How AI Contributes to Business Profitability

🤖 Future of Work: Are Business Leaders Ready To Work With Cyborgs And Centaurs? A recent study with Harvard, Wharton, MIT, Warwick's Business Schools, and Boston Consulting Group (BCG) unveils the remarkable impact of AI, specifically GPT-4, on the world of high-skilled professionals. This study provides a valuable perspective for senior executives exploring the integration of AI into their business strategies. 1️⃣ Remarkable performance uplift for underperformers with AI Assistance: The study's astounding revelation was the dramatic performance improvement among the initially lower-scoring professionals. When these professionals (consultants) were equipped with AI tools, they experienced a whopping 43% jump in their performance. This highlights AI's potential as an incredible workplace equaliser, particularly for those striving to enhance their capabilities and output. It's a testament to how AI can empower every tier of talent within an organisation, boosting overall productivity and the quality of their work products. However, let's note that creating work products is a small part of a consultant's job, and most successful consultants I have met spend most of their time gaining and/or bulldozing buy-in for their ideas/recommendations. 2️⃣ Boost in Productivity & Quality: For tasks well within AI's grasp, the results are impressive – a 12.2% increase in task completion, 25.1% faster execution, and a staggering 40% rise in quality. This isn't just for the tech-savvy; benefits span all skill levels. 3️⃣ A Cautionary Note: When tasks fall outside AI's capabilities, reliance on AI can hinder performance. In a unique twist, the study designed a task to expose AI's blind spots, where consultants who did not have AI to help them excelled with an 84% success rate. However, their accuracy dropped to 60-70% when AI was employed, highlighting the critical need for discerning AI use in complex decision-making. Why did this happen? It's called the 'sleeping at the wheel' effect, where one relies heavily on the AI system to problem-solve and forgets to apply their intuition and knowledge. This underscores the importance of discerning where AI can be valuable and where human expertise is irreplaceable. The only way to know which task is which for your business? You have to dive in, test and learn. 4️⃣ New Collaboration Models: The study highlights two innovative approaches to working with AI: 'Centaurs' - professionals who delineate cleanly between the work that they will do and the work that AI will do, and 'Cyborgs' - who seamlessly inject AI into most of their work processes. Both models open new avenues for business problem-solving. Thanks for the fascinating study, Fabrizio Dell'Acqua, Edward McFowland III, Ethan Mollick, Hila Lifshitz- Assaf, Karim Lakhani, Katherine Kellogg, Saran Rajendran, Lisa Krayer, PhD and François Candelon. 🔗 [Links to the study in comments] #BusinessTransformation #LeadershipInAI #FutureOfWork #Innovation

Yesterday, I helped my client begin to add an extra $80K in profit this year. Growing a business is HARD. With so many moving parts, it's tough to identify and fix problem areas limiting your potential. I recently helped a video production agency owner struggling with this. Despite a talented team and consistent work, he felt trapped working IN his business instead of ON it. There were just too many manual, repetitive tasks that kept him chained to his desk. Here are the 4 steps we took to help him grow faster: 1️⃣ Identified areas of “operational waste”. I had him walk through his entire production workflow from start to finish. It revealed over a dozen clunky processes - from creating GDrive folders to sending Notion/Slack status updates to posting deliverables. These manual tasks created friction that slowed down operations and hindered growth. 2️⃣ Prioritized the biggest pain points. Next, we estimated potential time/money savings for each inefficient process. This made it easy to pick the low-hanging fruit - things like data entry, reporting, and internal notifications. Fixing those would free up 20+ hours/week for higher-level creative work. 3️⃣ Defined the role of his AI-powered “Employee”. Using the insights from 1 and 2, I mapped out every job that his AI-powered helper would have. We gave it job responsibilities like: ✔️ Automating folder creation ✔️ Sending project status notifications ✔️ Posting finished videos to correct channels Screenshot of just some of these workflows below. 4️⃣ Built and launched the AI prototype. I set up AI Automations for the urgent tasks first. We tested them for a week before expanding the AI Employee's duties. After a few rounds of feedback and tuning, our AI helper was fully deployed yesterday. 📌 All of these steps were taken strategically to ensure we were adding the most value possible, in the shortest amount of time. And now, because we took a structured approach, he’s got an AI-powered Employee working 24/7/365 in the background. Helping him grow faster at all times. And adding $80K in profit while we’re at it. Man, I love go-live day. --- Want to install an AI-powered Employee in your business? Click my name above, and DM me “Growth” to learn more about how I can help. 🔝

LLM field note: super interesting new paper shows 25%+ increase in speed and a 40% increase in human-rated performance from AI assistants. Worth reading the whole paper from our friends at Boston Consulting Group (BCG). Two groups, each tackling different kinds of tasks. Results are fascinating. Let's dive in. Task 1: new product ideas and innovation. ⚡️Use of AI significantly improved the quality of the work by more than 40%, depending on the scoring system (consistently). 💫 12.2% increase in task completion rates, and interestingly, the most significant beneficiaries were those who were initially less skilled, improving their performance by 43%, compared to a 17% boost for the more skilled group. 🌟 finished tasks up to 22.5% faster, and the GPT Only group was 27.63% faster than the control. 👾 However, the study also noted that while the use of AI led to higher quality ideas, it also resulted in less diverse or more homogenized responses. This is noteworthy (but not surprising). -- Task 2: problem-solving tasks involving quantitative business data and customer interviews. 📋 Analyze a company’s channel and brand performance. Required nuanced interpretation of spreadsheet data and interviews with company insiders. Provide actionable strategic recommendations to a hypothetical company's CEO. 🦺 The primary evaluation metric was 'correctness.' Human participants outperformed AI, scoring 84.5% in correctness versus 60-70% for AI. ⏱️ AI treatment groups showed a significant reduction in the time taken to complete tasks but at the cost of reduced accuracy. ✍️ Quality of recommendations was also examined, and surprisingly, AI treatments led to a higher quality of strategic recommendations, despite the lower correctness scores. ✨ Implication is that AI could enhance the quality of outputs even if the recommendations were not entirely correct. A lot to think about from this one. Need to noodle some more on the take aways for performance, cautions, and skills - but a common thread is emerging as more of these studies appear... AI is a tool which improves work by turbo charging creative and analytical thought, not replacing it. -- (thanks for sharing Brad Porter and Allie K. Miller - who are both insightful, and posted about this too)

Boston Consulting Group (BCG) consultants completed 12.2% more tasks, 25.1% faster with 40% higher quality when using AI, according to a new report from Ethan Mollick. This is one of the best analyses I've seen on the impact of AI on professional work and insights on how to best collaborate with AI. In addition to the improved performance with AI, Ethan shared: ► AI improved lower performing workers by 43% more than higher performers in the BCG experiment, reducing skill gaps between employees. But over-relying can make people "fall asleep at the wheel" and miss AI mistakes. Staying alert is key. ► There is an unpredictable "jagged frontier" to what AI can and can't do well. Knowing where AI excels and falls short is crucial. ► To best collaborate with AI, be "Centaurs" to strategically divide work or "Cyborgs" to closely intertwine work with AI. This combines the benefits of both humans and AI. Ethan's paper provides valuable insights into effectively leveraging AI to enhance productivity and performance. I highly recommend reading the report (link in comments) to learn more about optimizing human-AI collaboration. What has been your experience working with #AI so far? I'd love to hear your thoughts in the comments! #FutureOfWork #AIAdoption #AIProductivity #WorkforceProductivity #WorkPerformance

🚀 𝐀𝐈 𝐢𝐧 𝐂𝐮𝐬𝐭𝐨𝐦𝐞𝐫 𝐒𝐮𝐜𝐜𝐞𝐬𝐬: 𝐅𝐫𝐨𝐦 𝐁𝐮𝐬𝐢𝐧𝐞𝐬𝐬 𝐂𝐚𝐬𝐞 𝐭𝐨 𝐁𝐮𝐬𝐢𝐧𝐞𝐬𝐬 𝐈𝐦𝐩𝐞𝐫𝐚𝐭𝐢𝐯𝐞 At a recent FunnelStory 𝐂𝐒 𝐋𝐞𝐚𝐝𝐞𝐫𝐬𝐡𝐢𝐩 𝐃𝐢𝐧𝐧𝐞𝐫 in downtown LA, one thing became crystal clear: forward-thinking CS leaders aren’t just talking about AI — they’re securing budgets and driving real impact with it. Here are some standout insights from the discussion: ✅ 𝐀𝐈-𝐃𝐫𝐢𝐯𝐞𝐧 𝐂𝐨𝐬𝐭 𝐀𝐯𝐨𝐢𝐝𝐚𝐧𝐜𝐞: When AI prevents costly hires or slashes manual workloads, budgets get approved fast. One leader shared how their internal tool, “Data Whisperer,” eliminated the need for multiple data analysts — a huge win for both efficiency and ROI. ⚡ 𝐄𝐟𝐟𝐢𝐜𝐢𝐞𝐧𝐜𝐲 𝐆𝐚𝐢𝐧𝐬 𝐓𝐡𝐚𝐭 𝐌𝐚𝐭𝐭𝐞𝐫: Weekly reports that used to take 3 hours? Now done in 5 minutes. RMA analysis? Streamlined. AI is freeing CS teams to focus on strategy, not spreadsheets. 🗣️ 𝐕𝐨𝐢𝐜𝐞 𝐀𝐈 𝐰𝐢𝐭𝐡 𝐄𝐦𝐩𝐚𝐭𝐡𝐲: AI isn’t just saving time — it's creating better experiences. From 24/7 sales support to empathetic voice agents assisting patients with chronic diseases, AI is enhancing human connection at scale. 📈 𝐏𝐫𝐨𝐟𝐢𝐜𝐢𝐞𝐧𝐜𝐲 𝐌𝐮𝐥𝐭𝐢𝐩𝐥𝐢𝐞𝐫: With AI, junior team members are performing like seasoned pros. It’s leveling the playing field and dramatically speeding up onboarding. 🔍 𝐂𝐨𝐧𝐭𝐞𝐱𝐭 𝐢𝐬 𝐊𝐢𝐧𝐠: To get meaningful results, LLMs need a robust context layer — business-specific data and relationships that make outputs accurate and actionable. 🧠 𝐑𝐞𝐭𝐡𝐢𝐧𝐤𝐢𝐧𝐠 𝐔𝐈: The future isn’t just prompts and chat windows. We’re seeing new AI interfaces deliver exactly the right data, right when leaders need it. 💡 The "aha moment" around AI adoption looks different for every organization, but the results are consistent: more efficiency, more insight, and a leap in capability. Jim Goldfinger Sofia Kiriukhina David Hayes Matt Collier Cesar Sanchez Brendan Bencharit Ram Shenoy Adnan M. Arun Balakrishnan Preetam Jinka #FunnelStory #CustomerSuccess #AI #Leadership #Innovation #VoiceAI #Efficiency #Proficiency #BusinessImperative #CSLeaders #CostAvoidance

There is a conversation happening in every boardroom: 'how can gen AI help us become more efficient and productive, and therefore profitable?' This naturally leads to a question about the biggest cost in many organizations: wages. Let me start with the point that the purpose and value of gen AI is not just about driving down costs or ‘doing more with less’ - but the question still has merit. We have seen research showing that a call center with over 5,000 agents using a generative AI-based assistant saw a productivity improvement of 14%, on average. Interestingly, the improvements were more noticeable with less experienced staff, who saw productivity gains of 35%. The theory being that the more experienced call center agents already have the required knowledge. Does this mean that gen AI has more value helping lower-paid or less experienced workers? Furthermore, will it drive down wages? We have to be careful leaping to conclusions. Erik Brynjolfsson, who co-authored the paper about this example, speaks of the 'Turing Trap'. The idea that when AI is used to replicate the tasks humans do, it drives down costs. When it augments humans, it makes them more valuable and drives up wages. So, perhaps we need to see this as a balance of forces: to imitate, because of the benefits to efficiency and costs; and to complement, to drive innovation and productivity. I have seen both Gianni Giacomelli and Rodrigo Madanes post about these very topics in the past week, so welcome your comments - and anyone else who has experience or opinions to share. 🔗 Links to posts and articles in the comments below.