].

How behavior will change:

  • If Voice AI is active, look for tickets with the source "Voice AI". Always review the transcript text attached before calling the client back.
  • If Reminder Agent is active, you no longer need to manually chase stalled tickets. The system will send up to 5 automated reminders and update the internal notes. Intervene only if the ticket reaches "Awaiting Close."

Keep a close eye on these tickets this week and report any anomalies.

Thanks, [Manager Name]

Weekly win

Subject: AI Win of the Week – [Team Name]

Team,

A quick shoutout and update on how our Thread AI tools are performing this week. The results are fantastic.

Metrics for the week:

  • Auto-resolved: The Triage Agent successfully resolved 47 tickets without any human intervention.
  • Time saved: Categorization and data entry automations saved us an estimated 12 hours of dispatch work.
  • After-hours captures: Voice AI successfully captured and transcribed 8 calls over the weekend, creating P2 tickets for the Monday morning shift automatically.

Excellent job adapting to the new workflow. Let's keep the momentum going!

Best, [Manager Name]

Part 2: Customer-facing rollout

Transparency builds trust. Don't hide AI from customers — position it as an upgrade that benefits them.

Stage 1: Pre-launch preparation

Before sending any customer-facing communication:

  • Customize Thread Messenger branding — bot name, colors, logo
  • Pick a friendly bot name like "Aria" or "Max" — avoid "AI Bot"
  • Prepare an FAQ document (template below)
  • Brief account managers on how to discuss AI with clients
  • Prepare a Quick Start Guide with your branding
  • Identify a Champion Contact at each client — the person who will drive adoption
  • Draft and schedule the announcement email for 3–5 days before go-live
  • Set up CSAT via Recap Templates to measure satisfaction post-launch

Account manager script

"We've invested in upgrading our support platform with AI-powered tools. This doesn't replace our team — it makes them faster. When you reach out, our assistant collects the details our engineers need so they can jump straight to solving your problem. You'll notice faster responses and fewer back-and-forth emails."

Stage 2: Wave rollout

Pilot customers should be:

  • Tech-savvy or from digital-native companies
  • Low-complexity (reduces risk during pilot)
  • Friendly and communicative
  • Not mission-critical accounts with low tolerance for change — save those for Wave 3+

Pilot process:

  • Select 1–3 pilot customers meeting the criteria above
  • Call the primary contact before sending the announcement email
  • Send the Introduction email 3–5 days before go-live
  • Send the Quick Start email on go-live day
  • Follow up personally after 7 days to gather feedback
  • Document feedback and adjust bot responses or workflow as needed
  • Approve expansion to Wave 2

Stage 3: Define AI vs. human (customer-facing)

Language to use vs. avoid

Use

Avoid

"Our AI assistant will handle your initial request to gather the details our team needs — this means faster resolutions for you."

"Our bot will answer your calls." (impersonal)

"You can reach a human engineer at any time by typing 'speak with an engineer' in the chat."

"This is an automated system." (cold, frustrating)

"Our Voice AI is available 24/7 — even when our team is offline, your issue will be logged immediately."

"AI will handle your tickets." (sounds like a service reduction)

What customers should expect

Scenario

What customer experiences

Human involvement

Submitting a chat ticket

Instant AI response, 1–2 clarifying questions

Human assigned after triage

Calling after hours

AI greets, takes message, creates ticket

Human reviews transcript next morning

Not responding to open ticket

Receives friendly AI follow-up reminder

Human intervenes after 5 attempts

Complex multi-system issue

AI gathers initial info, escalates

Human takes over with full context ready

Frustrated or upset contact

AI continues but Sentiment Alert fires

Human immediately notified to step in

Stage 4: Share proof with customers

QBR talking points

  • "Since deploying our new AI support system, your team's average first response time dropped from [X hours] to under 1 minute."
  • "Our AI handled [X] after-hours calls last quarter — that's [X] issues logged immediately vs. waiting for business hours."
  • "We're now auto-resolving [X]% of your lower-complexity tickets, freeing our team to focus on your critical needs."
  • "[X] follow-up reminders were sent automatically for stalled tickets — keeping your team productive without manual chasing."

CSAT framing

"After each resolved ticket, we now automatically send a brief satisfaction survey. This helps us continuously improve the support we deliver to your organization."

Customer email templates

Introduction (new customer)

Subject: Introducing a Better Way to Get IT Support — [Your Company Name]

Hi [Customer Name],

We are always looking for ways to provide you and your team with faster, more efficient IT support. That's why we're excited to introduce our new, real-time support platform!

Starting on [Date], you will have access to our new Support Chat directly from your desktop. This means no more waiting on hold or wondering if your email was received.

The benefits to your team:

  • Instant access: Open the widget and start a chat immediately.
  • Smart triage: Our intelligent assistant will ask a few quick questions to gather necessary details before connecting you to an engineer.
  • Faster resolutions: By gathering info upfront, our team can solve your issue the moment they step into the chat.

We will send a Quick Start Guide next week with instructions on how to use it. We can't wait for you to experience this upgrade!

Best regards, The Team at [Your Company Name]

Quick Start / go-live (new customer)

Subject: You're all set! Here's how to reach us instantly

Hi [Customer Name],

Our new Support Chat is officially live for your team!

3 steps to get started:

  1. Look for the [Bot Name/Icon] icon in your system tray or desktop portal.
  2. Click to open the chat window.
  3. Type your issue or request.

Our digital assistant will greet you, confirm the details, and automatically route your ticket to the right engineer. You can check the status of your ticket directly in the chat window at any time.

Need a human right away? Simply type "speak to an engineer" at any time to bypass the assistant.

If you have any questions, feel free to reply to this email — or better yet, test out the new chat!

Best regards, [Your Company Name]

Triage Agent upgrade (existing customer)

Subject: We've upgraded your IT support experience

Hi [Customer Name],

To ensure we continue providing you with the fastest response times possible, we've upgraded our support systems with intelligent triage capabilities.

What's changing?

When you submit a request via chat or email, our new AI assistant will instantly respond, ask any clarifying questions needed (like "Which device are you using?"), and securely log the information.

This ensures our engineers have 100% of the context they need the moment they look at your ticket, drastically reducing back-and-forth emails and resolution times.

You'll always have the option to connect with a human engineer, but we're confident this new workflow will make getting help much smoother.

Best regards, [Your Company Name]

Voice AI announcement (existing customer)

Subject: Call us anytime — our AI is now available 24/7

Hi [Customer Name],

We are excited to announce a major upgrade to our phone support. During after-hours, or busy times when all our engineers are assisting others, you won't be sent to a standard voicemail box. Instead, you'll be greeted by our intelligent Voice Assistant.

Our Voice Assistant can take your detailed message, transcribe it, and instantly generate a high-priority ticket in our system. This guarantees that your emergency is logged immediately and our on-call team is notified without delay.

We're committed to being there when you need us most.

Best regards, [Your Company Name]

Full suite upgrade

Subject: Your IT support just got a major upgrade — here's what's new

Hi [Customer Name],

We are overhauling how we deliver support to make it faster, more responsive, and more personalized than ever before.

Over the next few weeks, you will notice a few exciting changes designed to respect your time and keep your business running smoothly:

  • Instant Chat Support: You can now chat directly with our team via our new desktop application.
  • Smart Triage: Our automated assistant will handle the initial information gathering, ensuring our engineers get straight to solving the problem.
  • 24/7 Voice AI: After-hours calls will be intelligently processed and transcribed immediately into our system.
  • Proactive Follow-ups: If a ticket is waiting on your response, our system will gently remind you so nothing falls through the cracks.

We'll be in touch with a Quick Start Guide soon!

Best regards, [Your Company Name]

Adoption nudge (existing customer)

Subject: Did you know you can get help faster with Chat?

Hi [Customer Name],

We noticed your team recently submitted a support request via email. While we always process those promptly, did you know you can get help even faster using our new Support Chat?

Why Chat is faster:

  • You receive an instant acknowledgment — no waiting to see if your email was received.
  • Our intelligent assistant gathers the details upfront, so the engineer can start working immediately.
  • You can track your ticket status in real time, right from the chat window.

Next time you need help, try clicking the [Bot Name/Icon] on your desktop. It only takes 30 seconds to submit a request.

Best, [Your Company Name]

CSAT follow-up

Subject: How did we do? [1-minute survey]

Hi [Customer Name],

Thank you for giving our new support experience a try! We'd love to hear how it went for you.

Could you take 60 seconds to let us know how we did? Your feedback directly helps us improve the service we deliver to your team.

[Insert CSAT survey link button here]

As always, if there's anything we can improve — or if you have questions about the new system — don't hesitate to reach out.

Sincerely, [Your Company Name]

Customer Quick Start Guide

Distribute this to end users alongside the announcement email. Customize with your branding.

Step 1. Locate the [Bot Name/Icon] icon in your system tray (Windows) or menu bar (Mac), or click the support portal link on your desktop.

Step 2. Click "Start a Conversation" and briefly describe your issue or request in plain language.

Step 3. The smart assistant will reply instantly and ask 1–2 short clarifying questions to better understand your issue.

Step 4. The assistant hands off to a real human engineer with all your details already prepared — no repeating yourself.

Step 5. Re-open the chat window any time to see the current status of your request and any updates from the team.

A few tips for end users:

  • Type "urgent" or "speak to a human" at any time to be routed to an engineer immediately.
  • Chat is available 24/7 — the AI never sleeps.
  • Check the chat window for ticket status updates — no need to call in to check.

Customer FAQ

Include this in customer onboarding communications and the customer-facing knowledge base.

Will I be talking to a robot or a real person? When you first reach out, you'll interact with our AI assistant, which is designed to gather your details quickly so our engineers can resolve your issue faster. A real human engineer is always involved in the resolution — the AI simply does the prep work so they can help you without delay.

Is my data secure? Yes. All communications are securely transmitted and logged directly into our protected ticketing system. Thread is SOC 2 compliant and enterprise-grade secure. Your data is never shared with third parties.

What if the AI can't solve my issue? The AI's job is triage and information gathering. For anything complex, it will immediately hand off to a human engineer with the full context it gathered — so you never have to repeat yourself. You can also type "speak to an engineer" at any time to bypass the assistant entirely.

Can I still call or email? Yes. Chat is simply a faster option. All traditional support channels — phone and email — remain fully active.

Will I be notified about my ticket? Yes. You'll receive updates in the chat window and via email as your ticket progresses. If we need more information from you, the system will send a friendly follow-up reminder rather than leaving you wondering about the status.

How do I request a human engineer right away? Type "speak to an engineer" or "urgent" in the chat at any time, and you'll be immediately connected to a technician. If it's after hours, the request will be flagged as priority and handled first thing in the morning.

Reference

Customer adoption best practices

Lead with benefits, not technology. "Faster support" and "24/7 availability" resonate with customers. "We deployed AI" creates skepticism. Frame the change around the outcome, not the tool.

Name the bot something human-friendly. Avoid generic labels like "AI Bot" or "Chatbot." A friendly name like "Aria from [Your Company]" or "Max" builds familiarity and reduces friction.

Call the champion contact before launching. A proactive phone call before the announcement email sets the stage. Ask: "Is there anything specific about your team we should know before we launch?"

Don't launch without testing internally first. Techs must be comfortable with the new workflow. If they're confused, customers will suffer. Run at least one full internal test cycle before enabling customer-facing features.

Start with tech-savvy, low-complexity clients. Build team confidence and iron out issues before rolling out to mission-critical or change-resistant accounts.

Set clear escalation paths in every communication. Customers trust AI more when they know humans are always available. Every onboarding message should include explicit "how to reach a human" instructions.

Share quick wins publicly in the first QBR. Pull the metrics early — even 2 weeks post-launch. Show reduced response times, captured after-hours calls, or positive CSAT scores.

Use CSAT to capture positive momentum early. Enable CSAT via Recap Templates on day one. Satisfied customers become advocates and help drive upsell conversations.

Monitor sentiment analysis proactively. Sentiment flags frustrated or unhappy customers in real time. Check the Inbox daily for negative sentiment threads and ensure a human tech takes over before the customer escalates through another channel.

Iterate monthly with Ask Magic. Use Ask Magic each month to identify top recurring issues, CSAT trends, and triage accuracy rates. The more threads close, the smarter Client Intelligence becomes.

8-week onboarding timeline

Phase

Internal

Customer-facing

Week 1–2 — Foundation

PSA integration, workspace setup, Magic AI enabled, brand customization

None — internal only

Week 3–4 — Internal training

Thread Inbox training, AI automation testing, Companion App setup

None — internal only

Week 5–6 — Pilot launch (Wave 1)

Monitor Wave 1 closely, daily feedback loops, adjust AI configuration

Announcement email (3–5 days before), Quick Start Guide on go-live day

Week 7–8 — Wave 2 expansion

Expand to 25–40% of customers, begin Voice AI configuration and testing

Wave 2 announcement emails, follow-up calls with pilot accounts

Week 9–10 — Voice AI launch

Activate Voice AI in Overflow mode, configure escalation keywords, update Views

Voice AI announcement emails to applicable clients

Week 11–12 — Full rollout

Remaining customers onboarded, all AI features active, Reminder Agent live

Full suite announcement, adoption nudge emails to low-engagement users

Week 13+ — Optimization

Monthly Ask Magic QBRs, Client Intelligence review, iterate on AI performance

First-QBR metrics shared with clients, CSAT summary sent

KPI dashboard

KPI

Baseline (before Thread)

Target

Time to initial response

< 1 minute (AI)

Triage accuracy (category/type)

96%+

Issues auto-resolved (no tech)

0%

10–25%

Technician dispatch time saved

30% reduction

After-hours tickets captured

0 (voicemail)

100% captured

Customer response rate

+30% improvement

CSAT score

Upward trend

Tech adoption rate (Thread Inbox)

0%

100% by Week 4

Escalation matrix

Scenario

AI action

Tech action

Manager action

Priority

Critical system down

Create P1 ticket, alert

Immediate callback

Monitor SLA

P1 Critical

Frustrated customer (negative sentiment)

Flag for human review

Take over chat

Review if escalated further

P2 High

After-hours emergency call (Voice AI)

Create ticket, attach transcript

Review next morning (or on-call)

P2 High

Triage Agent unresolved after 2 rounds

Escalate to human queue

Take over with full AI context

P3 Medium

No customer response (stalled ticket)

Reminder Agent sends up to 5 reminders

Intervene after 5th attempt

Account review if persistent

P4 Low

For account-specific questions, contact your Thread Customer Success Manager.


How did we do?