Brands over the Last 3 Years in AI Startups Today, the winners are those who respond first to the needs of customers, who customize their services and offer the best price compared to others.

Customization is what the modern customer wants from any product or service but to provide it, companies must know the demand of their audience accurately. Although many administrative processes have already been automated with the help of sophisticated systems such as CRM, CMS, WMS, ERP and other tools, these still do not provide any analysis or answers on how to customize the approach for each customer.

The main obstacles companies must overcome are:
  • Inability to process collected data (Big Data).
  • Lack of staff dedicated to customization.
  • Legacy technologies that have been implemented for years.

This is the reason why P2P markets and large companies are implementing AI in their systems and processes, to automatically analyze the data or information obtained. Solutions based on artificial intelligence have already found their way into areas such as e-commerce, financial services, health, tourism, transportation, and others.

The most important technology companies on the planet are in a frantic race to acquire startups dedicated to the study of Artificial Intelligence.

Machines learn from the context they have and become more and more precise in the suggested decisions. As we write this, companies can use the power of artificial intelligence and machine learning to improve their customer service:

  • Predict user preferences.
  • Optimize prices.
  • Detecting fraud.
  • Personalizing communication.
  • They explore the use of IA to attract customers.

Predicting User Preferences

Finding the key that moves each client's heart is possible only when you begin to treat him or her as an individual by providing a personalized experience. World-renowned companies such as Airbnb, Uber, Pinterest, Starbucks, and others actively use artificial intelligence technologies to provide personalized suggestions.

Custom Airbnb Search

Airbnb, the leader in p2p residential real estate markets, with more than 200 million guests in total and 3 million exclusive listings, focuses its efforts on enhancing the personalized search experience.

It's no easy task to compare a large number of listings and select just a dozen to match the user's exact preferences. That's why, starting in 2015, the engineering team built an artificial intelligence model that analyzes more than one hundred variables at a time, to customize the search in real-time.

Previously, search results were retrieved according to predetermined software rules, taking into account only a few factors, such as price and number of rooms. Its ML algorithm, called Embedding Listings, analyzes the data collected from the user's previous search history and wishlist pins to combine them with the desired features of the list that may include location, availability, services and other options. The system sorts the results into a real-time personalization search and a carousel of similar listings to display the most suitable variants.

The success of such an approach was demonstrated during the first tests of the AI-powered search mechanism. The percentages of CTR and reserve increased by 21% and 4.9%, respectively.

Airbnb has grown rapidly since its inception in 2008 when only 1 apartment was rented each day.

Optimized Prices

How do you stay profitable without increasing prices? This is the question that affects p2p market service servers, ride-hail drivers and online retailers. Analysis of multiple variables to suggest the right price here and now: this is the task most frequently devoted to artificial intelligence technologies.

Airbnb Smart Pricing

The Airbnb platform works with two types of customers: hosts and tenants. Therefore, by putting "cultural ideology" and affordable prices on the table, Airbnb managed to win the love of travelers looking for accommodation; however, Airbnb has to struggle with the no less important tasks of retaining hosts and attracting new ones.

Airbnb's analyses always explore their customers' progress and quickly discovered that the biggest challenge for the hosts was pricing. In monitoring user activities, analysts noticed that many new hosts left the site because they were unable to calculate the price. You must set a price that is competitive between similar listings but also provides you with sufficient compensation, so you don't have to worry about economic losses.

This is why Airbnb engineers implemented an ML model that provides daily suggestions for price optimization, taking into account multiple indicators such as location, seasonality, amenities, prices of similar accommodations, current listing prices, and availability. Hosts are not required to follow the recommendations, but the study says that "when a host selects a price that is within 5 percent of the suggested value, they are almost 4 times more likely to be reserved than hosts whose prices are more than 5 percent different.

Fraud Detection

One of the most exciting benefits of artificial intelligence technologies is how they work with cybersecurity. No one knows when it may happen, but it is extremely important to react to unauthorized actions in real-time and try to avoid any repetition.


PayPal is one of the largest electronic payment services, as it processed 7.6 billion transactions with 227 million registered customers in 2017, with astronomical revenues. No wonder this is a desirable platform for hackers.

PayPal engineers combined AI with their security system to analyze user behavior in transactions and identify unusual activities. If a behavior is reported as fraud, the system marks it and the files as a "feature. If the pattern repeats itself, the fraudulent transaction will be identified and prevented immediately.

Also, artificial intelligence solutions make it possible to differentiate suspected fraud from actual security breaches and thus reduce false alarms.

Personalized Communication

Most of the services we use today can be accessed online, from booking a taxi to buying a new dress. However, people are sociable beings and often need to ask for advice or follow guidelines to complete a task. It can be quite redundant to read neat instructions, desperately search for trustworthy reviews or wait while an operator on the phone ends up with another customer before coming to you. Fortunately, a way out is possible with quick learning programming algorithms that allow the creation of a conversational user interface that will respond with patience and education to all of the client's needs.

The North Face

Buying clothes for extreme sports and hiking can be a real challenge. Will this jacket be good for snowboarding at a resort you're heading to next season? Who do you ask for professional advice if you're shopping online?

The North Face seems to understand its customers' problems and has enabled a conversational user interface powered by IBM's Watson for a seamless customization experience. The bot (artificial intelligence chat) asks you when and where you're going and by analyzing weather conditions and landscape peculiarities, it gives you the best articles.

Virtual Sephora Artist

Selecting a makeup product from a wide variety of brands and color palettes is always an experiment that can be very expensive. Fortunately, you can now try the new "Sephora virtual artist" online, to choose a new style and suitable products in an interactive environment.

Sephora, one of the most advanced and growing retailers in the beauty field, strives to better understand its audience with this solution.

The artificial intelligence algorithms underlying the application scan your photo and determine where your eyes, lips, and cheeks are. Some configurations allow you to play with colors and styles, selecting from 20,000 products sold on Sephora.

As a result, more than nine million customers have already tested the "virtual assistant" feature since its launch. Organic revenue in LVMH's selective retail, of which Sephora is a part, increased 9 percent in the first quarter of 2018 thanks to the AI implementation.

Key Points

As you can see, from unification we inevitably run towards personalization in all spheres of our life. For years, we collected data that is now permanently stored in Excel files, CRM files, and database tables, although no one noticed what was happening.

Modern clients tell us, "You know everything about me, why don't you give me what I want? There is no ready-to-use solution in AI.

Any system must be individually adjusted to its details. Leading companies have already plunged into this group of unlimited opportunities to personalize the customer experience in search, pricing, security, and conversation.

By studying the experience of the pioneers who dared to implement AI and ML in their businesses, you can learn how to improve your services.

Show All Articles