A lean approach to achieve product-market fit

Project Overview

As part of Aksorn’s Research and Development team, we chose to explore and test the viability of entering the self-study university test prep market. With over 400,000 Thai students taking the O-NET university admission exam each year, we felt there was a business opportunity to meet those student needs. Aksorn’s main market is classroom textbooks and supplemental materials, so our challenge was to identify a product-market fit in this new space, where we have limited knowledge and presence. This project case study details the process of design, production, and release of OnetGo, a MVP test prep application.

Objectives
  • How can R&D develop a digital university admission test prep tutoring product that can achieve product-market fit.
  • How can R&D use digital channels and mechanisms to optimize funnels without requiring the resources of the company’s traditional marketing and sales teams.
Constraints and challenges
  • Build and distribute the product using our 3rd party Learning Management System (LMS) distribution platform and content authoring tool. The platform’s main use case is intended for in-classroom content management and delivery between teachers and students. The challenge is how best to utilize this system for at-home, personalized student learning.
  • Create an optimal user experience without any custom development in the LMS.
  • Using internal team capacity and capabilities only, produce and release the product aligned to the studying and exam schedules.

Role and Responsibilities

UX product strategist, Project Lead

Align the product’s vision and objectives with team and stakeholders by facilitating co-creating/design thinking workshops in every phases of product production cycle. Design research plans. Conduct ethnographic and market research. Synthesize research insights into personas, user journeys and task flows.  Create concepts and testable prototypes. Provide guidance to the visual design and marketing team on final production design and landing pages.

Team: Business Manager, Marketing Strategist, Tutor Subject Matter Expert, Technical Lead, Visual Designer, Editorial Production

Lean Path to Product-Market Fit

A lean process approach to validate the mvp
Discover, Build, Measure, Learn

Achieving product-market fit in our case, meant providing a valuable product that users were willing to pay a currency for (effort, time, money, etc), in order to achieve their needs and goals.  In order to do this, we needed to deeply understand are users and their behaviors, know how to target them, create unique value, and validate that these values are worth the user’s exchange of currency.  Due to our limited knowledge in this new market, development and resource constraints, we felt the best process was to follow a lean methodology. Applying a lean process allowed us to quickly test and validate our assumptions and adjust our path to target accordingly.  

Discover

Discovering the needs, pains, and behaviors of our self-study students

The Discover phase is all about understanding our users, their environment, and what the market is doing now for them. We set about understanding grade 11 and 12 students by getting outside and talking to them in their outside studying environments: cafes, school libraries, bookstores, and tutoring centers. As we listened and documented to their stories, we extracted valuable insights in their current studying behaviors, their academic goals, what pains and fears they encountered, and what and who were their motivational drivers and influencers.

Methods

3 days of ethnographic observations, 20 field interviews,  350 online surveys

Representations

personas, affinity maps, current behavior user journeys

Discovering the existing market

The admission prep tutoring market is quiet saturated with products and services.  Students have many existing options from: personal and tailored – private 1-1 tutoring, social and semi-tailored – group tutoring and after-school sessions, to massive informal – Youtube and FB videos, online platform courses, mobile apps, and traditional – physical textbooks and mockup exams.  It was our immediate goal to quickly scope down and  identify where we could gain market share and provide unique value to our users.  We chose the traditional physical textbook as our direct competitor to our product.

Methods

market research with tools like SimilarWeb, competitor SWOT analysis

Key Discover insights

We began analyzing our findings by affinity mapping. This process exposed patterns of existing pains around the physical textbook learning and practice experiences.   We found our self-studying students shared these common pains:

Complex explanations created confusion and frustration within the answer key.

Rigid paths for skill development  forced students to learn and practice topics in a determined and linear way. The path was not personally relevant to their skills and areas of need.

Single way to learn assumed everyone learns the concept the same way with usually with one explanation

No immediate help / feedback to let students know where and why they are getting confused along the steps to learning

Not enough variations of practice concept questions to build testing confidence.

Unique Value Proposition (UVP) Assumptions

Now that we uncovered the common pains, we explored on what potential solutions would become our unique value propositions.  I ran a brainstorm session in which the team rapidly produced a range of ideas. First we dreamed big. What would be the most amazing solutions, limitless, without concerns? The next step was to think a lot more realistic. What can be done and still be unique? We democratically voted and came up with our short list of unique value proposition assumptions.

Progressive learning that breaks topics into microlearning components and guides students to concept mastery.

Multi-method explanations allow students to learn concepts through a multitude of different techniques and representations – short and concise, scaffolded, textual, and visual.

Skill placement that identifies a student’s  topic weakness and recommends a level to begin.

Immediate feedback given to students that is both contextual and relevant to their mistake and offers guidance on how to improve at the skill and exam level.

Practice real test questions at both the topic level and as comprehensive full exams to build up the confidence of test taking.

BUILD, MEASURE, LEARN

Validating and creating the Minimum Viable Product (MVP)

The path to building an MVP  that has product-market fit begins by always testing our assumptions with users so we can adjust our plan forward.   In order to quickly validate the product’s UVPs, I worked with marketing to craft a series of concurrent Facebook Ad campaigns.  Each ad ran with a separate unique value proposition message and directed users to a beta product landing page. We collected and compared the click through rates and sign-ups for 2 weeks. The data allowed us to make informed decisions about which UVPs  would be included and prioritized in our MVP features.

A split test Facebook ad used to test our key values

Methods

FB Ad split testing for 2 weeks

Key Insights

The FB ad campaign revealed that our students wanted multi-method explanations and practice real test questions as features for the MVP product.  This insight paved the way forward for our pedagogical design plan, concept app design, and our fit/gap study of how we can build our app using our 3rd party LMS and authoring tools.

Pedagogical Content Design Flow

Now that we confirmed which features would be part of our MVP, I worked with our math tutor subject matter expert to co-design a system flow diagram based on a pedagogical multimethod learning and practice process. This would become the foundation behind our content architecture and  task flows.

Pedagogical system flow

Prototypes  

After sharing knowledge with the production and technical team, I understood the capabilities and constraints of our LMS authoring tool and what usage data can be collected.  From there, I sketched out concepts and designed low-fidelity prototypes aligned to the task and pedagogical flows.  High-fidelity look and feel, was not the objective here. We wanted to first work out how this system allowed user’s to achieve critical tasks, such as being able to choose multiple method explanations.

A user-testing prototype representing the multi-method task flow.

Representations

Sketch and Invision prototypes

User testing  

Testing early and often with your users is critical in finding problems of your product before running the risk to costly development and redesign fixes.  So, after I designed the initial flows and basic UI, it was a great moment to test it out with our users. We brought in students for user testing sessions to identify any breakdowns in task completions, misunderstood copy and labels, common errors, and thoughts on how they felt about using the product. We followed a rapid and iterative testing approach which allowed us to learn, improve, and test quickly with many users. In each 1 hour session, we observed and probed our student as they completed a set of tasks with a interactive prototype.  The session continued with a 30 – 60 min internal review discussing and fixing revealed UX issues. After the UX fixes, we began another testing session with a different student.

A user testing session

Method

8 RITE usability sessions

Final Application Design  

After the completed rounds of UX improvements and visual design, the Trigonometry topic was built and released first as part of our trial version product.  This topic design served as a scalable production framework for the other 11 math topics, by incorporating reusable and modular components. 

The visual design look and feel served an important function in maintaining a calm and fresh mood.   Our bright yet cool color palette de-stresses the moments of test prep. We also included personified illustrations of student characters at critical and emotional touch points. They were a sort of  “friend along the way” support system – celebrating success and encourage prevailing.

Screen designs showing the flow of practice and multi-method explanations.

Launch of OnetGO

OnetGo was first released as a beta free trial version.  As our production pipeline finished additional math topics, we bundled them in updated paid product versions. We informed our current users about new topic releases on LINE, and used Facebook Ads campaigns to drive new traffic.

Measuring OnetGO

As our primary business objective was to identify product-market fit, we collected and tracked usage data focused on growth and retention on new, churned, and resurrected users over each week.

A new user is a student who accessed the application and completed at least one full question, answer,  explanation sequence.

A churned user is a previous week new user, but failed to complete at least one full question, answer,  explanation sequence at the end of the current week.

A resurrected user is a churned user who now came back to complete at least one full question, answer,  explanation sequence at the end of the current week.

In addition to retention data, we collected micro usage event information such as:  number of topic visits, page visits, subject visits, and session durations, time and days.

Measuring the usage behavior

Our secondary business objective was to use digital channels and mechanisms to optimize marketing and sales. Using Facebook Ad campaigns, and A/B testing of Product Landing Pages, the marketing and sales team worked to optimize the funnel and establish business monetization models of free trial and paid products.

Methods

Google Analytics, BiGQuery

Representations

Weekly Active User (WAU) Growth Accounting Charts, Google Data Studio Reports

Conclusion

The project’s progress and insights were given to the Executive Board to decide on the next stage of OnetGo. Launching OnetGo was an internal success in gaining valuable knowledge in the online university test prep market. Our new understanding about lean product-market design and development gave us guiding principles and best practices in how the R&D department approaches new digital projects. Below are some of the key takeaways:

  1.  Have realistic monetization business models. Determine at the start of the project what those are, how viable it is, and what and how much your users are willing to pay for.  Be clear at the project’s beginning that seasonal, short lived and nonseasonal recurring products bring different lifetime customer values.
  2. Season, short-lived products should be fully released or in proper timing. Students expect to have full access to all topics when they are ready to study at their own time. Don’t have a production cycle that compromises their expectations and trust in your product.
  3. Without custom development  and proper UX solutions, you risk losing your potential customers. Users expect a simple and streamlined process  to access to your product’s services for the first time and everytime. Any obstruction they encounter risks them dropping off for good.  So when planning, understand the risks when opting for using 3rd party services and tools that don’t allow for custom development.