MacPaw: image 1


MacPaw, a software company founded in Ukraine, has been dedicated to developing and distributing software for macOS and iOS since 2008. The company is behind CleanMyMac X, Setapp, ClearVPN, and several other well-known products. In 2017, MacPaw acquired The Unarchiver and has been actively supporting the program ever since.

Today, MacPaw boasts over 30 million users worldwide, with their programs installed on a fifth of Macs across the globe.

One of the main priorities for MacPaw is the digital safety of their users, team, and organization. Their mission is to help machines help people.

The Challenge

The growing number of users has led to the need to improve the current customer support service. MacPaw acknowledged the necessity for a bot solution as the volume of сontacts surged. The company tried to launch its own simple chatbot, based on the “if/else system,” but it was not enough and didn’t cover all support service needs.

So, they wanted a completely new approach and asked for a solution to minimize human involvement in daily customer sessions. Given the specifics, customers need clear and structured answers with step-by-step instructions that will help to solve the difficulties they might face.

The chatbot would need to pick up the conversation after a few seconds and provide a correct, relevant answer using the customer knowledge base. The combination of efficiency and continuous learning that MetaDialog’s chatbot provides, promised a solution that could meet MacPaw’s customer service standards.

Most Suitable Conversational Bot

Here’s how we tackled this task to make MetaDialog the most suitable conversational bot for client purposes:

1. We taught the bot to provide straightforward and clear answers to customer questions. To achieve this, we began by feeding in our conversational chatbot data, derived from MacPaw’s interactions with customers.

2. We then proceeded to test the bot’s capabilities. The goal was to throw all the frequent questions that MacPaw’s consumers typically ask at the bot, and find any weaknesses. Whenever a problem emerged, we established new rules for AI to fix the issue.

MacPaw: image 3
MacPaw: image 4

3. Afterwards, we ran internal tests with the MacPaw team to demonstrate the results, gather their feedback, and identify any additional features desired.

4. We then released the chatbot and observed how it interacted with real customers to improve patterns of AI-human interaction. You never see how creative customers can be with their questions until you see the bot acting in real situations.

5. Within just two weeks of preparation, testing, and improvement, the bot was fully trained and implemented into MacPaw’s workflow.

Integration Team

To optimize the work speed, we initially assembled a lean team of four experts – two developers, a QA tester, and a team lead. As we progressed, we expanded our team by two more developers, enhancing productivity without adding complexity. This efficient team composition enabled us to successfully complete the project within just 2 weeks.

MacPaw: image 5

Check Out Our Pricing Packages



Response Interval

The response interval to users’ inquiries dramatically reduced by 5 times — from 50 seconds to 8 seconds.

Satisfied Users

60%+ of users were satisfied with the bot and agreed that they would happily refuse human help. Others liked the responses but still preferred to interact with real people, so they continued to communicate with live agents.

Chatbot Helped

The chatbot helped the customer service team by handling easy questions, letting the team concentrate on solving harder problems.

Save 87% of your customer support costs in 1 hour with MetaDialog

Automate 87% of Your Customer Support