Impact

50% of users request a tutor right after onboarding

27% improvement in drop-off rates

158% increase in requests for tutors YOY

Context

Knack is a free platform for students to get tutoring from peers who have walked in their shoes

Knack offers free, in-person/online peer tutoring at universities (using a variable annual fee based on student access). Universities pay Knack, who then compensates tutors.

In short, students get free and accessible tutoring while student tutors have a fulfilling side hustle - it’s a win win!

Problem

43% of new users on Knack drop off after sign-up without attempting to find a tutor

In Q4 of 2023, our Onboarding/Activation team was developing the product roadmap. I analyzed current onboarding data, revealing a 43% drop-off rate after signup. This prompted us to re-evaluate the existing onboarding flow because the high drop-off rate was jeopardizing our north star metric: number of students served.

The existing onboarding flow was designed to make sign-up as quick as possible.

But, what’s wrong with minimal onboarding?

Unconvinced by positive generic user feedback of our current onboarding, I drove discussions to dig deeper on why our current onboarding, which was made to be fast and easy, was not retaining users.

⚠️ Students are giving up before even trying

As a result, they’re losing out on accessible and personalized support that could help them reach their academic goals.

Goals

Provide context to normalize tutoring by highlighting its benefits and affordability while setting expectations for connecting with a tutor.

Design goals

Business goals

⭐️ Solution

A personalized onboarding experience that addresses students’ hesitations, replacing doubt with empowerment

Before

After

Instead of a one-size-fits-all onboarding, we crafted a flexible, personalized journey that anticipates and addresses students' unique concerns, nurturing engagement at every step.

Goal #1: Provide context to de-stigmatize tutoring and reinforce affordability

Result: A dynamic model that targets students' concerns and highlights Knack's value propositions.

Goal #2 and #3: Equip students with the knowledge and confidence to seek out tutoring early on.

Result: Set expectations on next steps and drive a sense of urgency to motivate students.

⬇️ Want to know more about the process? ⬇️

User research

Why were students not making a request for a peer tutor?

After sign-up, many users were leaving the platform before requesting a peer tutor. We ran user interviews and surveys to understand the top 3 reasons behind this behavior:

Now, the fog was clearing.

Our current flow failed to capture a large demographic of students:

those who are curious about Knack yet have no urgent need for tutoring.
And …it didn’t help that the onboarding flow provided zero context on what Knack is and how to use it. 🙄

Seeking tutoring is a lot like seeking therapy

After seeing the hesitations around tutoring, I was strongly reminded of my own hesitations when first exploring therapy.

So, how can we change this narrative?

Ideation

Our initial approach was to make it feel like a personalized survey

Because the research pointed to a lack of context in how to use Knack and students weren't prioritizing tutoring, we sought to provide personalized tips on how to get started with Knack, based on their needs and study habits.

Due to the flexibility of contract options we offer universities, there needed to be an option for every combination of features available.

Validation

Redefining our personas to validate the efficacy of our recommendations

During the validation process, users were asked questions to assess the following aspects:

1. Ease of onboarding flow

2. Expectations on what to do next after

3. Relevancy of the recommendation they landed on

Results

• ✅ Students found the flow easy to use

• ❌ Students still didn't know what Knack was and how to use it

• ❌ Students did not find the recommendations relevant

Oh shoot, we made the wrong assumptions…

The results of the usability tests led me to believe that our assumptions for what students want was wrong altogether. I conducted another survey to test the validity of our user personas and analyzed the data using pivot tables.

The culprit: outdated information!

Prior to this survey, we were operating with the company’s previous (and outdated) user personas. Before, we were bucketing student types by study habits and the grades they got:

We learned that studying isn’t one-size-fits all.
Instead, we should meet them where they’re at.

Rather than approaching it with “amount of time spent studying” and “grades”, it was more realistic to view students in these buckets:

New personas, new goals

• Approach social concerns with empathy, providing them transparency of how normal it is to ask for help.

• Communicate value propositions of free tutoring and affordability.

• Provide relevant recommendations based on users' current needs and beliefs.

Plan of action

Prioritize communicating core value propositions. Then, find a path to personalize.

Our immediate plan is to provide value to users by updating the website copy with positive and encouraging language and developing a library of finance recommendations for users to access.

Step 1. Financial Accessibility

Emphasize that receiving peer tutoring from Knack is free, and includes peer tutoring as well as on-demand tutoring from professional tutors.

2. Social Proof

Introduce live-updating numbers of others who have used Knack at their school. Showing how many other students have used Knack helps normalize the usage of tutoring.

3. Personalized onboarding

Allow users to be shown messages that targets their specific doubts, challenging their limiting beliefs about tutoring.

Vision

Step out of the traditional Prussian model of education. Peer to peer, personalized education is the future.

Just as therapy shed its stigma and became integral to mental health, we envision a similar transformation for education. One World Schoolhouse (written by the founder of Khan Academy) paints a powerful picture of this future: a world where accessible, free education empowers every student.

This is the future Knack aspires to be a part of. We want to be the go-to platform where students find diverse academic support, not just through professional tutors, but also peer tutoring, study groups, and more – a curated toolbox for their unique learning journeys.

But this vision rests on dismantling the outdated narratives around tutoring. Let's shed the labels of "expensive," "ineffective," and "uncool." Asking for help isn't a sign of weakness; it's a powerful act of self-advocacy and a catalyst for growth.

Knack’s vision is to redefine the narrative. We are building a world where every student feels empowered to seek support, embrace collaboration, and unlock their full potential.

Takeaways

I learned to validate earlier on in the process to save time

Don’t let a business problem lead the design solution

I learned that a business problem is simply a symptom of a product failing to meet our user’s needs. This project took longer than it should because we were starting from a business problem without anchoring the design goals to a strong human problem. A lot of rework could have been saved if we invested more research time early on in the process, rather than conducting it mid-way.

Follow your intuition

Despite users reporting the current onboarding flow was great, fast, and easy, I knew there was more to the story. By following my intuition, I realized that the issue needs to be looked at from a different angle.

Be careful of user biases

It was tempting to listen to users’ positive regard for the previous onboarding flow but I learned to keep in mind that their opinions may be biased, as they did not fall within our target demographic of research.