Guided Mentorship

• Role: Lead designer providing hands-on work and project oversight alongside my report, Renga
• Core problem: organizations had a hard time getting users to engage on the platform, making it hard to showcase the value of our product, therefore making it difficult to justify the investment
• Research insight: Mentees and mentors were unsure how to utilize the platform, and how to approach the mentorship experience in general. Additionally, many participants were meeting, just not on our platform
• Focus areas: Improved connection onboarding flow, guided conversation templates with AI personalization, and a completely reimagined video call experience
• Outcome: Post-match retention rose from 65% to 81%, along with ~8x more meetings scheduled and action items created, and ~2.5x more goals set

Chronus exists to help organizations run mentoring programs. Our software matches mentees with mentors, and provides features that support those connections. Our leadership team tasked us with exploring how to improve churn, so the product team worked closely with sales and customer success to understand why customers had left, or were at risk of leaving. By far, the theme we heard most was that organizations had a hard time getting users to engage on the platform, and therefore couldn’t showcase the value of our product.

Users were signing up and getting matched, but…

Within 30 days of being matched, 35% of connections had NOT scheduled their first session.

While the problem was clear, we didn’t have a good sense of why it was happening, so we drilled a bit deeper. After speaking with mentees, mentors, program admins, and our customer support team, two key themes emerged:

First: both mentees and mentors were unsure about how to utilize the platform, or how to approach the mentorship experience in general. They’d get confused, overwhelmed, or nothing would happen, and they’d just abandon the whole thing. They had questions like:

“What am I actually supposed to do here? What do we talk about? Who’s driving? How often should we meet?”

Second: a fair number of our connections were meeting off-platform — either in person, or hopping on whatever video tool their company already had. On the surface, that’s not a failure. NPS stayed consistently high, and by every signal we had, those relationships were going well. But it did mean people were getting matched and then never touching the rest of what we’d built. This was a critical business problem:

If the mentorship happens somewhere we can’t see, the only demonstrable value we are providing is the matching, and, “Are we paying just for matching?” is not a question you want to hear from customers.

Program admins felt the fallout of both. They had a hard time visualizing the value our platform was bringing, and with engagement lagging on top of that, their ability to tell a compelling story about the program’s benefits was severely hindered. That ties directly back to the most common reason for churn.

“I can’t prove this program is working.”

At this point it was pretty clear — we’d done a good job of getting two people matched, but a poor job of helping them progress and get the most out of our platform.

This problem area ended up being the highest priority roadmap item for over a year, spanning multiple work streams that all had to come together to form a seamless mentorship experience. While I was still largely hands-on, a lot of my involvement was also as a mentor (how meta, I know 🤪), helping steer or make decisions when things felt stuck. Sree owned most of the production of the onboarding work.

To find key opportunities for improvement, we looked at the entire mentorship lifecycle, and explored which areas had the best chance of moving the needle on the problems discussed above. At the same time, I worked closely with a few cross-functional leaders who were helping define, from a program design perspective, what ideal mentorships look like.

Our core hypothesis was that if we solved the user problems, particularly for mentees and mentors, those improvements would bubble up to solve the admins’ problems, and in turn the business problem. That left us with a question to design against:

How might we help mentors and mentees start with confidence and keep coming back, so the value of their mentorship is visible to the people investing in it?

We organized the work around three pillars:

1. Provide clarity in key moments: Communicate value, set expectations, and instill confidence, starting with product and connection onboarding
2. Set them up for success: Give people a clear starting point instead of a blank canvas, and make them want to have the conversation in our platform
3. Make value visible: Improve engagement in general, but also give admins more concrete proof that the program is working

The solution(s) tackled different aspects of the pillars mentioned above. Here’s a glimpse into what we delivered.

Connection onboarding

A quick intro flow that introduces the mentor, confirms what you’re working towards, and nudges you to set up the first conversation. Sree owned most of the production of this work.

A focused widget that gives a high-level outline of the connection journey (how many meetings should I expect, etc.) and provides key details about each proposed meeting. Users can always add or skip recommended meetings, and create their own agendas.

Agenda templates & updated meeting experience

Our biggest bet to encourage on-platform meetings was a completely reimagined video-call experience. It brought the meeting guide directly inside the meeting surface, providing specific agenda items so participants can have efficient and meaningful conversations.

AI-powered personalization

Based on your goals, added tasks, and recent meeting notes, AI can personalize the next conversation agenda to keep you on track. And if the recommended guide isn’t quite doing it for you, you can use AI to generate a new one.

Visible value

Getting more meetings onto the platform solved half of the admin’s problem. The other half was giving them something to show for it. We provided two touchpoints that allowed users to provide feedback about their meetings. One was a simple rating of how helpful they found each meeting. The other, known as "in-a-word", had users identify one word that described how they felt before and after their meeting. Together they gave admins a quick read on the health of their program, and a much better story to tell about it.

We started with 35% of connections not scheduling a first session within 30 days of being matched. After launch, that dropped to 19%. Put another way, early retention rose from 65% to 81%, nearly halving the drop-off.

The rest of the numbers say the same thing: people came back, and committed. We saw about an 8x increase in meetings being scheduled in our product, and a similar increase in the number of action items created alongside those meetings. We also saw a 2.5x bump in users who created goals, further showcasing a willingness for people to invest in the experience.

This success certainly didn’t come without challenges, and looking back, there are definitely things I’d do differently. Some quick thoughts:

• I would’ve pushed harder on the note-taking experience within the guided conversations. Because of the structured nature of the agendas, we ended up having individual text fields for each agenda item’s respective notes. I had really hoped for an open canvas kind of approach, like a Notion or Google Doc, but the engineering team wasn’t open to doing that at the time, so I let it go
• I would’ve also encouraged more exploration into integrations with existing video call tools like Google Meet or Zoom. Both had began offering some 3rd party app or plugin support, and I thought it would be awesome if we could bring the guided conversation experience directly in those products. At first that feels counterproductive to the goal of keeping people in platform, but in my mind, the specific video service being used wasn’t all that important so long as they can utilize the conversation guides and then follow up based on the created action items. I think we could’ve easily nudged people back to the product to manage those elements of the experience.