EIU: MARKET EXPLORER
JUNE 2016
Project introduction
Market Explorer is an interactive data exploration tool used by consumer industry professionals to identify the best markets in the world for their products. It was launched in 2016 and represented two-years work from initial concept through to MVP, launch and then post-launch releases.
Market Explorer was a groundbreaking product for the EIU. It was the first new product in over 5 years that did not represent a predictable, incremental change to existing products that had been par for the course for the EIU’s 70-year old subscription service. Instead, it trialled a new approach to product development that was far more user-centred and design driven than previous products and the eyes of the business were trained on it looking for validation of the approach as well as an end product that stood out in the market place.
Audience
The target users for Market Explorer are a fairly niche audience with specific and identifiable needs, encompassing marketing, forecasting, business development and strategy professionals in fast moving consumer goods (FMCG) companies. Their primary goal is to scan the globe looking for emerging markets and growth opportunities for selling their products and services. They would traditionally do this through a lengthy process of finding and combining different data sources to conduct analysis. The aim of Market Explorer was to streamline this process for them, giving them a tool with which to gain immediate results and insight.
Team
The Marker Explorer team consisted of a development team in New York with scrum master and QA testers, then in the London office we had myself as the lead designer, a business analyst, product owner and a cities economist from our Economics Unit. For 3 weeks of the project we also employed a contract visual designer, who defined the initial design language for the application.
Throughout the project I was responsible for communicating our vision, intentions and progress to key stakeholders and internal teams through regular presentations and sharing sessions.
Constraints
In spite of this being a project that was given carte blanche to design a new product in the way that we wanted, the business at this point was still wary of including externals in the design process. In particular we faced opposition from some senior stakeholders who were not keen on letting the design team talk directly to existing clients or prospective customers, nor share unfinished work in progress with them. This made getting external feedback on the product difficult and we were sometimes required to use intermediaries in order to talk to users or even to work covertly to bypass roadblocks.
Design process
Product definition
The first few weeks of the project were all about taking the initial business insight and developing it to a point where the digital team could articulate a strategy and develop a plan to bring it to market. I conducted desktop research to understand the market, did some empathy mapping around target customers and created a straw man concept. This early prototype was shown to some of our existing FMCG customers as a way of testing the water with our initial position and I was pleased to find that assumptions we had made around what problems Market Explorer would address, who it would best serve and what it would enable users to achieve were largely validated. I articulated the vision for the product in a simple, solution-agnostic statement.
In less than 10 minutes, consumer industry professionals will be able to scan all global markets and identify the best opportunities for selling their products.
I also completed some high level thinking around the service design of the product. Market Explorer would be accessed as a subscription product from the EIU's customer platform so I needed to consider how this would integrate with the existing offering, how users would be able to login to the product, how the data might flow into other services such as the EIU's existing data tool, how new users might find out about the service and how it would be marketed.
Starting with data
After an initial direction had been established I started to look at the data and some sample econometric calculations with the Economics Unit. I wanted to understand the demographic and income data series that would be feeding into the tool, the variables that were available in these series and how the various data entities were related. I spent a lot of time with SME's in the business asking questions and attempting to understand the economic models behind the tool.
I prototyped the data in Excel to test a few ideas for ways of visualising the data and solidified my thoughts around creating a scoring and ranking system for markets to make them easily comparable.
Mapping user journeys
I next began to work on mapping out the end to end journey from logging in, initial selection of target market through to results pages and outputs. One of the key ideas for Market Explorer, as its name implies, is that users are encouraged to explore the data and one of the things I worked hard at was creating an organisation of screens, a consistency between them and a binding narrative flow to support this.
As soon as I was able to present some ideas with the team they were able to critique, validate and make suggestions. Presenting ideas early and often helped to move on and develop shared understanding quickly.
As our ideas progressed we started to converge on the concept of having 3 screens showing different views of the data. We would have a table to give access to the raw numbers and show the ranking of markets, a chart to visualise the results and a map to show geographic clustering.
Market Explorer flow
Prototyping
Using Sketch and Invision I then started to work up some of my initial ideas into a higher fidelity prototype. I created a click through prototype to illustrate the end to end journey from target market selection through to the visualisation of results.
The advantage of a clickable prototype at this stage was that it brought to life what up to this point had been a fuzzy concept. When shared with the team and stakeholders it served as a great conversation starter and a way to quickly review, build feedback and ultimately test the strength of my ideas. I continued to prototype, share and refine my ideas over a couple of weeks.
In order to get feedback from the target audience I set up some remote user testing using a service called Validately that allowed me to link in my Invision prototypes and create a series of tasks for users to complete. I then conducted a number of moderated tests where I was able to observe the users undertaking the tasks and then follow up with them afterwards.
Requirements gathering
One of the things we were keen to do quickly was to get out of a fictional state and to build a first stage real data prototype. This was primarily to test the data and some of our initial concepts with actual users.
Working with the engineering team, project managers and product owner, we agreed on an MVP for the data prototype and I worked with a BA and project manager to define and write the epics and user stories. I then prepared my deliverables for hand-off to the engineering team.
For each sprint I prepared a number of deliverables that would typically include user stories with accompanying written requirements, a click through prototype with interaction notes, flat images of the pages or components with detailed notes, a style guide and motion guidelines.
The sprint cycle
We worked on a 2-week sprint cycle and I was typically working a couple of sprints ahead of the development team. Each week I would review with the whole team my work in progress, in differing levels of fidelity depending on the stage of the work. Sometimes sketches, flat designs, click through prototypes or animation sequences. Presenting was always done with the New York office using teleconferencing software, which made things quite difficult at times. I learned to be very clear about explaining what we were looking at, at what stage we were with the design, what decisions had or hadn’t been made and what feedback I was looking for.
As part of a typical Agile sprint cycle I also attended daily stand ups, backlog grooming, planning and review and retrospective sessions.
MVP
The MVP was developed over a 2-month period and proved to be a really important phase for the project. It immediately allowed us to test out some of the high level concepts we had developed and we quickly saw both where our ideas were working, but equally where they were failing. We saw that users were often overwhelmed with their first view of the results and we decided that we needed a better entry point into the tool that would give users a softer-landing before exposing them to the more complex, data-rich visualisations of the chart view.
Once I could see how the data was visualised for real, I also realised that the style of chart we were using didn't work well. My original concept was to map the data points in a Gartner Magic Quadrant style so that the best performing markets appeared in the top right quadrant. In reality I found that this was rarely the case and after playing around with some ideas for asymmetric axis and garnering feedback, I decided that a simple chart with x and y axis was best matched for displaying our results.
Following the release of the MVP we then moved into a second phase of development that would respond to the areas of improvement we had already identified as well as building out further features for filtering, deep diving into individual markets, saving and exporting data and graphics and support and help screens.
Design language, interaction and motion design
For a 3-week period a contract visual designer was brought in to work on refining the design language for Market Explorer, while I continued to finesse the interactions, micro-interactions and motion design.
Launch release
From the point of releasing the MVP we continued to build, test and refine Market Explorer over a 6-month period. In June 2015 we launched the new product. At this time I was involved in the marketing effort behind the launch and I designed a one-page brochure-style site to support this. I also helped to train the sales teams in the US, Europe and Asia, giving demonstrations of the tool alongside our cities economist.
Post launch
There was significant interest in Market Explorer from the point of its launch and immediately some large clients took out subscriptions. With real users on board we could now start to collect quantitative data through analytics as well as regularly speak to clients or receive feedback through account managers.
We have found that Market Explorer is providing value for customers for early exploration of market scanning, but we are repeatedly hearing that users would like to be able to include more industry specific data when using the tool. The EIU's data is typically macro and broad and doesn't go deep on specific industries, but we are now looking at ways to potentially enhance the data offering for some sectors.
Retrospective and lessons learned
Personally, this was the most challenging project I have worked on - it was complex, high profile, but also one in which I learned a great deal.
Internal communication was an important part of the project. The new approach to designing required a larger internal comms effort to demonstrate that this new way of working was having the desired results. There was a lot of exposing people who were not designers to the design process, with the purpose of getting better buy-in and involvement. I met some resistance as would be expected, but I chose to pick my battles and fight for what I thought was important. Twice monthly playback sessions ensured that senior stakeholders were kept in the loop and were able to give feedback or voice concern over misalignment.
The project itself has given greater visibility to design in the business and the C-suite are now talking enthusiastically about making the EIU a more user-centred business and actively advocating getting closer to customers. This organisational change alone feels like a big win for me.
Other projects
