Digital Product Development: 5 Trends of 2024 and Which of Them Work

Every fall, dozens of predictions come out about which technologies will reshape development the following year. Most of them repeat the same words: artificial intelligence, virtual reality, blockchain. The problem is that for a business owner or product manager, a list like that is useless until it's clear how much implementation costs and what it actually delivers for users.

We took five trends most often named as the top ones for 2024 and broke down each one honestly: what it actually is, where it genuinely pays off, and who should wait. Prices are approximate, for 2026, for custom development in Russia.

1. Artificial Intelligence and Machine Learning

This prediction came true, and then some. Language models and machine learning have made their way both into products themselves and into the process of building them. Within a team, AI helps write and review code, generate test data, draft interface copy, and sort through user feedback. Inside products, it answers support questions, searches documents, generates recommendations, and recognizes documents.

• Where the value is obvious: customer support, knowledge-base search, processing requests and documents, personalized recommendations.
• Where to be careful: decisions that affect people's money and health, made without human review.
• What to plan for: the cost of model queries as your audience grows, and requirements around storing personal data.

Adding an assistant built on an off-the-shelf model to an existing service costs roughly 300,000 to 1,500,000 ₽, depending on the volume of data and integrations. Training your own model from scratch isn't something most companies need: it's cheaper and faster to fine-tune or configure an existing one.

2. Virtual and Augmented Reality

This trend was expected to break into the mainstream, but by 2024 it stayed niche. VR headsets never became a household device, while augmented reality took hold wherever it solves a specific problem: trying out furniture in a room through a smartphone camera, trying on glasses and makeup, training on the factory floor, and navigating large facilities.

For most products, AR makes sense as a standalone feature, not a foundation. A simple “try it in your room” feature through a browser or app costs roughly 500,000 ₽ and up, with 3D models of the products as the main expense. It pays off in furniture, home renovation, and retail with a high price tag, where buyers are afraid of getting the size wrong.

Virtual reality, meanwhile, has found its place in corporate training: simulators for working with dangerous equipment and practicing emergency responses cost less than mistakes on a real site.

It's also important to remember the people involved. Employees need clear rules: which data can be sent to outside AI services and which can't. Trade secrets and customers' personal data shouldn't go into someone else's model without a contract and review, or the time saved will turn into legal risk.

A trend is worth adopting not when everyone is writing about it, but when you can name a specific task it will make faster or cheaper.

3. Progressive Web Apps (PWA)

A PWA is a website that behaves like an app: it can be installed on the home screen, works on a poor connection, and sends notifications. For the Russian market, this trend turned out to be one of the most practical. When some apps disappeared from foreign app stores, many companies released web versions and PWAs to avoid losing their iPhone users.

If the product mainly shows information and takes orders and payments, a PWA often covers 80 percent of the need for half the money. Native development is worth it when you need deep access to the camera, background location, Bluetooth, or high performance.

4. Voice Interfaces

Voice assistants are in almost every smartphone and many homes, but a dedicated voice interface for a specific product is rarely needed. Scenarios work well where a person's hands or eyes are busy: navigation, smart-home control, the kitchen, a warehouse, drivers. In other cases, voice remains an add-on to the regular interface.

Practical uses for business: voice search through a catalog, speech recognition in requests and inquiries, voice bots for first-line support. The latter have started sounding noticeably more natural since good language models arrived, but they still need a careful script and a quick handoff to a human agent.

Before launching a voice bot, it's worth listening to recordings of real calls and picking out the five to seven most common scenarios. A bot that confidently handles those and immediately hands everything else to a person annoys customers less than an ambitious bot trying to answer any question.

5. Blockchain

In the 2024 predictions, blockchain promised transparent supply chains, digital identity, and new ownership models. In practice, for most products it turned out to be overkill: transparency and data-protection needs are usually solved more cheaply with a regular database that has a change log and access controls.

Blockchain is justified where several independent parties don't trust each other and aren't willing to hand data to a single operator: interbank settlements, tracking digital assets, select logistics consortiums. In Russia, working with digital financial assets is regulated separately, and launching such products requires legal groundwork. A regular online store or booking service doesn't need blockchain.

How to tell a trend from a fad

The 2024 predictions showed clearly how fast the hype around a technology can diverge from real demand. AI and PWA became working tools within a year, while virtual reality and blockchain remained conference topics for most industries. To avoid spending money chasing fads, it's worth asking yourself a few questions before any technology decision.

• What specific user problem does this solve, and how do they handle it right now?
• Can the idea be tested with a cheap prototype in a month, rather than a big project over six months?
• Do your users already have the right devices and habits, or will you need to teach them first?
• How much will support cost a year from now, not just the launch?
• Does your team or contractor have real experience with projects like this, not just slide decks?

If you don't have a clear answer to most of these questions, it's better to shelve the technology and revisit it in a year. In the meantime, the market will show whether it works for others, and implementation costs will typically drop. Being second to adopt a new technology is often more profitable than being first: you pay less and learn from other people's mistakes.

Bottom line: what to take on right now

1. Write down the three biggest problems your product's users have.
2. For each one, check whether one of these trends solves it faster or cheaper than your current approach.
3. Run a small experiment for 4-8 weeks with a clear metric.
4. Scale only what actually improved the metric, and honestly shut down the rest.

Technology trends are useful as a list of ideas, not as a plan. The teams that win are the ones that take exactly what solves their users' problems from them, and don't spend budget on features just to get a mention in a press release.

This article is for informational purposes only and does not replace professional advice. Prices are approximate and current as of the review date. Found an error? Let the editors know via the contact page.