Ecosystem of basket-building features

In this project I applied system thinking with the goal to gain a holistic understanding of the ecosystem of features in the Ocado Smart Platform (ecommerce) that enable users to build their baskets.

The goal of this initiative was to obtain a big picture view of how those features can be combined, where are the overlaps and the gaps in the experience. This project embraced systems thinking, examining how the features are interrelated and connected in relation to user intention, goals, and needs.

My role: UX Lead // Project timeline: 6 months // Company: Ocado Technology (OSP e-commerce platform)

Problem

Ocado Smart Platform (OSP) offers several features that enable users to shop and build their baskets. These features operate independently and often cause confusion among customers. To find products and add them to cart users can use different features like Browse, Search, visit their Favourites page, and also Repeat Orders, use the Regulars, Recurring Orders, Lists and Recipes... The goal of this initiative was to obtain a big-picture view of how those features can be combined, where the overlaps and the gaps are and empower users by making the platform easier to navigate and improve the shopping experience.

Outcome

This initiative played a pivotal role in shaping the product roadmap and guiding product priorities. Through this initiative, the team gained a holistic and shared understanding of the big-picture view, enhancing our ability to make informed decisions. I took proactive steps to share acquired knowledge across the team, ensuring alignment and fostering collaborative insights. By leveraging the Opportunity Tree framework, I facilitated the identification and connection of opportunities, ultimately shaping a cohesive vision that deeply influenced our product roadmap and priorities.

This project was conducted in several phases including analysing the current state, defining the scope, running internal workshops and finally conducting field research in the form of in-depth interviews with online shoppers in Spain and in the UK. We also ran interviews with Retailers to get insights from their point of view.

I generated multiple deliverables to visualise the ecosystem, including diagrams, user journey mappings, and information architecture mapping. I also gathered existing research, analysed relevant studies and created a benchmarking focused on user automation and control. Additionally, I planned, organised, and facilitated several workshops to communicate updates and gather input from the wider team, through ideation exercises and promoting team alignment.

Pre-discovery and project definition

Current state analysis

I gathered and collected existing research about the features in scope. I mapped them in an affinity board and analysed the current state including qualitative insights and quantitative data.

Workshops

Ran workshops with internal stakeholders to get people involved. I invited and engaged main team members from domains and feature owners that should be involved. This was essential to collect internal knowledge and get buy-in from all stakeholder.

Defining the scope

Together with the PM and Researcher assigned to this initiative, after current state analysis and team alignment we defined the project objectives and scope.

Visualising the ecosystem of features

To visualise the ecosystem of feature and their connections, overlaps and gaps I generated a set of visual outputs and diagrams:

Feature connections diagram

Information architecture – where the features live in the system

Journey mapping

Levels of automation – user control vs. system automation

I designed the axis above to show how much of user control vs. automation we were enabling with our system and set of features.

Mapping the relationship between users and features usage

A diagram was generated to map the mission-behaviour personas and the feature usage to see any overlaps and where the features were connected or cannibalised, with insights from previous research per feature. The features colored in green are what we already had data to confirm our assumptions, features marked in orange is what we needed to dig deeper with research to confirm the assumptions, and what is marked in pink are the potential missed opportunities to enhance the experience.

By designing those artifacts to visualise the ecosystem of features I crafted a set of visual tools for the team to gather and have discussions around a very abstract concept (features in a system) and start identifying how we could improve the experience

Research

I could extract some common insights from previous research conducted but for achieving a holistic view of the experience we had to dig deeper.

While I was working on a set of visual tools, we started to plan and work on a research guide and strategy, together with the researcher assigned to the project.

Research goals

Understand the common pain points, delights and problems across features for users
Understand how features can be connected or complemented
Learn which features are the most valuable for retailers

Primary research method:

1-hour in-depth Interviews with users
1-hour in-depth Interviews with Retailers

The research was conducted by 2 researchers. After the collection of all data, I also participated in the analysis and preparation of insights.

Overview of research conducted

Research analysis and data

Journey maps were generated after interviews with an analysis of each shopping journey and the most used features. They were a great tool to help the team visualise how features were combined and used in the shopping journey and helped us to draw conclusions and insights.

Overview of data collected and analysed

Data was collected with the help of a Data Analyst in the team. I wanted to check a collection of metrics for all ecosystem features for:

% of active customers
Where features are used, touchpoints, access points
Are features used in conjunction? eg. Favourites with Shopping list?
What is the value generated in $? (cash added)
Most added items from which feature? (calculated after removal in Trolley)

The data helped to support the qualitative insights that emerged from the interviews with online shoppers and retailers.

Ideation

After extensive and thorough research. We set up sessions with the team to share the research insights and generate opportunities. For that, I chose the SCAMPER method which is a brainstorming method to encourage teams to view current problems through different lenses.

3 different sessions were run and in total there were 19 team members involved. It was a multidisciplinary team of PMs, Designers, Researchers, Engineers and Data Analysts.

Opportunity tree

An opportunity solution tree is a tool that is used to identify and organize potential opportunities to help us achieve our main outcome: Effective basket-building. To connect ideas to opportunities I used the Opportunity Solution tree framework.

After collecting insights including needs, pain points and delights from both users and retailers, we invited a multidisciplinary team to brainstorm potential ideas and organised them into categories.

From there, I used an opportunity solution tree to map out the connections between the main outcome, opportunities and the team’s ideas that could potentially lead to the main outcome and generate a pool of ideas and experiments to run.

Final outcome and proof of concept

From all the opportunities draw in the opportunity tree, we selected the ones that better addressed the main outcome: Effective Basket-Building and I designed design drafts. Here below is the main proof-of-concept:

Conclusions and problem statement refined

The complexity of our platform’s current information architecture is causing difficulties for users to discover and understand the shopping tools’ functionality, impacting both feature usage and user experience.

The naming and navigation hierarchy of the tools are unclear, leading to poor findability and a lack of awareness of the tools available to users.

Despite offering varying degrees of automation, the unique selling points of our tools are not emphasised, making it challenging for users to comprehend their value proposition and differentiate their functionality

Opportunities

"I want to easily access the features I use the most." // "I want to clearly distinguish between the features in the platform and how to use them"

How might we?

How might we highlight the unique selling points of our tools better to help users understand their value proposition? How might we improve the clarity of tool naming and navigation hierarchy to enhance findability? How might we create a more user-centric information architecture that takes users' needs and shopping behaviour into account?

Hypothesis

We believe that categorising our tools under the label "Easy Shopping" reflecting users' shopping behaviours according to the level of control they desire over their shopping journey, will simplify and align the information architecture with users' contextual actions. This will enhance findability, clarity, and differentiation of the tools' functionality and value proposition and lead to more efficient navigation and increased adoption of features.

Proposed experiment:

Group and categorise our tools under the “Easy Shopping” label with the goal of aligning our tool offerings with users’ actions.

Our research indicates that users have diverse shopping behaviours, with some seeking full control over their shopping journey and others opting for greater automation.

This proposal allows an envelope for scaling features that can be grouped and aligned with users intent.

Overview of other experiments generated and how I structured them to share with the team

Ecosystem - basket building - Inspiration

Ecosystem - basket building - Improve user control (1)

Ecosystem - basket building - Experiments - Keyword Lists

A set of opportunities and ideas were generated in this project. The POC and design drafts played an important role in shaping the product vision for future improvements, influencing the roadmap and providing the team with a holistic and shared understanding of the big-picture view of the e-commerce Ocado Smart Platform.

Let's have a chat!

This was a long-term project with a lot of discovery research, design exploration and great team discussions.

I would love to walk you through the process and show more of the outcome.

Note that this proof of concept is designed in a testing environment and is currently under revision for implementation.