About
Most Customer Success teams have the same problem. Too many customers to onboard, not…
Articles by Kae
-
Driving Process Excellence - 4 Key Insights from Nintex's Sydney Executive Lunch
Driving Process Excellence - 4 Key Insights from Nintex's Sydney Executive Lunch
A 5 bedroom house with a pool can take many shapes and sizes! That was one of the great analogies that our keynote…
2 Comments -
3 lessons in driving process transformation from FST's South Australia Government Conference
3 lessons in driving process transformation from FST's South Australia Government Conference
First of all - I wanted to say a massive thank you to FST Media for running the 6th annual FST South Australia…
1 Comment -
A Hypocrite’s First Post
A Hypocrite’s First Post
At LinkedIn, we've just launched our Agency Influencer program, which encourages our agency partners to write long form…
13 Comments
Activity
5K followers
-
Kae Hum shared thisI was just catching up on a Customer Success webinar recording. I found the below facinating. Customers rank technical implementation as their #1 priority in a new relationship. The post-sales teams responsible for delivering it rank it #6. That single gap — straight from Bain — explains more churn than any pricing page or missing feature ever will. Because here's what's sitting on the other side of it: → 75% of SaaS users abandon a product within the first week → 90% churn probability if they don't engage in the first 3 days → But onboard them properly, and they're 5x more likely to still be around at 90 days — and 30% more likely to buy more. Read that again. The thing customers want most is the thing we quietly deprioritise. So here's my question: why do you think the people closest to implementation are the ones who rank it lowest? Is it incentives? Org design? P.S. This is the exact problem we're building Obi to solve at Cor — so the first 72 hours stop being a coin flip.
-
Kae Hum shared this"Skip that. Go to integrations." No hello. No please. That's how users talk to an AI agent. When I compare that to how users talk to our CSMs on calls, the difference is stark. With humans, there's warmth. "Hey, how's your day going?" "Thanks so much for your help." With the agent, it's pure function. Tell me what I need. Show me where to click. Move on. The data backs it up. 50% of users skip ahead to the section they need. 38% open with their own question instead of following the default flow. They're not here for a tour. They're here to get something done. People behave differently when there's no social obligation. They don't need the agent to like them. They just need it to work. The longest sessions aren't the ones where the agent is most conversational. They're the ones where the agent gets people through their task fastest and they stay to do the next one. Users don't want the agent to sound human. They want to get the value. Fast.
-
Kae Hum shared thisSome AI pilots get value in within weeks. Others take months or not at all. I've run dozens of AI onboarding pilots over the last few months. The ones that succeed do three things differently. 1. They don't try to do everything at once. One customer said it best: "This is an additional resource that covers ground our team physically can't." They started with one specific onboarding use case. Users engaged after hours and on weekends. Time a CSM was never going to fill. Once they saw the value, they expanded to other use cases. 2. They committed to given it a real shot One customer said on the first call: "I like your agent, but need to prove it works with a small set of customers first." Then he assigned a team member and mapped out a plan with committed objectives, responsibilities and a set timeline. 3. They invest in real training content. The companies that stall upload short video clips with no context. The ones that see success take the opportunity to identify the best structure (and in some cases completely re-designing what they've done in the past) and provide the agent with supporting content. AI can help you scale. Some effort is required upfront. The companies that do this see results fast. 87% of user questions get resolved without a human. A few weeks ago we published a research report and implementation blueprint based on everything we've learned from these pilots. Ping me if you want a copy.
-
Kae Hum shared thisMost onboarding and support tools treat all of your customers the same way. Obi doesn't. In the video: a healthcare client played by yours truly is logging into Cor for the first time. Obi recognises that the customer is in healthcare and surfaces the privacy configuration steps that are so important to this industry. We know as we've been working with quite a few healthcare platforms recently. It walks the admin through: 1) How to configure masking rules for sensitive information 2) How to delete session recordings and transcripts This is what context-aware onboarding looks like. Let me know if you have any industry or regional requirements that we should consider. Keen to hear!
-
Kae Hum shared thisIn Australia, we've just had the luxury of an extra long weekend away from work. I was having a conversation over a family dinner with my brother. He's a lawyer. He told me he's never once received training on any of the legal software for his entire career! It's been on him to figure it out himself - always. Here's the thing - he's not alone. According to Recurly, over 20% of voluntary SaaS churn is linked to poor onboarding. When a new user onboards with Cor, the average session with Obi lasts 31 minutes. 31 minutes of questions answered, gaps filled, and value understood. That's 31 minutes that used to be a support ticket, a churned account, or nothing at all. For self-serve customers without Obi - that's 31 minutes of silence instead. My brother is a highly educated professional (Probably the nicest thing I've ever said about him). If he's left to figure it out himself, imagine what's happening with your self-serve customers in other sectors. How long does your average self-serve user spend figuring out your product on their own?
-
Kae Hum shared thisIf you have self-serve customers, many of them have already failed before they've even started. Users sign up, get a welcome email, and then silence. Here's what that silence actually costs you. When a new user onboards using our agent (Obi), they ask an average of 11.7 questions. 11.7 questions answered = a user who understands your product and stays. 11.7 questions unanswered = a user who churns quietly and never tells you why. I had a call today with a Head of Implementation at a company currently trialling Cor. After just 3 days of testing, she decided to roll-out our agent to all her customers. Her reasoning? "What could go wrong? They don't get any training now anyway. If Obi doesn't work - we'll turn it off." The fix isn't hiring more CS reps. It's making sure every self-serve user gets answers at the moment they need them. That's exactly what Obi does. The product you ship is rarely the problem. It's the silence after signup that kills retention. How many of those 11.7 questions about your product are being answered right now?
-
Kae Hum shared thisGreat work from Arwa Lokhandwala and the rest of our engineering team who are evolving Obi - our customer onboarding agent at pace. This time, they've given our agent persistent screen awareness, which allows us to provide pro-active guidance resulting in 3.6X longer sessions. Come have a chat if you're thinking of scaling your onboarding/training efforts.Kae Hum shared thisVoice UI gets the "wow" in every demo we run. Turns out, it's not what makes the biggest difference in live AI onboarding sessions. Screen awareness does. When users share their screen with Obi, sessions run 3.6x longer. They ask 3x more questions. They complete more. They come back more. Not because the voice got better. Because the agent could get live context and just be on the same page. Until now, Obi waited for a signal. A question, a hesitation, a mention of something going wrong, before it looked at the screen. It was reactive by design. Arwa changed that this week. 👏 Obi now reads the screen in real time. It tracks whether the user is completing steps correctly and moves forward automatically when they are. If it detects an error or sees the user drift, it checks in. The voice is what people remember from the demo. The 3.6x is what shows up in the sessions. --- 👉 Give it a try yourself at getcor (dot) ai
-
Kae Hum shared thisThank you Customer Success Collective for running the "Using AI to Scale Support without sacrificing quality" webinar. Great insights from Rob Edmondson and Seth Halpern As a go deeper in the customer success space, one thing I realize is that there's a huge amount of care and thought goes into any initiative that impacts their customers and their own internal teams. https://lnkd.in/eeg_dnvT #customersuccess #customersupport #AIUsing AI to scale support without sacrificing qualityUsing AI to scale support without sacrificing quality
-
Kae Hum shared thisGreat insights from Mark Deacon and Marcel Peifer on the HubSpot AI Agents in the Real World Webinar. From what I've seen in market, one of their first thoughts are spot on. You need good content to start off with and then treat your AI as you would an employee with proper education and support and your "AI employee" will pay itself off in multiples (paraphrased). Also appreciate the call out Mark, using Cor's AI platform to help you scale your customer onboarding and training. https://lnkd.in/e5hTq5DBFrom Experiment to Impact: AI Agents in the Real World | WebinarFrom Experiment to Impact: AI Agents in the Real World | Webinar
Kae Hum replied to a comment
15h
Chisom Emmanuel Thanks for your input and feedback Chisom. Agree that incentives don't typically seem to bind people together to achieve the same objective for the vendor let alone a unified focus on the customer.
-
Kae Hum liked thisKae Hum liked thisAI isn’t just for small tasks anymore. Real marketers are saving hours each week with AI tools, right now. Discover how you can transform your workflow with Active Intelligence.
-
Kae Hum liked thisKae Hum liked thisOkay... I was QUITE impressed with monday.com I had a call with Einat (senior product designer) last week, giving feedback on their new AI + agents. I see tons of potential there. Had a major gripe though. The stupid sidekick button was in the way of clicking submit. Like really annoying. And I told her as much. Week later, she sends this email. And I notice in the platform it's moved. I felt heard. It's fixed. As a customer, that makes me quite happy. I've been on these types of calls before and I never get a follow-up email (props Einat! you are fantastic). Nor are changes made this fast. I'm impressed Monday. Nicely done.
-
Kae Hum liked thisKae Hum liked thisHad a great time up in Brisbane kicking off our Reddit IRL sessions with the team, Sydney you're up next! Reddit, Inc. Reddit for Business Hayley Saddleton Cassie Gilbert Georgia Wetzell Martin Suchy Tom Tilney Josh Frangi Alexandra Campey Balraj Saini Carvey Truong Edwin Hughan Ella Keddie Emily Dooley Georgia Pigrome Goran Naumoski Gracie Mutton Heidi Wildermuth Jenny Le Grand Joanna Woods Kandiese Villella Larissa Foy Lucy MacCulloch Maia Pang Mark Ferguson Max Bernstein Molly Hanrahan Sammy Elazab Scott Mitchell Serena Wong Simon Connolly Stephanie Shang Robert Turnbull
-
Kae Hum liked thisKae Hum liked thisI love talking about RaaS, Revenue as a Service, TP’s fastest growing business segment and how we can help CX leaders and business owners grow together with real-time AI intelligence. See some of you in Scottsdale to get in on the convo! #TPAPAC #TPImmersive Daniel Trueman, Mallory Fahy, Tom Mac-Kenzie, Dr. Anubha Upadhyay, Dave Rizzo, Miranda Collard, Ryan Wray
-
Kae Hum liked thisWith AML obligations for all real estate agencies commencing from 1 July, agencies need to be making decisions now about their AML provider and how those workflows will fit into their business. The goal is to help agencies manage AML in a way that feels practical, connected and built into the systems they already use every day. I'm excited to be working with APLYiD on making this compliance burden as simple as possible. If you're not already up to speed on your AML obligations from July 1, join our webinar (link in comments) for a deeper dive into how it will work and what agencies should be thinking about as they prepare. Russell Smith Toby Taylor Tom Clark Murphy Scott Shepherd Kate SunolKae Hum liked thisWe’re excited to introduce APLYiD as an integration partner of PropertyMe ⭐ APLYiD is a no-nonsense Anti-Money Laundering platform built for real estate agents, helping agencies onboard, verify and reverify vendors and buyers, manage risk and stay on top of ongoing obligations. With APLYiD integrating directly with PropertyMe, agencies can support their AML due diligence without adding unnecessary complexity to the way they already work ✨ 🔗 Find out more information here: https://lnkd.in/gDE-jQ4n
-
Kae Hum liked thisKae Hum liked thisAgent feedback is clear: Luxe delivers results. Senior Sales Manager Paul Loftus outlines how agents are achieving increased visibility, stronger enquiry volumes, and greater momentum by using Luxe as part of their campaign strategy. Learn how Luxe can help you achieve extraordinary results > https://bit.ly/47SiEDN
-
Kae Hum liked thisKae Hum liked thisHad a full room last week, all trying to work out why the most recent wave of AI tools is still leaving advisers feeling bottlenecked. Turns out we mostly agreed the AI isn't the constraint. The constraint is the work of reshaping a firm's work and data so advisers actually get value out of it. At least in this wave, you can get to that end state without locking yourself into a single vendor. Great to chat Maria Lykouras Rob Talevski Hugh Robertson CFP® CIMA® ifa (Independent Financial Adviser)
Experience & Education
-
Cor
View Kae’s full experience
See their title, tenure and more.
or
Languages
-
English
Native or bilingual proficiency
-
Chinese
Elementary proficiency
Recommendations received
7 people have recommended Kae
Join now to viewOther similar profiles
Explore more posts
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore More