5 weeks
Digitalising Novelty Orders (The LEGO Group)
Overview
One of my first assignments as the Lead Designer in the B2B WebShop (The LEGO Group) was to design how retailers could order new products from The LEGO Group. I loved to put to use previous B2B knowledge and engaging the product team to get it done within tight deadlines.
Approach
When I started working on this new functionality there was an existing concept left my the former designer in the team. The concept was visually appealing and used components already existing in the design system (KLIK Design System)
In my first review of the concept, I was intrigued. The flow used tiles, which is a common UI pattern for B2C e-commerce, but not as usual in B2B in my experience. I quickly outlined my existing knowledge about the B2B ordering patterns and decided to start collecting data (lean research).
The concept also used an approach called "Packages” which would allow packaging a few products together and sell them as regular subscription. But this concept raised concerns early on from the Legal specialists I talked with during my research.
Key Differences between B2B and B2C e-commerce
(From my observation)
B2C Shopping
- I browse to find what I want to buy/impulse buying
- I inspect details
B2B Shopping
- I buy based on existing selling data
- I order large quantities, for different locations
So knowing these key differences outlined a few unknowns:
- How many products are launched per year?
- How many products are launched per launch window?
- What information we have about the products 9-12 months before they launch?
Resolving these knowledge gaps quickly would allow a better concept and avoid engineering waste later on.
Fast learning loops & how context matters
I learned quickly showing and testing the concept to internal specialists.
Observation: at this point we were was not allowed to test it with the real users - something we resolved months later.
Key learnings & risk mapping:
- Only a few retailers were allowed to order novelties due to the high cost to enable the process (which was manual using spreadsheets)
- I also discovered we didn’t have access to the artwork for new products 9-12 months before launch
- From a business point of view, releasing anything that allowed retailers order novelties was better than trying to do a perfect solution (this was valuable to ensure progress over perfection later on)
- We received feedback the packages concept could be risky, due to regulations in a few countries. It was also highlighted retailers had their own purchase patterns, for example, customers in certain regions would only buy a certain theme such as LEGO Technic, Disney Princess, etc - therefore packaging products would not enable their usual process.
With the information at hand, I updated the concept and made a few changes:
- Opted for a table instead of tiles. This would help achieve:
- Faster management of quantities
- More familiar learning curve, due to high similarity to spreadsheets
- Seamless micro interactions by adding fresh UI patterns such as + and - buttons inline with product list
- Removed custom artwork to lower complexity
- I also took the opportunity to take a bold decision to simplify the concept to the barebones. Just ordering products, no packages - so we could have time learn and mitigate risks in more complex solutions.
I believe in making simple systems that can evolve, and getting to the most simple version seemed like the right next step.
Testing simplified approach
After review and the buy-in from the Product Trio (PM and Tech Lead), I tested the new concept with SMEs (Key Account Managers) and stakeholders.
The feedback was super positive and we could feel the excitement to the solution coming to life!
Co-creation with product team
Engaged the team in co-creation both remotely (using Miro) and physically in office
With the first prototype tested but still early on, I decided to get the product team involved to ensure feasibility. When talking to a few team members about doing this, I learned the team was never involved in problem framing before, as they were used to get why, what & how and they would get to implementation mode. But in my experience, this order taking mode leads to a passive team. I saw an opportunity to engage the team before we had too much defined.
To speed up the process and allow the team to join the collaboration, I stepped back from the high fidelity prototype and created low fidelity mockups in Miro so they would feel more comfortable to contribute.
Given they had never done this before, the team needed time to get engaged and had had a great session. With the technical path decided we were able to make significant progress in the next few days.
By the end, I always ask for feedback so we can improve on future sessions. Some of the feedback:
So cool to be a part of the process at this stage already, great way to be engaged in the feature, and I feel much more passionate about developing this now. — Senior SW Engineer
Bringing in the whole team early on makes me feel like we are a proper autonomous team making our own decisions and setting our own direction <3 — Security Engineer
UI Strategy, refining and further testing
With the technical feasibility cleared, I worked in quick feedback loops with the Frontend Engineers to refine the UI interaction. We worked so closely that they would even help in Figma, so I could tackle other UX tasks.
We have opted for a immersive flow to reduce any distractions during checkout, allowing retailers to get the task done with reduced friction.
Risks & Release Strategy
We had mapped a few risks during the discovery, and based on them I influenced the team to consider a phased approach for the release instead of a general release to all markets and regions which is something my team would do in a former role (and usually a sign of product team maturity). But the team had never worked this way and they lacked the knowledge and infrastructure.
In a few talks with the Product Manager and Tech Lead, we decided the phased release was the right way to go. The Tech Lead found out another team (who worked on Lego.com) had access to the right infrastructure (Optimizely) so we could do a safe, phased release strategy.
I consider that a great win for the team, and then could safely test and resolve all edges before releasing the new functionality to all regions and markets.
Outcomes
This was a fun project where we worked fast and had a great momentum. The product team has done a great job mitigating the technical constraints to meet the release deadlines.
I love that I had the chance to use different skills e engage the engineers in the problem framing phase - as well as how open they were to working this way.
A few outcomes:
- When launched, the novelty orders functionality reached 94+ CSAT (Customer Satisfaction) and it has sustained a 90+ average one year later.
- Reduced cost to serve through self-service
- Improved demand planning by allowing retailers order novelties 9-12 months ahead of time
- Increased solution ownership within product team
- Established ways of working in Product Trio
- Increased maturity in release strategy (segmentation/feature toggles)
- Seamless adoption by retailers (only minor adjustments after launch)
You achieve perfection, not when there is nothing else to add, but when there is nothing else to take away. — Saint-Exupéry