The Impact of UX Writing on Conversion through A

Active vs. Passive Voice

Active voice makes the text more dynamic and understandable. For example, compare these two phrases:

The second phrase sounds more personal and inspires more trust. The active voice gives the feeling that there is a specific person behind the actions, not a faceless algorithm.

Use simple and clear language

Example:

Сложно: Для того чтобы начать работу с приложением, вам потребуется выполнить ряд предварительных настроек.
Просто: Чтобы начать, выполните несколько простых шагов настройки.

A person who encounters your product for the first time wants to understand it as quickly as possible. Use short, clear sentences without complex phrases. Simplicity is the key to effectiveness.

Brevity and visual separation

Break up your text into short paragraphs. No one wants to read huge blocks of text, especially on a screen. Visually highlight key points, use bullet points, and use headings.

Example:

Сложно: Мы предлагаем множество услуг, включая разработку приложений, анализ данных, оптимизацию бизнес-процессов и поддержку пользователей.
Просто: Наши услуги:
- Разработка приложений
- Анализ данных
- Оптимизация бизнес-процессов
- Поддержка пользователей

A clear structure makes it easier to understand and helps the user quickly find the information they need.

Examples of how small changes in syntax can have a big impact on conversions

Buy Now vs. Place Order Button:

<!-- Кнопка 1 -->
<button>Купить сейчас</button>

<!-- Кнопка 2 -->
<button>Оформить заказ</button>

An A/B test on a major e-commerce site found that replacing the phrase “Buy now» to «Place an order” increased conversion by 17%. Users perceived “Place an order” as a softer and less aggressive proposal, which lowered the psychological barrier.

Registration form: “Create an account” vs. “Register”:

<!-- Заголовок 1 -->
<h2>Регистрация</h2>

<!-- Заголовок 2 -->
<h2>Создайте аккаунт</h2>

In another test, replacing the word “Registration” to a more personalized “Create an account” resulted in a 22% increase in registrations. “Create an account” implies more active and personal participation of the user, which turned out to be more attractive.

Clarity and straightforwardness:

Сложно: Наше приложение предоставляет полный набор инструментов для оптимизации вашего рабочего процесса.
Просто: Наше приложение помогает вам работать быстрее и проще.

In one test, simple language and straightforwardness increased app installs by 12%. Clear and specific promises are perceived better and inspire trust.

Conducting A/B testing for UX Writing

Defining the purpose of testing and hypotheses

The first step is to formulate a specific business goal. For example, you need to increase conversion on the registration page. This is a general goal, and the hypothesis might sound like this:Replacing a technical term with a friendlier expression will increase the number of completed registrations by 15%“.

On one of the projects I worked on, we were facing a problem with low registration completion on a B2B platform. The hypothesis was that replacing the technical term “Client number» to «Your ID in the system“will allow users to better understand what is required of them.

Preparing text variations

Creating text variations should be based on clear data about users and their behavior. It is important not to just change words at random, but to rely on real data.

Example:
1. Option A: “Please enter your customer number to complete registration.»
2. Option B: “Please enter your ID in the system to complete registration.»
3. Option C: “Please enter your ID to continue.»

For each option, you need to think about tone, style, and formatting. For example, it may be better to use the second person in the text than impersonal formulations.

Integration and technical implementation

Launching A/B testing requires proper integration of testing tools, such as Google Optimize, Optimizely, or your own solutions, if available.

Example with React:

import React, { useState, useEffect } from 'react';

function RegistrationForm() {
    const [variant, setVariant] = useState('A');

    useEffect(() => {
        // подключение Google Optimize или вашего кастомного решения
        window.gtag('event', 'optimize.callback', {
            callback: (variantId) => {
                setVariant(variantId);
            }
        });
    }, []);

    return (
        <div>
            <h2>Регистрация</h2>
            <label>
                {variant === 'A' && "Введите ваш номер клиента для завершения регистрации."}
                {variant === 'B' && "Введите ваш ID в системе для завершения регистрации."}
                {variant === 'C' && "Укажите ваш идентификатор для продолжения."}
            </label>
            <input type="text" />
        </div>
    );
}

This allows text content to be dynamically changed depending on the variation displayed to the user. It is important that these changes are made quickly and without noticeable delays.

Selecting and monitoring metrics

Metrics should be configured to minimize impact on system performance. The choice of metrics depends on the goals, but some of the key indicators are:

Conversion Rate: measuring what percentage of users completed the target action (registration, purchase).
Form Completion Rate: A form-specific metric that measures what percentage of users who started filling out a form completed it.
Time for First Interaction: how quickly a user interacts with text or an interface after the page has loaded.

Example of monitoring with Google Analytics:

ga('send', 'event', {
    eventCategory: 'RegistrationForm',
    eventAction: 'Completion',
    eventLabel: variant,
    eventValue: 1
});

Completing the test and analyzing the results

It is important not only to collect the data, but also to interpret it correctly. For this, statistical analysis is often used, taking into account the p-value to ensure that the test results are not due to chance.

For example, in one project, the analysis showed that Variant B (the text “Enter your ID in the system”) actually increased the number of completed registrations by 18%, which confirmed the hypothesis. The data analysis also revealed that users over 40 responded significantly better to this version of the text, which influenced further changes to the texts in other sections of the site.

Approaches to audience segmentation and analysis of test results

Audience segmentation is the key to A/B testing. Proper segmentation allows you to understand how different groups of users react to changes in text.

1. Segmentation by demographics

It is important to consider age, gender, location and other demographic parameters. For example, a text that is well received by young people may be incomprehensible to older generations.

2. Segmentation by behavior

Segmentation based on user behavior can yield valuable insights. For example, new users may need more detailed and understandable texts, while regular users will appreciate brevity.

3. Segmentation by devices

It is equally important to consider the device type: mobile users tend to prefer short and concise texts, while desktop users may spend more time reading.

On one project, we A/B tested the texts for a newsletter sign-up page. For mobile users, we used the headline variation “Get updates“, and for desktop – “Subscribe to our newsletter and receive fresh news every week” As a result, the mobile version with a shorter headline increased conversion by 18%, while the desktop version showed a 12% better result due to a more complete description. These data led to the development of a strategy in which text elements are adapted to different devices, taking into account the behavioral characteristics of the audience.

Summing up

As it turns out, text is not just a set of words, but a tool for influencing the user. Here are some key findings worth taking into account:

Text matters, even if it is short. No, this is not an excuse for lazy copywriters! Even one word can have a significant impact on conversion..
Adaptive texts are our everything. Users of different devices perceive information differently, and this should not be ignored.
Hypotheses are not always confirmed, but that's okay. Yes, tests may show that your brilliant, perfectly crafted copy is a complete fiasco. But that’s what A/B testing is all about: learning from mistakes and adapting quickly.
Metrics are your best friends.

Don't be afraid to test, don't be shy about changing the wording, and remember: the right words can work wonders, especially if they are tested correctly. And who knows, maybe the next A/B test will lead you to creating the most converting text in the history of your project.