Informations
Jobs
What is the article about?
In a cross-functional project, OTTO wanted to generate insights on the topic of the customer journey and use them as a basis for measures (e.g., improving the customer experience). The aim was to create an analysis basis that would make potential visible and offer added value for OTTO. The Business Intelligence (BI) and Customer Service and Logistics (KL) departments were involved in the project.
BI is a central driver for the development of the "data driven company". We use the latest BI technologies and make the customer journey and core processes transparent through Data Science! We gain insights from data that enable us to develop ever better strategies and processes for a successful digital future. To achieve our customer-centric "360° approach", we implement exciting use cases. Our core services are data, data science, business analytics and reporting. Among many other services, we design professional BI products such as reporting and analysis tools and develop prototypes for testing hypotheses and quick win solutions. The Pathfinder is one such solution.
KL is home to the experts for developing and redesigning services and optimizing customer-relevant processes from ordering to delivery. Service and customer orientation is KL's passion, and for this we have a whole range of tasks in this area, from project work, controlling, conception and organization to strategic topics.
Our claim is quite simple - only the best for the customer. And by this we understand the responsibility for the most personal customer service.
Other goals of the project:
In addition to the overarching goals, BI wanted to make a tool selection for customer journey analyses in order to offer specialist departments explorative analysis options; on the other hand, several data worlds (unstructured data and relational data) were to be linked. In order to offer our customers more targeted support on all channels and to avoid customer irritations, KL had, among other things, the goal of optimizing services on otto.de. To do this, KL was looking for a way to more easily identify problems and visualize the customer journey.
Problem:
After a comprehensive market analysis, it was clear that no tool existed at the time of the project that met our requirements:
A visualization of customer journeys was possible in some tools (e.g. Google Analytics or using D3), but the interactivity was either not given or only very limited. Thus, if one wanted to analyze journeys more precisely, one was forced to generate a new visualization for these same journeys, which proved to be very cumbersome and impractical.
Changing the level of detail, if at all, was also only possible for the entire visualization and not, as desired, for sub-areas.
An exploratory analysis of how customers move around otto.de and which paths the majority of customers take thus proved to be very time-consuming.
Solution:
For this reason, we opted for in-house development. In agile mode, both areas worked in close cooperation on the OTTO Pathfinder: KL programmed the tool and BI supplied the data. In regular exchanges, we reported to each other on current interim statuses and discussed how to proceed.
What exactly is the OTTO Pathfinder?
With this tool, it is now possible to interactively visualize and analyze any paths. The paths are displayed as so-called Sankey diagrams:
pathfinder
A Sankey diagram graphically depicts quantity flows. It can be used to depict any path, including onsite customer journeys - i.e., the paths that visitors take on a website. The diagram is to be read from left to right, i.e. the pages that visitors saw first are shown on the left (here: dark orange). The thickness of the arrows represents how many visitors used that path. When viewing, the user has the ability to filter paths, drill down specific nodes to three levels of detail (e.g., from assortment to product or page types to increasingly finer page names), select the number of steps displayed within the journey, view only the paths of visitors from specific marketing channels, and many other options. The tool can be used using Google Chrome; no additional software is required on the user PC.
Technical:
OTTO Pathfinder is based on Node.js technology, the Sankey diagrams are visualized using D3 library and Mike Bostock's Sankey plugin modified by us. The different data bases that the user can select in the frontend are stored on a MongoDB.
Sample Applications/Use Cases:
- Onsite Customer Journeys
- How did visitors behave on otto.de?
- Which pages were visited before and after a particular page?
- How does the behavior of different visitor types differ?
- Marketing insights from completed customer journeys
- Which marketing channels did customers use before they made a purchase?
- Which marketing channels typically appear together in a Customer Journey?
- Do customers who call about a given topic miss certain information on otto.de?
Source code:
The OTTO Pathfinder has reached a level of stability that we are very proud of and are happy to share with the public.
The source code of the OTTO Pathfinder can be found at the following link: https://github.com/OTTO-Pathfinder.
We would love to see the Pathfinder used by many users, both private and business, and welcome suggestions and feedback! Feel free to share features developed by you via our Git repository!
Have fun testing!
0No comments yet.
Write a comment
Answer to: Reply directly to the topic
Written by
Mario Hienen
Similar Articles
COSTNovember 12, 2024How we used a simple trick to save USD 500,000 in data transfer costs
03Increased efficiency of cloud usage ✓ Cost reduction ✓ Migration to dual-stack IPv6 ✓ Find out more now!DevelopmentCloud
DominikSeptember 30, 2024Developer Hacks – Modern Command Line Tools and Advanced Git Commands
014In this article, you learn more about a modern development setup, state-of-the-art alternatives to classic shell programs, and advanced git commands.DevelopmentWorking methods
