Hello fresh

Hello Fresh

HelloFresh SE (HelloFresh) is an online food delivery company. It offers meal-kits, fresh ingredients, recipes, and ad-on offerings which include soups, salads and desserts, among others. The company delivers homemade and fully cooked meal-kits and pre-portioned ingredients to the door-step of its customers.

‍

What is System Inventory View?

An inventory tooling is important as Hello Fresh scale their business model.

This tool will ensure to buy just the right amount of inventory to avoid choking the warehouse capacity, avoid expirations and most importantly sub-optimal working capital utilization.

Systematic Inventory View (SIV)  aims to answer “which SKUs to order, when to order, for which DCs & how much? Giving trust and confidence to the numbers.” answering these questions requires complex data ETL procedures while ensuring data accuracy throughout. Without automation & flexibility these ETL processes can be tedious, repetitive and error prone due to manual intervention(s).

SIV has been developed from the ground up to deliver an MVP to democratize data concerning inventory & demand forecast for our users (buyers, category managers & inventory managers).

‍

How SIV used to worked?

While the MVP (built on a tableau dashboard) ensures that the users are given an aggregated data view and replaces the effort to manually query data from multiple sources, the next evolution of the tool is to enable information & actionable insights from the aggregated data.

This warrants a customer user interface which allows flexibility on the way we want to display the information & actionable insights to our users , far beyond the capabilities of tableau.

‍

Research

As SIV matures, its core purpose as a BI tooling extends to actionable insights and decision making which requires a modern User Interface which isn't supported by Tableau. SIV will need to support Actionable insights beyond general B.

The current setup is transactional in nature and at its best helpful for data aggregation, it doesn’t provide insights for decision making. An inventory explosion as the business scales warrants solutions which process data to provide insights, lack of which could result in suboptimal capital utilization, warehouse space constraints and/or expirations.

To ensure we have excellent procurement functions supported by an excellent inventory management system, we would need a custom user interface which can support the business’ needs by providing relevant, timely, accurate information as it scales over the next few years.

If we enable a customer user interface that displays actionable insight and allows data correction by our users, then we can reduce data preparation effort for ordering by 50% vs today.

‍

Unreliable data

At the moment, SIV doesn't get real-time warehouse information -- only once a week. Buyers have to gather stock level counts from multiple systems [WMS / FCMS]. In some markets, buyers have to check POs against every SKU -- a time-consuming manual process sometimes for around 200 SKUs.

This results in the service's low adoption rate, with only 2/5 markets using SIV.Buyers also need to check the SIV accuracy dashboard [Grafana] to troubleshoot inconsistencies, such as # of SKUs. Inaccurate data and additional processes are the main pain point preventing other markets from adoption.

To find: Metric around avg time spent referring to inventory

‍

Lack of defined processes

SIV [as a product] doesn't have defined use cases or user journeys. This leads to markets interpreting the data in different ways. It is also not clear to the user how to incorporate SIV into their buying process, and lacking guidelines for which usable stock levels to take into account. This mainly relies on buyer knowledge, which is not scalable.

To find: Metric around time spent for workarounds or sub-processes
Ex. NL reducing stock levels [manually deduct from orders]: 6-7hrs per/week

‍

Tableau limitations

Tableau is proving to be a frustrating user experience, since the refresh time takes 15-30min after the upload of the forecast into OT. The view is only on a SKU level, which is an additional step for buyers since they purchase by supplier. Education [around the tool] is also a pain point, with lack of awareness of the tool's functionality, as well as confusion around the data point calculation.

‍

‍

‍

‍

Definition

User needs + scenarios

I conducted interviews with buyers and inventory managers, in order to do a benchmarking of the different buying processes and stages.

The interviews were made for UK, Nordics, Italy, Dach and NL.

‍

Future planning (primary use case)

Markets make preliminary forecasts 6-10 weeks in the future [to reserve stock with suppliers]Around the 3-5 week mark, they update weekly forecasts to get more accurate numbers

‍

Emergencies / troubleshooting

Scenario A: Warehouse notifies buyers stock is low because it has gone bad or was unusable upon delivery

Scenario B: Supplier contacts buyers about an issue with delivery

‍

Getting actuals

Tracking performance and accuracy of ordering for operational / business goals Supplier management: tracking a supplier’s reliability [ordered vs. delivered]

‍

Interaction model

Market Tools

Design plan

‍

SIV Road map

‍

Wireframing & Prototyping

I created a series of wireframes for desktop, optimizing the journeys and aligning the wireframes with the business requirements and the information I gather from the interviews and data analysis.
‍
User now will be capable of experience the flexibility and functionality of the system interface, allowing them to explore different ways of displaying information, accessing actionable insights, and interacting with the data.

By going through the wireframing and prototyping process, I ensure that the final user interface for the SIV tool meets the specific needs and expectations of its users. This iterative and user-centric design approach reduces the risk of costly redesigns and usability issues later in the development process, ultimately leading to a more effective and user-friendly interface.

‍

‍

‍

Take away

Once the wireframes were iterated and approved by the client, I created fully interactive prototypes and conducted testing with final users.
The Systematic Inventory View (SIV) has played a crucial role in democratizing data access and decision-making related to inventory and demand forecasting within HelloFresh.

By providing users with a customize user interface that offers flexibility in displaying information and actionable insights, the SIV tool has empowered users to make informed decisions about inventory management.

This improved visibility into inventory data has facilitated proactive planning, enabling HelloFresh to meet customer demands effectively while minimizing excess inventory or stockouts.

As HelloFresh continues to evolve its inventory management practices, the next phase of SIV's development involves expanding beyond the capabilities of Tableau. This indicates a focus on building a more sophisticated and customizable interface that can provide advanced insights and recommendations based on the aggregated data.

By investing in this system, HelloFresh aims to further improve its inventory management processes and strengthen its ability to adapt to changing market demands.

Overall, the implementation of the Systematic Inventory View (SIV) has been a valuable asset for HelloFresh, helping the company streamline its inventory operations, reduce costs, and ensure optimal utilization of its resources. By leveraging data-driven decision-making, HelloFresh is well-positioned to boost customer satisfaction through timely and efficient delivery of meal-kits and fresh ingredients.

‍

‍