Acuto / Case Studies / AI Call Analysis
Enhancing Call Tracking Data With AI

Combining AI with Call Intelligence for Deeper Insights

We helped revolutionize Marketing32‘s call tracking by leveraging OpenAI’s API for intelligent call analysis, leading to precise categorization and better data-driven decisions.

10,000

Calls

Analyzed with AI every single month

50x

Cost Efficiency

Compared to human labellers

80%

Accuracy

Compared to human labellers

Acuto has been a game-changer for us. Among other projects, setting up AI-driven call labeling transformed our insights, allowing for scale that previously would have cost 50x what it does now. We’re thrilled with the results — it’s efficient, accurate, and has genuinely boosted our ability to make smarter marketing decisions. 
marketing automation ai team
Jeff Kingston

Co-Founder & COO at Marketing32

How Marketing32 Worked With Acuto

The Challenge:

Marketing32 needed to go beyond merely counting calls as conversions. While many agencies track any call as a conversion, the nuance between a cancellation call and a new booking call is significant. Simply tallying calls falls short of a data-driven approach, offering limited insights into campaign effectiveness. 

To truly understand and optimize their campaigns, Marketing32 required a solution that could discern the nature of each call, enabling them to measure meaningful conversions accurately.

Our Solution:

Acuto tackled this by utilizing the OpenAI API to analyze call transcripts. By feeding the call content into ChatGPT, Acuto refined the prompt to specifically identify new users, the most valuable conversions for Marketing32. 

The data was then integrated into a BigQuery warehouse, and a custom dashboard was developed to monitor accuracy. This setup allowed for a comparison between the AI’s labeling and the human team’s previous efforts, ensuring a high standard of accuracy and reliability.

How We Did It

The Results

Initial promising results led Acuto to further train the model with additional data and adjusted parameters, enhancing accuracy. The model excelled in correctly identifying call labels, with minimal false negatives. While false positives were a greater challenge, accuracy remained above 70%. 

Additionally, Acuto leveraged IBM Watson for natural language processing (NLP), extracting sentiment and emotion from the calls. This dual approach not only categorized calls more precisely but also provided deeper insights into customer interactions.

The enhanced call intelligence significantly improved conversion tracking by categorizing calls based on content. This allowed Marketing32 to distinguish between new bookings and cancellations, optimizing their marketing strategies. 

The integration of data into a BigQuery warehouse and the development of a real-time dashboard facilitated advanced analytics and data-driven decisions. The accuracy in call labeling saw substantial improvement, benchmarking AI performance against human standards. 

Sentiment and emotion analysis via IBM Watson provided additional insights into customer interactions, contributing to continuous model improvement and operational scaling. This automation reduced manual labor, freeing up resources for more strategic tasks and enabling personalized marketing strategies through comprehensive customer insights.

Facing Similar Challenges in Your Agency?

If your agency is grappling with accurately tracking call conversions and optimizing marketing efforts, Acuto offers a custom AI-driven solution tailored to your needs. 

Discover how our advanced call intelligence can transform your data into actionable insights. Contact Acuto today to achieve unparalleled accuracy and efficiency in your marketing strategies.

Book a Meeting With a Marketing Automation Expert