How Feeld Fosters Explorative Human Connection With Better Intelligence
Feeld reduced time-to-intelligence by 20 hours each week to spearhead data-driven decision-making at speed with Holistics.
The Story Of Feeld
Feeld is a UK-based innovative platform for alternative dating inspired to create an open space for people who are beyond labels and unchained from the necessity of coming from a very defined place with a very defined requirement. After 7 years of developments, Feeld has now grown beyond a dating app - advocating gender equality, racial equality, identity freedom, becoming one of the largest open-minded, progressive dating spaces worldwide with over 700K+ connections made and over 100 million messages sent every month.
The Challenges Faced By Feeld’s 3-Person Data Team
Every swipe made, every great match and every heartfelt conversation ignited come from the collective efforts of the entire team - from data, engineering to business, and marketing team
As the company grows, the increasing influx of data puts a dam on this effort: The bottleneck at the data team is continuously widened as business users are always hungry for the most updated data to make timely decisions in response to the growing customer base.
When Olaf Vogel, a Senior Intelligence Analyst, joined Feeld, his first mission was to build a data platform that can empower everyone to freely explore data and have all the necessary dashboards to support their OKRs and business goals, without having to teach users SQL and BigQuery.
Having experience with Google Data Studio and Looker, Olaf was not quite satisfied and set out to find a data platform that can answer one of his most urgent needs: the need for speed.
“We were searching for a self-service tool and we had previous experience with Looker, but we were looking for a more modern and faster tool to which we can implement a lot faster with basically all features that we were missing previously.
– Olaf Vogel, Senior Intelligence Analyst
Before using Holistics, Olaf and his data team had a longer implementation time for each data project. From gathering the requirements or making additional fields available, up to the development of the reports and making them available. By switching to Holistics, they were able to speed it up rapidly.
“So with Holistics, I think especially with us, in our setup with BigQuery and dbt, if we have a new dataset available and we want to implement it, we can literally do it in like two hours from start to finish, including all the descriptions and metrics, as compared to probably 10 to 12 hours in our previous tool
– Olaf Vogel, Senior Intelligence Analyst
With Holistics’s Self-Service capabilities, Feeld was able to remove the problem of having recurring requests coming from the business team towards the analysts - everyone can now create their own dashboards to suit their needs.
Feeld’s Favorite Features
Robust and Easy-To-Use Data Visualization: The data team uses Holistics to define business metrics, datasets, and reports for business teams. Business users can consume the predefined reports, or build their own reports and ad-hoc analysis with a simple, drag-n-drop, point-n-click report builder.
- “No cluttering on reports or dashboards, it’s really clean so we can have people actually focus on what’s going on with the data instead of looking at overly fancy dashboards with all kinds of distracting things in there. We really like the clean look and feel of creating reports and adding them to a dashboard. The main feedback from our business users was when are we getting more data ready to use?”
Caching Feature: Olaf’s team know that when people start creating their own dashboards and sharing them amongst each other, they will eventually hit a bottleneck where people have to wait and get discouraged from using the BI platform, so having proper caching in place is one of the key criteria when they search for a BI tool.
- “Looking back when I used Looker in my previous company and we had 900 users in Looker, when some people created the dashboards that were used very often, it created actually a massive queue. And because there was no caching in Looker, properly in place, it would just run through the database, create a queue, people were like waiting and waiting and waiting. I think that’s something that’s really optimized with Holistics which I don’t have to worry about anymore in the future.”
How Feeld Advances Internal Data Ecosystem With Holistics’ As-Code Analytics
Just as how Feeld is redefining the nuances of human relationships and pushing forwards dating space, their data team, in the same vein of spirit, continuously innovates their data workflow with analytics-as-code.
Feeld quickly adopted Holistics 4.0 - the latest version of Holistics with the capacity to do analytics work by writing code. This is done through a text-based markup language which is named AML (Analytics Modeling Language).
With Holistics As-code, a more streamlined and collaborative workflow can be quickly implemented at Feeld using a wide range of benefits that doing analytics using code provides:
- Version Control: Since analytics is code, analytics code can be easily checked into Version Control Systems like Git. This allows you to track who changed what when.
- Code Review: Every change to analytics code can be submitted for review before deploying to production.
- Code Reuse: Analytics logic can be expressively defined, refactored and reused across multiple instances.
“Holistics’ AML is a bit simpler to use, more optimized towards making sure that you have more reusability, easier to create your fields whereas Looker has a little bit more complexity in the structure, especially if you compare the models, files and the data sets files in Holistics versus the way you have to set it up in Looker - where you’re actually working with a lot more files to create the same.
A lot of what we could not find in other BI tools has to do with the speed of usage. Let’s say for example the caching system using Holistics is not only a lot faster when you’re scaling up with users, but also, the speed of implementing new datasets and making them available for self-service
– Olaf Vogel, Senior Intelligence Analyst
For Olaf and his team, the future shines bright: They expect to lean into Holistics to advance data at Feeld - having more data models, doing more advanced concepts and making them available for users.
In the past, their data team spent 20% of their time on maintenance, 40% on helping users and the other 40% developing new concepts.
Now, with Holistics As-code, they can go for 10% on maintenance and 70-80% on advancing data, and the other 10% on supporting business users.
“We want to make sure that we’re like building this whole data ecosystem, instead of gradually moving towards like, “Okay, we don’t really have time for that.” Which was like a lot of data departments jobs in the past, like, you know, maintenance, answering questions, we just want to like, keep on going with this speed and making them eventually show up Feeld is staying on top of their data game. In front of all the competitions. They left behind, not us.”
– Olaf Vogel, Senior Intelligence Analyst