BCG suggests doing this to get value from AI

FIRST PASS

BCG highlights the great AI divide; provides a blueprint for companies to get value from AI (BCG)

BCG’s latest research on AI adoption reveals a stark reality: while 98% of companies are experimenting with AI, only 26% have developed the capabilities necessary to move beyond proofs of concept and generate actual value.

Even more striking, just 4% are at the forefront of AI innovation, systematically building and scaling cutting-edge AI capabilities across their organizations.

The research, which surveyed over 1,000 companies worldwide, found that these AI leaders are significantly outperforming their peers in revenue growth, shareholder returns, and returns on invested capital.

Key points:

1. More than half of top performers are traditional incumbents, not tech companies

2. 62% of AI value comes from core business processes, not support functions

3. Leaders focus on fewer, high-impact projects rather than spreading resources thin

4. The key challenge isn’t technical – 70% of implementation challenges relate to people and process

Merlin’s Notes: The report challenges the common narrative that digital natives and tech companies are the only ones winning with AI. With more than half of the top-performing companies being traditional incumbents, it suggests that any company, regardless of industry or starting point, can become an AI leader if it builds the right capabilities and focuses on the right opportunities.

What sets the leaders apart? BCG identified six key differentiating characteristics:

1. They focus on core business processes rather than just support functions

2. They're more ambitious, with expectations for revenue growth from AI by 2027 that are 60% higher than other companies

3. They invest strategically in fewer, high-priority opportunities rather than spreading resources thin

4. They integrate AI in both cost-reduction and revenue-generation efforts

5. They follow the 70-20-10 rule: 70% of resources into people and processes, 20% into technology and data, and 10% into algorithms

6. They've moved quickly to adopt GenAI alongside predictive AI

The sectors with the highest percentage of AI leaders are those that experienced digital disruption earliest: fintech (49% are leaders), software (46%), and banking (35%). However, the impact extends across all industries, suggesting opportunities for every sector to create value through AI adoption.

BCG provides a 7-step AI value playbook for any company:

1. Start with strong leadership commitment

2. Build a balanced portfolio of initiatives

3. Focus on lighthouse programs (high ROI, easy to implement)

4. Ensure basic infrastructure

5. Address capability gaps

6. Implement end-to-end governance

7. Establish responsible AI guardrails

HIGHLIGHT

Microsoft shares 200+ Real-world AI Success Stories (Microsoft)

In a comprehensive post released last week, Microsoft's Chief Marketing Officer, Alysa Taylor, shared stories from over 200 organizations leveraging AI to transform their operations. The study, conducted with IDC, revealed that companies are seeing an average return of $3.70 for every $1 invested in GenAI.

The stories broadly fell into four key categories:

1. Employee Experience Enhancements

Some standout examples:

• Access Holdings: Reduced coding time from 8 hours to 2 hours, and slashed chatbot development time from 3 months to just 10 days using Microsoft 365 Copilot

• Honeywell: Employees are saving 92 minutes per week (74 hours annually)

• Softchoice: Achieved 97% reduction in time spent summarizing technical meetings

2. Customer Engagement

Notable implementations:

• Telkomsel: Increased customer self-service interactions from 19% to 45%, reducing daily call volume from 8,000 to 1,000

• Pacific Gas & Electric: Built a chatbot using Microsoft Copilot Studio that saves $1.1 million annually on helpdesk support

• Vodafone Group: Achieved 70% resolution rate for customer inquiries through digital channels and reduced call times by at least one minute

3. Business Process Transformation

Examples:

• Cineplex: Saved over 30,000 hours annually in manual processing time through innovative automation solutions

• Qatar Charity: Reduced call center handling time by 30% while increasing customer satisfaction by 25%

• Saphyre: Reduced manual efforts by 75% in financial trading workflows using Azure and AI

4. Innovation Acceleration

Breakthrough examples:

• Brembo: Reduced development time for new brake pad compounds from days to minutes using Azure OpenAI

• Pacific Northwest National Laboratory: Discovered new battery materials in weeks instead of years

• Khan Academy: Partnered with Microsoft to bring AI tools to millions of educators

The impact spans across industries, from manufacturing and healthcare to financial services and education. What's particularly noteworthy is how AI is not simply optimizing existing processes but enabling entirely new possibilities.

The variety of use cases demonstrates that AI's impact isn't limited to any particular sector or size of organization – from multinational corporations to local government bodies, organizations are finding innovative ways to leverage AI for tangible business results.

JARGON BUSTER

USE CASE SPOTLIGHT

Carlyle-backed Acentra Health used AI to save 11,000 hours and $800K

Acentra Health, a healthcare solutions provider backed by The Carlyle Group, partnered with Microsoft to develop MedScribe, a GenAI application that revolutionized their Medicare appeals process. The goal? To free up valuable nursing time while maintaining high-quality, empathetic communication with patients and providers.

In just six months, their team created an innovative solution that exceeded expectations, transforming a time-consuming manual process into a streamlined, AI-assisted workflow.

The impact was dramatic and measurable:

1. 50% reduction in letter writing time (from 6 to 3 minutes per letter)

2. 11,000 nursing hours saved

3. Nearly $800,000 in cost savings

4. 99% approval rate for AI-generated letters

5. 2x increase in daily letter processing capacity (from 12-14 to 20-30 letters per nurse)

The Challenge

Acentra Health's nurses faced a demanding task: translating complex medical terminology from physician notes into clear, empathetic language for Medicare appeal responses - all while meeting strict 24-hour deadlines.

Each letter needed to be personalized and written at an appropriate reading level for either beneficiaries or healthcare providers.

The Solution

Working closely with their clinical teams, Acentra Health developed MedScribe, a web application powered by Azure OpenAI Service. The system was carefully designed to:

• Generate initial draft letters based on physician notes

• Automatically adjust reading levels and tone for different audiences

• Maintain HIPAA compliance within a secure environment

• Allow for nurse review and feedback through a simple thumbs-up/thumbs-down system

Key Success Factors

1. Strong Clinical Partnership: The development team worked hand-in-hand with nurses to refine the system's prompts and outputs

2. Rapid Development: From concept to full deployment in just 6 months

3. Focus on User Experience: Simple interface with immediate output and easy review process

4. Continuous Monitoring: Power BI dashboards track metrics and nurse feedback to ensure quality

The success has been so significant that Acentra Health is already expanding MedScribe to their quality-of-care division for handling more complex correspondence.

As Rosemarie Hartnett, Director of Operations, puts it: "The nurses constantly tell us that of all the improvements we've attempted over the years, MedScribe has probably been the most impactful."

It's a clear demonstration of how thoughtfully implemented AI can enhance healthcare operations while maintaining the human touch that's so crucial in medical communication.

TOOL OF THE WEEK

Zapier

Zapier, the veteran automation platform, has undergone a significant AI transformation that's worth noting. At their recent ZapConnect event, they unveiled a suite of AI-powered features that position them as more than just a workflow automation tool.

Key AI updates include:

• AI by Zapier: Built-in GPT-4 capabilities that can be integrated into workflows without requiring a separate API key. This allows for automated content generation, data extraction, and analysis within Zaps.

• Copilot: An AI assistant that helps build workflows by translating natural language descriptions into actual Zaps - handling everything from account connections to data mapping and testing.

• AI-Powered Chatbots: Custom chatbots that can be trained on your website or help center content, with the ability to automatically update based on schedule.

• Zapier Central: A personal AI assistant that works across 7,000+ apps, capable of analyzing data, drafting emails, and pulling reports autonomously.