Taste Editor 2.0

MVP 1

(Due to a non-disclosure agreement the information on this project is limited.)

“A curious case of

Taste-Editor“

This application has quite possibly helped make 50% of the products that you and I consume in our daily lives. That may be toothpaste, coffee, morning breakfast, that sandwich you had for lunch or a nice dessert. Taste-Editor is an integral application for the business, garnering little over $2 Billion worth of profit for the year 2020. Needless to say redesigning it was indeed a curious case.

Problem statement.

A flavorist is someone who uses chemistry and creativity to engineer artificial and natural flavors. They are the Da Vinci of food and beverages. On a daily basis they use the legacy Taste-Editor application to create flavors for IFF’s customers. The current application is over 15 years old and is in dire need of redesign to make the flavorist’s process easier. Also, the company and processes have greatly evolved in the past 15 years. The innovative functionalities need to be built in new design. IFF is a 130 year old organization with a lot of historical data. Why do we utilize this data by using AI & machine learning and build a predictive model as flavorists build their formula? Simply put- to give IFF an upper hand over competition.

My role What I did

In 2021 I was tasked with taking what we learned from the legacy application and exploring what the new Taste-Editor might look. I was the lead designer documenting research, leading design sprints, and best practices.

UX research & analysis
Design Sprint
Wire framing
Interaction design
Prototyping & motion studies
Documenting patterns & best practices

UX Research & Analysis

(Formula creation is one of the best kept secrets within IFF so I’m not able to document a majority of the work I have done)

There was already a strong foundation for us to leverage when it comes to formula creation. What the UX journey users had to go through was very terrible. We felt this problem could be easily remedied by the inclusion of latest technical capabilities within IFF .

To kick off this exercise, we engaged in user interviews on a global scale. With a number of conversational models to gauge user interest - including their current experiences. Using these interviews a detailed user journey and empathy maps were articulated.

Aligning on goals

Before beginning any design work, I brought together the PM, PO, research, engineering and architect leaders as part of a 5-day sprint in Union beach, New Jersey. People joined from the US, Brazil, London, France and India, with the goal of agreeing to a unified vision.

Journey Mapping: How does users currently use Taste Editor?

IFF has flavorists that have been working with company for over 30 to 40 years now. These experienced users have developed their own methods of working. At the end of the day this is an art. So it was apparent that if we were to redesign this application then it should be adaptable and fully customizable as per the user’s need. All while keeping in mind that stricter business rules are followed and every work item gets tracked and reported. For these purposes, an intense and laborious mapping was conducted. The end result was equally rewarding.

Note-N-Map

As we progress further into design sprint, all stakeholders individually map the user's current experience, with all of its flaws. “Don't improve things or add new ideas.” This was imperative given the creative nature of the business where every user has their own defined way of working. I would highly recommend this activity as it is a great way to pin existing pain points and avoid future mistakes.

Competitive Analysis

Design sprint board

It’s always fun watching people learn new agile ceremonies. Taste Editor design sprint was done fully remote. The team loved the gamification of idea sharing. One feed back we received repeatedly was that sprint was a good structure, rather than just interviewing stake holders for a longer period of time. This way we can validate ideas faster and as a bonus receive a prototype as a result. During this particular sprint we came across a lot of technical sprint questions, which the team had to go out and do R&D on.

Building a smart grid

Due to the complexity of business use cases, we built a smart grid that is not just responsive but AI and ML driven. The user feels like they are not alone while formulating. As they build the formula the system works with them showing relevant data, adjusting columns, exposing relevant regulatory checks and errors the user might have committed. On top of this, it predicts the success rate of certain formulas in the market, which is a huge cost saver as IFF’s biggest cost is wastage of samples. In order to come up with such a grid we partnered with a vendor who is an expert in such use cases along with IFF’s internal development & engineer team. This activity was an eye-opener and rewarding as I was fortunate to be given the opportunity to work with some of the industry’s best.

In MVP 3 one of the future functionalities we are planning to bring is voice command capabilities using NLP. POCs are already in the process. How cool will it be to talking to “Jarvis“!

Final Look

Here’s the final look at the product which is MVP 1. It has 300+ screens. Due to the NDA some of the data and screens have been altered to maintain confidentiality. Having done the design sprint process speeds up the user validation, technical feasibility and iteration process. As described in design sprint, users wanted an intuitive system that works for the users, not make the user work for the output. A system that guides the user along the way, run regulatory check and other AI/ML driven functions in the background.