Skip to content
back to the presentation
Talk summary2020CTO

The Web in the Mobile-First Era

Prioritizing mobile applications does not make web engineering disappear. The talk examines how web technologies support customer acquisition, interfaces across platforms, and data preparation for applications. The history of browsers and tools provides context for a practical question: which responsibilities remain with web teams, and how does their ownership of the product expand?

GDG Party3 min read

This summary is shorter than the original material; examples and assessments reflect the author’s experience and views.

The main thread of the material
01

Mobile products change customer acquisition

The 2020 statistics cited show widespread mobile internet use and the influence of Google and Facebook on platforms. The historical overview connects browsers, JavaScript, Node.js, interface libraries, and mobile tools. Its purpose is to show the expanding reach of the web stack, rather than make a once-popular framework the default choice for every future project.

In the Tinkoff example, the traditional customer journey started with advertising and a landing-page application form. The mobile app came later, for using the product. A newer journey invited customers to install the app first and apply within it. Mobile therefore took on part of customer acquisition as well. The web retained its role, but channel boundaries shifted with the user journey.

02

Platform choices depend on product maturity

Web technologies extend beyond the browser: Node.js brings JavaScript to servers, Electron supports desktop applications, and mobile approaches offer different forms of code sharing. A shared language does not remove platform differences. Device capabilities, interface requirements, and technical limitations must be considered separately from the attraction of maintaining a single codebase.

For testing an idea, the author suggests starting with a web application and, where appropriate, PWA capabilities, then considering a mobile wrapper. A validated idea with sufficient resources may justify native development. React Native and Flutter offer another trade-off when maintaining several separate applications is undesirable. These are investment choices, not a claim that one technology serves every product equally well.

03

Backend for Frontend brings ownership closer to the interface

A general-purpose API can force clients to make several requests for one screen and combine the results themselves. This logic is duplicated across platforms, while the interface team depends on another team's priorities. Backend for Frontend addresses this through a server layer tailored to a particular client and owned by the corresponding team.

Calls to shared services and response composition move into that layer. Client code becomes simpler, and the server component can be updated independently of installed mobile apps. However, it introduces another failure point, and response latency depends on downstream calls. BFF therefore expands both the capabilities and responsibilities of frontend teams, requiring an understanding of server development and the operational consequences of their design.

Takeaways

What to take away

  1. 01Mobile first changes how user journeys are divided among channels without eliminating web engineering.
  2. 02Native, web, and cross-platform approaches should be chosen around product maturity and platform constraints.
  3. 03BFF simplifies clients and improves control over changes while introducing another server layer to operate.

Sources