Case study

We have created a dedicated AI assistant for Brand24 to support all important metrics and analytics, which are important for a SaaS company.

Offices:

Poland / USA

Company size:

51 - 200 employees

Industry:

Marketing

Revenue:

SaaS

For Brand24, the primary objective was to streamline the extraction and analysis of its internal data, with a particular focus on sales metrics.

Brand24 is the #1 online brand monitoring tool in the world. The world’s largest corporations, prominent public institutions, and even the White House rely on it.

Bards.ai has designed and built a dedicated AI assistant for Brand24 to support in-house data analytics

Brand24 is an organization built from the ground up with a data-driven approach. However, having vast amounts of data also presents the challenge of interpreting it effectively. For Brand24 executives, the most critical aspects of data are time, accuracy, and availability. In a globally operating product company, making metrics-based decisions is a fundamental operational practice.

An AI assistant delivers instant insights and data through clear text, detailed tables, and insightful charts—making analysis faster and smarter.

Focused on business side

Solution

  1. Our solution is built on a specialized analytical pipeline powered by LLM, designed to extract the necessary information, evaluate and compare it, iterate when needed, and ultimately deliver precise, well-structured answers to user queries.

  2. AI assistant keeps track of the conversation context from start to finish. To make things even easier, it’s fully integrated as a Slack bot, so no extra apps are needed. This way, Brand24 could keep their existing workflow without adding another tool to their tech stack.

Hittin' the road

Implementation roadmap

The project was divided into several phases:

Phase 1 - Week 1

An early prototype has been developed based on the data provided by Brand24, utilizing 1-2 data sources. This has provided an initial preview of how integration with the potential final product has taken shape.

Phase 1 - Week 1

An early prototype has been developed based on the data provided by Brand24, utilizing 1-2 data sources. This has provided an initial preview of how integration with the potential final product has taken shape.

Phase 1 - Week 1

An early prototype has been developed based on the data provided by Brand24, utilizing 1-2 data sources. This has provided an initial preview of how integration with the potential final product has taken shape.

Phase 2 - MVP - Weeks 2-4

The first iteration of the BOT has been created, supporting up to 8 data sources. This version has enabled real-world testing, allowing us to assess how effectively it has handled the most frequently asked questions from actual end users.

Phase 2 - MVP - Weeks 2-4

The first iteration of the BOT has been created, supporting up to 8 data sources. This version has enabled real-world testing, allowing us to assess how effectively it has handled the most frequently asked questions from actual end users.

Phase 2 - MVP - Weeks 2-4

The first iteration of the BOT has been created, supporting up to 8 data sources. This version has enabled real-world testing, allowing us to assess how effectively it has handled the most frequently asked questions from actual end users.

Phase 3 - Weeks 5-6

Once the MVP has been launched, an evaluation phase has taken place in a production environment. During this period, the Brand24 team has needed to be actively involved. This phase has served as the first real-world test of the solution, where we have analyzed its performance and have adjusted the pipeline as needed to better meet actual demand.

Phase 3 - Weeks 5-6

Once the MVP has been launched, an evaluation phase has taken place in a production environment. During this period, the Brand24 team has needed to be actively involved. This phase has served as the first real-world test of the solution, where we have analyzed its performance and have adjusted the pipeline as needed to better meet actual demand.

Phase 3 - Weeks 5-6

Once the MVP has been launched, an evaluation phase has taken place in a production environment. During this period, the Brand24 team has needed to be actively involved. This phase has served as the first real-world test of the solution, where we have analyzed its performance and have adjusted the pipeline as needed to better meet actual demand.

Brand24 loves it!

Here's a list of example questions that AI assistant can answer:

Present the total MRR results for the last 12 full months in a table. How much MRR did we lose in January 2024 due to churn?

Provide a comparison of July vs. June in terms of trial numbers, sales, MRR, and churn. Plot the number of trials, sales, and conversion rates for 2023.

In which countries did trials drop the most in April 2023? Analyze only countries that had at least 500 trials in the previous period.

Partnering with Bards.ai to implement a customized LLM model was a game-changer!

By leveraging a dedicated AI solution that pulls knowledge directly from our key resources, Brand24 now has a single source of truth. This has significantly streamlined workflows, saving time and enhancing collaboration by eliminating the need to sift through multiple documents. Bards.ai has shown remarkable flexibility and a deep understanding of our needs, delivering a solution perfectly aligned with our processes.

Michal Sadowski

CEO @ Brand24

Looking to integrate AI into your product or project?

Looking to integrate AI into your product or project?

Looking to integrate AI into your product or project?

Get a Free consultation with our AI experts.

Get a Free consultation with our AI experts.

Get a Free consultation with our AI experts.

Book a free consultation

Book a free consultation

Book a free consultation