Secret Story: ZELF
First Bank of the Metaverse. Instant money in messengers. No apps. No banks.
ABOUT THE COMPANY
ZELF is a leading Gen Z neobank embedded in Facebook Messenger, WhatsApp, Viber, Telegram and Discord.
Through our efforts, now any financial operation can be processed by an AI-powered voice assistant
#zelf_tags
Products and integrations which were run by me
CEO invited me to build a team and establish processes.
As a Chief Product Officer, I took over all development activities, i.e 5 product teams:

Messenger bot
Messenger, WhatsApp, Telegram, Viber and Discord
Mobile App
Play-to-Earn App (Android/iOS) without Banking Features
Marketing System (Web)

Cohort Analytics & Complex Referral Program
Back-office
Building a Modern Bank Backend
Treezor Banking API
Connecting Banking-as-a-Service Platform
Achievements in figures
+120 000
people issued ZELF cards
+400%
team growth
+27%
increased the conversion rate
+1
new country was launched (Spain)
Timeline
Jan’21
Jan’21
I started with a product discovery to understand ZELF customers and their needs. Conducting the market research, I found out that ZELF loses 40% of potential clients because of the absence of some features. I reprioritized the existing backlog and launched a new line of business. Summarizing all information, I created a new roadmap.
Feb’21
Feb’21
Updating a strategy, I made a recruitment plan. Then I optimized the HR system and personally interviewed more than 80 analysts, designers, product and project managers. After hiring the best candidates, I prepared an adaptation system for each.
Apr’21
Apr’21
The development process was chaotic. Employees often preferred to do easy tasks, but useless and low-priority. Working inconsistently a lot of features stayed in progress for many months. I divided the team into 5 products and implemented SCRUM. With new product managers, I established a new development process, so the company could predict the costs and execution time.
Jun’21
Jun’21
To finish the transformation project, I rebuilt the analytics and support system. I implemented cohort analytics to all reports and a hypothesis validation approach. So I found out that the support team missed 25% of requests, and the average response time was 7 hours. After changing the support software, scripts, and service standards, it changed to 0.05% missed requests and 8 min time response.
Let's talk! Hope to see you soon
Let's talk!
Hope to see you soon