Global Rollout of Starbucks.com
How do we scale Starbucks.com across global markets?
I worked on a project to localize Starbucks.com across its global markets. At the time, many countries or regions had homegrown, sometimes wildly different, version of Starbucks.com. This wasn’t ideal for people who were traveling, and it wasn’t ideal for the business or the brand. Starbucks wanted their site to look and feel consistent all over the world.
At the end of this project, each global market would be getting a new site from the global HQ in Seattle. While the UI would be largely the same all over the world, we didn’t want to be overly prescriptive with the UX. Market owners (the marketers and operators who run the business in a given region) have the best sense of what’s right for the business and customers in their region. Our approach would have to strike the right balance between constraints and choices, giving market owners guidance without being too limiting or too open ended.
Getting Started with a Site Audit
My first step was to go through the the U.S. site and create a list of features and content.
I started by looked at the site and documentation we had from the build. I also set up meetings with developers and the UX lead to a) make sure I didn’t miss anything or hadn't combined features and b) to understand the dependencies and requirements for each feature and content block.
Some parts of the site were static, while others required ongoing updates. Ongoing updates required people to make those updates, which required budget allocation. Then there were parts of the site that were required by law in certain regions: things like nutritional information.
By the end of this process, I had list of features and content mapped to dependencies in three areas:
- Customer Insights (attitudes, behaviors, and preferences towards digital, the brand, and the coffee category in general)
- Organizational Structure (team, budget, and business priorities)
- Market Constraints (laws and regulations)
In parallel to the site audit, I had to research each market. Other than language, what made the markets different and how might those differences affect the site?
I pulled market reports from sources like Gartner and Forrester, looked at internal research, and set up calls market owners . In talking with market owners, I got hear their perspective, better understand their market’s needs, and get web analytics for site usage.
These are some of the questions I answered with secondary research: How important and widespread is digital in each market? What do people use the site for and what's important to them? How do people feel about the brand and about coffee in general?
A Framework for Localization
I defined a minimum feature set that every market would get plus a menu of advanced options that could be mixed and matched as needed.
What's the minimum viable site?
At an absolute minimum, what does the site have to do to add value for people? Which features add the most value and require the least resources? These must-have features defined the baseline that every market would get: the minimum feature set.
The remaining features were labeled "advanced" and grouped into tiers. Each tier represented a level of "market readiness" required to support the feature. Market readiness was a function of things like budget, team structure, and business priorities.
Which of the advanced features are right for my market?
Some choices were obvious. For example, if a market doesn't have a loyalty program, then their site didn't need features and content related to the My Starbucks Rewards program. Others choices were less obvious: Should a market have its own blog with commenting enabled? Ultimately the decision came down to the market owner, but it was our job to provide guidance and recommendations that would inform their decision.
For each advanced feature, I created a diagnostic quiz that helped to answer the question, "Is this feature right for my market?"
To create the diagnostic, I went through a checklist of requirements (dependencies and resources) and compared them to the benefits and value of each feature. The diagnostic would help market owners make an informed decision by helping them consider the following questions: How much do I need this feature, if at all? Can I afford it? Is it worth it?