AI-assisted drafting, summaries, reply suggestions, knowledge lookups, and customer-safe support communications

Following up on my last post about AI in customer support, I wanted to share what that's actually looked like for me on the ground.

In my most recent role, AI wasn't one single tool. It was woven into different parts of the workflow: helping triage and route tickets, flagging sentiment and escalation risk early, and assisting with drafting responses so I wasn't starting from a blank page every time.

The biggest shift I've noticed? Fewer repetitive tickets landing on my desk. The routine, low-complexity stuff gets handled or surfaced faster, which means more of my time and attention goes to the issues that actually need a human brain, judgement calls, edge cases, customers who need to feel heard, not just answered.

It hasn't made the job easier in the sense of "less to do." It's made it different. More of my day is spent thinking, problem-solving, and building trust with customers, and less of it is spent on repetitive back-and-forth.

I keep coming back to the same conclusion: AI is changing what "efficient" looks like in support, but it hasn't changed why people need support in the first place. That part still needs a human.

Still learning, still adjusting, still here 😁

What's your experience been like? Has AI freed up more time for you too, or has it introduced its own kind of complexity?

#CustomerExperience #CustomerSupport #AI #ArtificialIntelligence #FutureOfWork #Tech

AI is a fantastic tool. But it should be a tool, not the entire customer service department.

We have reached a point where some businesses seem determined to remove actual people from the process. Personally, I cannot stand calling a company and being trapped in a never-ending conversation with an AI assistant.

“Please briefly describe what you need.”

I do.

“Sorry, I didn’t understand that.”

And now we are enemies.

I would much rather speak with a real person first. Let that person understand the situation and decide whether an automated tool can help. Not the other way around.

I have normal business hours. I answer my phone when I can, return calls, respond to emails, and make sure agents know there is an actual human behind the support they receive.

Do I use AI? Absolutely. It helps me organize information, research questions, review documents, and work more efficiently. But I do not rely on it to build relationships or make judgment calls. Some situations require context, experience, and the ability to recognize when the question being asked is not actually the problem that needs to be solved.

That is especially true in insurance contracting.

Contracting is not always easy, and trying to automate every step does not necessarily make it better. A knowledgeable person can notice missing information, explain carrier requirements, identify potential problems, and help an agent build their business on a stronger foundation.

AI should support good service, not replace it.

Quick and easy may sound appealing, but easy is not always efficient. Sometimes it is just cheap, unreliable, and nowhere near what you expected.

Use technology to save time. Use experienced people to make sure you are doing things correctly. The best businesses understand that you need both.

AI won't replace support teams, but support teams who use it - will.


I lead support, and I've stopped treating AI as a threat. It's the layer that clears the repetitive work off my team's plate, so they spend their energy where a human matters most.

Here's the stack I run, and where AI does the heavy lifting.

1️⃣ Intercom → where every customer conversation lands.
2️⃣ Claude → drafting replies and untangling the messy, multi-thread tickets.
3️⃣ Jason AI (Reply) → deflecting the repeat questions so my team spends its time on the tickets that need a human.
4️⃣ ChatGPT → fast rewrites when a reply needs a different tone.
5️⃣ Perplexity → quick answers with sources I can check.
6️⃣ Notion → our playbooks and team knowledge in one place.
7️⃣ Airtable → tracking tickets, trends, and how the team is doing.
8️⃣ Trello → the daily task board that keeps everyone on track.
9️⃣ Jira → bugs and escalations that need engineering.
🔟 Slack → where the team actually talks.

Four of these are AI tools, and none of them answer the hard tickets for me. They buy my team the time to handle those well. That's the shift worth making. Let AI carry the repetitive load, and put your people where empathy and judgment change the outcome.

What is your top tool in support? 🤔

We're a customer support team of around 70 agents handling a few thousand conversations every day. Like a lot of companies we've started rolling out AI over the last year. It started with ticket summaries and reply suggestions. Now we're testing AI for handling common requests before they reach an agent. Overall it's been useful so far. AI is great for repetitive questions and speeding up after-call work. The bigger challenge has been deciding where AI should step in and where a human should take over. If it hands off too late the customer is already frustrated. If it hands off too early you're not getting much value from the automation. Most of the reports I've read say AI adoption in customer service is growing fast and companies are using it to support agents more than replace them. That seems to match what we're seeing. Open to different opinions.

Copilot intelligent triage is now included in Suite and Support Professional plans and above, enabling automatic AI classification of incoming tickets by topic, sentiment, language, and entities at no extra cost. This helps identify ticket trends and provides AI-powered workflow recommendations.

Auto assist now uses similar solved tickets as a knowledge source across all channels and languages when no procedures or articles are available. This enhancement provides agents with more relevant reply suggestions based on real-time resolutions, helping teams respond faster and deliver consistent support. Agents can also view the rationale and relevant tickets used for suggestions.

With the General Availability of Service Agent and MCP tools in Microsoft 365 Copilot, service organizations continue to get the benefits of Copilot in customer service, now powered by the rich business context and grounded intelligence of both Dynamics 365 and Microsoft 365. Whether they’re new in career or experienced customer service representatives, your employees get tools from Microsoft to help deliver a better support experience for your customers. With this release, we have added rich, interactive app-in-chat experiences, ensuring that customer service representatives do not need to solely rely on text-based conversational UX. Employees can investigate issues, navigate complex processes, and complete tasks without leaving the conversation. The result is a more connected, action-oriented experience that helps service professionals move from understanding to resolution faster in Dynamics 365 Customer Service and across Microsoft 365 apps.

I opened my inbox this morning and counted.

23 emails from customers. 19 of them could have been written by the same person.

Same structure. Same tone. Same opening line that sounds almost apologetic for existing. Same three bullet points. Same closing that asks for "a quick sync."

They were not written by the same person. They were all handled by different CSMs at different companies using different AI tools.

And somehow they all arrived at the same email.

That is the problem nobody is talking about. Only 3% of CS teams are using AI in genuinely autonomous ways. Most are using it one task at a time. Which means every CSM is essentially starting from the same template, tweaking a few words, and hitting send. The output is technically competent and completely forgettable.

70% of CS organisations now use a dedicated platform. Only 28% say their automation has measurably reduced churn. The tools are not the gap. The thinking is.

71% of CS leaders say their platforms can predict churn risk but cannot explain why. So the alert fires. The score drops. And the CSM sends another email that sounds like everyone else's email to a customer who is already halfway out the door.

The CSMs I have watched actually move accounts forward do not send the email AI drafts. They read it, feel something is off, and rewrite it until it sounds like them talking to this specific person about this specific problem.

That takes judgment. And judgment is the one thing AI cannot hand you.

What is the one habit that makes your customer communication actually land? Drop it in the comments.

Meet my team of AI assistants and skills: Hunter, Clarice, the Board, and a few other tools I use every day.

I explain how I set each one up, and what's in it for you, whether you're a CSM or you sit in any revenue-generating, customer-facing role.

Still figuring this out myself, so I'd genuinely love to know how you're using AI.

https://lnkd.in/eyE4MaPs #CustomerSuccess #ArtificialIntelligence #CS

61 percent of your users will leave for a competitor after exactly one bad experince.

I was biting my tongue today during a hallway conversation so I would not scream this number out loud. A colleague asked why we even need human Customer Support anymore. "Why not just automate everything with AI and close the tickets?"

I love tech people. I really do. But sometimes they think human frustration can be resolved with a version update.

Do not get me wrong, AI is a fantastic sidekick. I use it to answer simple tickets. To summarize past chats, analyze sentiment, and pull up user history. But AI has absolutely zero judgment.

Take a classic edge case. A user buys a time-based package, but real life happens and they cannot use it. The database shows the purchase was delivered successfully. If an AI handles this, it looks at the policy, sees a successful delivery, and politely tells the user to go away.

A human looks at the exact same ticket, realizes we want this person to actually like our brand, bends the rules, and restores the package.

Trap a frustrated user with a bot that makes a mistake, give them no way to reach a real human, and trust is instantly broken. People just want to know there is an actual person behind the system.

Use AI to read the data. Use humans to read the room.

The Jeffri Profile

That's my customer support persona and connects to HelpScout.

Every morning at 6AM he scans the inbox and closes out spam or junk from our inbox. As new emails come in he writes draft, we say yes, he sends. Closes tickets once solved.

Writes a new “customer page” for every customer in Gbrain so we can track feedback and issues.

Used to be a manual task someone had to remember.

Now it just... happens.

Also helps scrub PII with customer support.