Are You New Here?

Streamlining onboarding for busy teachers (they’re all busy).

Company
Makers Empire

Time
July 2020 – Ongoing

My Role
Research, process design, prototyping and testing, final UI handover

Context

Makers Empire is the world’s most fun and easy to use 3D design program. The Makers Empire 3D app is used by elementary/primary-aged students to create over 100,000 designs a day, while the Makers Empire Teacher Dashboard is used by 38,000+ teachers around the world.

Makers Empire 3D is a native app which works on all platforms.

The Teacher Dashboard is web-based.

Problem

Teachers were confused about how to use Makers Empire.

There was a very high drop off rate through our “Getting Started” guide for new users. Too many trial users never reach key milestones and conversion rates were affected.

Additional considerations:

  • EdTech products generally have a very long sales cycle so it is challenging to measure the efficacy of any changes you make in the short term.

Objectives

  • Increase the number of teachers who got as far as setting up a class and using the Makers Empire 3D app with students during their trial.
  • Improve conversion rate.

Crunching the Numbers

* To respect commercial confidence I have omitted exact figures.
The working assumption at Makers Empire had always been that if teachers can see their students using our product they are more likely to reach their ‘aha moment’ and be more likely to convert from a trial to subscription. To find out why this wasn’t happening as much as we’d like, we started by digging into the data.

We used Google Analytics to track the teacher on-boarding funnel and found that of the teachers  who did start reading the Getting Started guide, only 20% got as far as reading the instructions for adding students to their account (step 4 of 13).

Clicking on a page to read instructions and following the instructions to actually do something are two entirely different thingsWhile 20% of trial users got to the instructions on how to add students, even less actually did it.

However, a small group of ‘super trial-users’ were adding students and creating thousands of designs during their trial period, suggesting it is possible to add students and engage thoroughly with our product, but something was blocking the average trial user, maybe motivation, tech issues or time?

We tracked the number of student accounts created on each school’s trial account to get a more accurate understanding of how many teachers were having a ‘hands-on’ trial experience.

We didn’t have any way of knowing how many students are in each teacher’s class, so we checked average class size figures from the USA and OECD countries. Average class sizes hover in the low to mid 20’s, so we reasoned:

  • less than 10 accounts did not represent an entire class of students,
  • 10-29 accounts did suggest that a teacher was preparing to trial our product with an entire class (or at least a decent cohort),
  • 30 or more student accounts suggested a teacher has set up more than one class on their school account.
  • 1 account means the teacher has created an account, but not added any colleagues or students
 

We then wanted to know how many of those students and teachers ever used Makers Empire 3D.

As a proxy measurement for student engagement, we tracked the number of designs created on each school’s account and divided that figure by the number of student accounts. New users were guided to a series of 11-15 tutorials the they first log in to the Makers Empire 3D app, so we reasoned:

  • if the average student was creating less than 10 designs they probably haven’t completed the tutorials,
  • if they create 10-20 designs they’ve had time to complete the tutorials and started experimenting with creating their own designs and/or engaging in other areas of the app, 
  • if they created 30 or more designs they were probably starting to achieve a decent proficiency.
 

 These assumptions were by no means rock solid evidence, but given our limited resources we decided we couldn’t spare the HR to compile exact figures for every student on every trial account.

When we cross referenced the number of designs created per student by the class size cohorts. The results were pretty grim.

We also reviewed some academic papers and some of the many thousands of articles written about on-boarding best practices. We found J Carrol’s Minimalist Instruction theory especially relevant.

"They scan a page looking for hints... and click anything that looks like it will result in less cognitive load than reading"

Early Analysis

  • Most elementary/primary school classroom teachers simply do not have the time or patience to learn how to use complicated programs.
  • Teachers, like most internet users, really don’t like to read. They do not read pages top to bottom and generally do not follow instructions sequentially. They scan a page looking for hints about how to achieve their objectives as quickly as possible, and click anything that looks like it will result in less cognitive load than reading.
  • There were some fundamental misunderstandings about our product which immediately confused new users, the main one being that Makers Empire is actually two products: the 3D design app and the web-based dashboard, and teachers need to familiarize themselves with both.
  • Low levels of technical proficiency and digital literacy among teachers were a more significant barrier than we expected in 2020.

Going Deeper

Once we had quantitative data to tell us with some certainty what was happening, we moved on to observation sessions (in-person and using Smartlook), surveys and interviews to help us better understand why teachers were struggling to get through the Getting Started funnel.

Some of the most brutal but valuable feedback came from an informal after work drinks event. Gin for the win.

We realised that our definition of a successful trial didn’t always align with the trial user’s goals (intention), we were overloading users with too much info (clutter) in terms that didn’t always make sense to them (language).

Intention

Not every trial user wants the same thing from their trial. They come with different intentions, levels of interest and/or commitment, and they define a ‘successful trial’ differently. There is no one experience that will suit all intentions. 

We grouped trial users into four categories according to their intention and purchasing power:

  1. Tourists
    They are at an info gathering stage. They’re clicking around to get a sense of what Makers Empire is about. They want to see what it does, but don’t want to do it themselves yet.
  2. Toe Dippers
    They want to test the waters. They will try the product themselves but under controlled conditions and for a short time, typically 20-30 minutes from the safety of their lounge room, after hours.
  3. Divers
    They’re ready to jump in and make a splash. They want to see how our product performs in real-world conditions. They have enough confidence in our product to invest at least one lesson with their class, and up to a whole term.
  4. Boss
    Teachers rarely have much buying power, where as a principal or district administrator can go straight from info gathering to purchase without going through a trial at all.

"Sorry, but there is no way I'm spending that long or reading that much"

5th grade teacher

Clutter

We had previously tried to reduce the amount of information trial users were bombarded with, but it was clear we still had a lot of work still to do to prevent cognitive overload.

There were still too many emails and they were too long.

When users did get to the Teacher Dashboard they were overwhelmed by the interface. The functionality was fine, but usability needed attention, i.e. if I showed a trial user how to use a feature, they could – but they struggled to figure it out by themselves.

Teacher Language

Like any field, education has its own vernacular and jargon. Put simply, teachers speak teacher language, not tech language. If we want to engage teachers, we need to speak to them in their language.

Reframing the problem

We distilled all that number crunching, observation and interviewing into a few big opportunities:

1.

How might we better assess a trial user’s intention and tailor the onboarding experience to suit their intention?

2.

How might we improve our product so it requires less instruction?

3.

And for bonus points; can we make a dashboard that teachers actually love?

What We’re Doing…

It’s a never ending process, but you’ve got to start somewhere.

Triage

We modified our trial sign up page to help us identify which category a new teacher best fits, based on their likely intention. The key question is:

How would you like to get started on your trial?

  • Try Makers Empire myself before using it with my students
  • Start using Makers Empire with my students ASAP
  • I’d like to get a bit more info before I start using Makers Empire

Intent-driven journey

We then deliver a different series of email prompts to suit the trial user’s intention, with the goal of nurturing teacher up from one level to the next: Tourist > Toe Dipper > Diver.

Declutter & Translate

I’m a minimalist at heart. Bidding farewell to superfluous UI elements and under-performing features comes easily to me. Emotionally, it’s not so easy for the techs who built them.

We also have thousands of teachers who already use our Teacher Dashboard, cluttered as it may be. We can’t simply rip functionality out or suddenly change the UI without carefully considering how these changes will affect our current users.

Rather than doing a full Mari Kondo declutter, we redesigned the interface to be more spacious and, through some co-design sessions with teachers, reorganized the navigation into more intuitive groupings.

Show, don’t tell

We’re experimenting with removing our Getting Started guide entirely and replacing it with a very simple, friendlier checklist that links directly to the relevant page and offers in-situ tips on how to complete the relevant task. 

Results

Many of the changes above have been implemented very recently or are still being implemented (at the time of writing) so we don’t have enough data to draw any conclusions yet…