-
Meetings
-

#20: Meet Qonto

written by
Clara Cros

For this twentieth meeting NoCode Series we were able to go to the Qonto offices to discuss with Maxime, Head of Product, and Palak, Product Manager.

Focus on Maxime and Palak

Maxime has been working in fintech for about ten years now. For him, it all started at IbanFirst, another payment institution. The first thing he did at this company, over a period of two and a half years, was core banking.

🤓 Core banking is the engine, the heart of the bank, which will allow the creation of bank accounts or even the sending of transfers.

Following this, the company had its first customers and started to scale. They also wanted to develop their offer via API: to be able to make transfers via this API. This was a first step for Maxime in the world of automation.

He then moved to Mailjet, where he was able to specialize more in APIs. A year later, he moved to his current position at Qonto, Head of Product. The objective from the start was to build the core banking in 9 months: a big ambition when you know that in his previous experience it took him 2.5 years!

In order to reach this objective, the means followed and nearly 80% of the Product and Tech teams were put on the project. Thus, 9 months later, the objective was reached.

As the Qonto product grew and became more complex (especially on the core banking part), they had to reorganize into CFTs (cross-functional teams). The Product Managers now had a more IC role and worked on very functionally defined scopes. Maxime was thus able to navigate between the different CFTs and today adopts a more managerial role as Head Of.

Now the Product at Qonto encompasses 150 people including:

  • Product Managers
  • Product and Brand Designers
  • Product Marketing Managers
  • UX Writers

And it's about sixty Product Managers, including 2 Head Of.

Palak, on the other hand, has been involved in NoCode since more recently. He has a developer/engineer background and gradually transitioned to Product. He was able to work in Silicon Valley before coming back to France to start his adventure at Qonto. Like Maxime, he was able to be part of different teams and different projects including the CBS (Core Banking System).

When given the goal of developing core banking in 9 months, they quickly realized how far behind other fintechs Qonto was in terms of partnerships and connections. However, there was not enough bandwidth on the developer side to catch up. So NoCode appeared to be the perfect solution to this problem. Palak had both a tech and a product profile, so this solution was the perfect link between the two worlds in his opinion.

Qonto x Slack: the first integration made by Qonto

When the push to catch up with NoCode came, the teams weren't starting from scratch because Qonto had already launched its first API 5 years ago - well before the PSD2 Open Banking regulation where banks are required to open their data. This API started to be consumed by accounting tools that started to retrieve bank data and do integrations. So it was an accounting ecosystem that started building integration use cases on the Qonto APIs. There were about ten direct integrations but also others indirectly.

It was then decided, 2 years ago, to give visibility to this whole ecosystem: this is where the Connect marketplace was born. This marketplace allows to reference the most qualitative direct integrations. To accompany the launch of this marketplace, Qonto has launched its own integration: Qonto x Slack. The objective is that for each new transaction, a notification is sent via a Slack channel. This notification can be configured with specific criteria to avoid being spammed after each transaction. All this was done in traditional development, over 3 months, with a team dedicated to the back-end and one dedicated to the front-end.

💡 Connect welcomes the means of integration between Qonto and SaaS applications

A few days after the launch of the Slack integration, the following observation was made: each customer has a different need for their business, with a strong need for customization. Such an observation implied that we would have to spend several weeks developing an integration that is not core business (Qonto's business being payment methods).

It was therefore decided to stop further investment in Slack improvements.

The arrival of NoCode at Qonto

In parallel with the development of the Connect marketplace, Shubham Sharma came to Qonto and allowed them to take their first steps in NoCode and thus create their own apps on Make and Zapier. Qonto started touching NoCode via these two tools and later on a connector with templates was set up, with lots of triggers and actions, allowing customers to customize their automations.

All the integrations built in their Connect marketplace are "if this then that": something happens at the partner, they get it from Qonto and vice versa. So it's algorithmic logic - something that can easily be done on Zapier or on Make and therefore be done in-house. A PM that builds that algorithmic logic, so that will not use technical bandwidth.

🤓 NoCode was integrated in 2020 by Shuhbam Sharma when he was working in the company as NoCode & API Product Manager.

The question of creating their own language like PayFit and JetLang also arose. However, this did not correspond to their expectations because the creation of such a language implied having a back-end team that would be responsible for this language and its evolution, and therefore a tech team that would no longer be focused on the core business.

Concerning the Buy part, they had to find a way to make this Zapier white label: the Integration Platform as a Service actor was the solution allowing them to create their flows. The first integrations were done in 2/3 months with the Product Managers. Very quickly, performance problems were encountered and NoCode developers, otherwise known as Makers in-house, were hired. They are now 3 within Qonto and on their side Maxime and Palak have mapped their operations system as follows:

  • Value analysis (PM)
  • Value engineering (PM and Makers)
  • Delivery (algorithmic logic specifications)
  • AQ Plan

Today, the Makers are attached to the tech team and especially to the back-end team. According to Maxime, each developer must have a specialization and that's why they chose to work with Makers who already have a strong technical background. Of the 3 profiles recruited, 2 have an engineering background.

Today there are almost no quality problems on integrations and a rhythm closer to 4 weeks per integration by Maker. It is also about fifteen Built by Qonto integrations in the last 12 months (accounting, productivity, document management...).

Why use the NoCode?

  • Facilitate the export of Qonto data into an accounting tool and the repetition of this use-case
  • Do not have to use technical bandwidth and make Product Managers 100% autonomous on these subjects

Vision of the NoCode

NoCode reminds Maxime of the emergence of APIs a few years ago. He sees it as the same revolution, the same evangelization.

According to Palak, NoCode editors learn a lot from their customers and evolve with them, thanks to the feedback they receive (especially regarding versioning and alerting). Maxime and Palak also point out that we are still at the beginning of a professional use of NoCode but that these technologies are still increasingly adopted by companies, which opens a gap for platforms. This is still a nascent industry, full of opportunities, and one that is likely to become more and more democratized in large enterprises in the future.

Increase in skills

A training has recently been set up internally, based on what has been learned on integrations. This training is open to all PMs in order to offer them an introduction to NoCode tools via a use-case. The final objective is to realize an integration between Notion and Slack so that the PMs can get their hands on it and thus allow them to :

  • Understanding the value of NoCode at Qonto
  • Feel comfortable with the use of these tools (for their professional and personal use)
  • Consider integrating NoCode, including automation, into their own projects

The tool they are mainly taught for automation is Tray and the feedback from the PMs, who like to test on their own and receive feedback, is already strongly positive.

Having this knowledge of algorithmic logic is an additional skill for PMs because it also allows them to speak the same language as the back-end engineers (the Makers). The goal is to keep a logic of pairing between PM and Maker, that each one keeps his role and his specialization: a Maker specialized on engineering and a PM expert on product creation and requirements gathering.

At Qonto there are two types of squad:

  1. Very verticalized technical teams that need to understand the project from the core to the interface. To be a PM in these teams, it is essential to have technical knowledge, especially in algorithmic logic, in order to work with APIs.
  2. Other, more transversal squads, with less depth, in which the user experience and the interface are worked on. Other types of PMs, with a less technical background, can be recruited.

At Qonto, there are PMs with a very technical profile and others who are less technical.

Future ambitions

The current goal is to become as stable as possible and to have full control over the integration build cycle and to offer a whole bunch of new integrations via their Connect marketplace.

Many thanks to Maxime and Palak for this greatly interesting exchange! 🙏