Why Mentorships Fail and How We Fixed Them

Why Mentorships Fail and How We Fixed Them

What do mentorships suck?

There are two answers:

  1. They’re either too difficult to set up, or
  2. They don’t result in skill advancement for the pupil.

Finding a mentor willing to devote time (and possibly resources) is rare.

Finding a pupil willing to devote all energy to following a strict regimen in order to develop his/her skills is also rare.

Finding both is a virtual nightmare.

So how might an app fix this problem?

No alt text provided for this image

Goal Setting for Skill Advancement

The main problem we decided to tackle with Sprout Mentorships was the fact that mentorships hardly ever result in actual skill advancement.

In order to combat this problem, we created a Goal Achievement / Mentorship hybrid feature.

Users, or “Pupils” as we call them, are able to set Daily Goals and Milestone Goals.

Milestone Goals are long-term goals (i.e. Get UX Job) that may be suggested by mentors and Daily Goals are subsets of milestone goals (i.e. Apply to 3 UX Jobs Per Day), set in order to reach their milestones.

For example, if my suggested a Milestone for me to “Get UX Internship at Apple,” my mentor would then consult me on some great Daily Goals that would help me achieve that.

No alt text provided for this image

Dark Mode

When deciding whether to include Dark Mode in our app design, I really only faced one question: Why dark mode?

I wanted to find a reason beyond the simple “because it looks nice.” I found the answer while designing the transition from light to dark mode.

Mainly, dark mode makes buttons and CTAs pop.

No alt text provided for this image

In fact, all elements pop more in dark mode.

According to Apple’s Human Interface Guidelines (HIG), “Dark Mode puts the focus on the content areas of your interface, allowing that content to stand out while the surrounding UI recedes into the background.”

In the end, the decision to include Dark Mode was easy.

In fact, I was tempted to make Dark Mode the default style, but decided to err on the safer, simpler side.

Besides, keeping Dark Mode as a hidden/exciting discoverable gem gives the app an additional charm in my mind.

The Messaging Hickup

Because of time constraints and primarily a decision made by our iOS developers, it was determined that we would not be able to include a messaging feature in our app.

This was disappointing, and presented an extremely frustrating UX challenge. Without in-app messaging, how do we connect mentorships with skill achievement/goal setting?

In the end, we decided to leave all messaging icons.

Rather than delete messaging altogether, I decided we could simply link users out to a third-party messaging app.

No alt text provided for this image

This would at least salvage the “feeling of communication” that our app would have had with a messaging feature.

I admit that I hate this solution; it almost feels like lying to the user.

(Update: We later fought for an in-app messaging feature and our dev team worked hard to be able to include it).

Changing the Goals Section

Our iOS devs suggested a change to the goals feature functionality. They did not see the need for daily goals, so they wanted us to remove them completely, leaving only milestone goals.

We disagreed with this idea for two reasons:

  1. It almost completely removes the need for pupils to get on the app daily.
  2. Milestone goals, while important, hardly ever result in skill advancement on their own.

It is the consistent achievement of daily goals that result in new skills.

Here was the compromise that we arrived at:

Instead of pupils creating their own goals and checking them off manually, the mentors would set milestone goals in behalf of the pupils.

The pupils wound then need to set 4 daily goals in order to achieve their milestone goals.

This would increase and force interaction between pupils and mentors.

However, there are a few problems with this design tactic.

It complicates the onboarding process and learnability of the app. Before, pupils could set any goal they wanted to and whenever they wanted.

With this change, pupils would have to match with a mentor before being able to set any goals for themselves.

In other words, the app could no longer act as a stand-alone goal setting app (admittedly, this isn’t a huge issue, but I liked the fact that pupils could use the app in a variety of ways, according to their own learning styles).

This design change ended up being impossible to implement within our time constraints. Also, I subtly fought against it.

Although I agree the goal setting would be more helpful with the change, I also believe that usability/simplicity is a more important characteristic with any given app.

But this did lead me to a few interesting questions that I don’t exactly have answers for:

How do you whether to go with the more intuitive design choice or the most helpful functionality?

Is it worth creating an app that’s helpful but more difficult to learn?

The Team

My team consisted of myself and a fellow UX Designer, Austin West.

Austin has extensive experience in game design. His familiarity with animation and coding was invaluable to making our app function smoothly.

I’m incredibly grateful to Austin for his creative and innovative mind. Even though we approach design from different angles, his measured insight and easygoing nature made working with him a breeze.

My Learnings

  • I’m not great at the research phase. I rush to get through to the sitemap.
  • I love the “information architecture” portion of the UX Design process.
  • Working in teams does not need to be difficult. Organization, clear communication, and task management are essential for smooth teamwork.

Conclusion

I’m surprised a decent app hasn’t been created for mentorships before this point. A lot can be done with LinkedIn, but it is mostly left to the user to take initiative and set goals for himself.

This is what we set out to achieve: a mentorship app that ACTUALLY results in skills advancement.

I believe our solution resolves the mentorship problem.

To view or add a comment, sign in

More articles by Chad Zollinger

Insights from the community

Others also viewed

Explore topics