Claudia Ring’s Post

The 1️⃣ place where AI truly stands out isn't in content creation. It's in data analysis. We've all seen the hype around AI being the new way to write your LinkedIn posts or personalise your emails. But, if we're being realistic: - AI is AWFUL at writing. - Everybody knows this. So, I started looking at new cases for the use of AI and what opened my eyes was the insights Kevin Gesellensetter brought to me on a call we had. And that insight was AI is AMAZING at Data Analysis. Your business is full of data about - Who responds to your emails - Who buys your products - What subject lines work - How much X contributes to your revenue Why not let AI look at all your data and extract those trends? Then, you can view the report and make informed forecasting decisions, without having to spend hours on end analysing the data yourself. ----------- If you think your data isn't being used to its fullest extent, drop me Devrim Celikay or Kevin Gesellensetter a DM. 📸 of a Go Karting trophy I won yesterday. Started 15th on the Grid too 🪖

AI is great and Tableau Pulse will help answer lots of questions. But it doesn’t solve the fundamental problems of user engagement and a proper data platform. AI is the same as any other tool...garbage in, garbage out. But when you get that right, Pulse will be a game changer. TIP: Pulse helps you know which metrics are most important to your business because people subscribe to them.

The new GPT-4o mini announcement has gotten me thinking about how such an extreme cost decrease can affect AI written cold emails. Of course, AI written lines are going to come down in cost and that's great. (Don't use AI to write your whole email if you haven't been aware). With such a big cost decrease, I think the next phase of AI in cold email will be to double check companies for the correct fit line by line before reaching out. Traditional methods of filtering don't account for incorrect self reported data. For example, many data providers report the industry from the company linkedin profile as their Industry in their database. You know what every 50% or more of software companies reports their industry as? The industry they sell into. If you look up companies in a Banking industry, you might still get consultants for banks etc. This can be cleared up with tools like Clay by setting up if/then type workflows or data providers that have manually reviewed data and industry tagging as well as better AI analysis and keyword research like Ocean.io, ZoomInfo, Owler - A Meltwater Offering etc. But then this leaves edge cases. Companies that are tough to filter by if/then statements and that are so niche, the standard data providers haven't made filters for them. Lists like Construction companies with large fleets Gaming companies with games on every platform Gyms that only do Yoga training etc With the models getting cheaper, it's going to be extremely cost effective to build a prompt that checks each company for true fit before you reach out and then you can even send a batch API to get the cost down 50%. There can be a review of everything company, every contact, every email before it gets sent out and it would still be cheaper than hiring full time people. The fact that Open AI keeps dropping their prices makes the usage so exciting.

🤖 "Certainly! Here is your answer..." If you've worked with AI responses, this might look familiar. ... When building AI automations, these polite introductions can actually break your entire flow. Consider this: You're building an AI-powered sales call analyzer that transcribes calls and extracts key data. One crucial data point is the prospect's budget. The AI correctly outputs: "The prospect mentioned a budget of $40,000 for Q1" But your CRM needs a clean numerical value to calculate forecasts. This is where Make's "Transform Text to Structured Data" module by OpenAI becomes essential. This powerful module takes your AI's conversational outputs and transforms them into clean, structured data you can actually use in your automations. This module helps you: 🎯 Finally get clean, usable data from AI responses. 💫 Stop wasting time cleaning up AI's polite fluff. 📊 Push exact values to your tools automatically. ✅ Get consistent "Yes/No" instead of lengthy explanations. 🔄 Trust your data flow with reliable formatting. Let's take my previous AI agent, the "Food Analyzer" as an example (If you missed it, check the first comment): Raw AI output: "In the attached image there is 150 grams of chicken breast and 200 grams of rice". Amazing output! But you need it structured for the next step in the automation as follows: "ingredients": [ {"name": "chicken breast", "weight": 150}, {"name": "rice", "weight": 200} ] How to implement: 1. Add the "Transform text to structured data" module after your AI response. 2. Choose the model (GPT-4o-mini in most cases). 3. Enter the prompt to describe your desired output. ↳ Be specific about format and requirements. 4. Define your exact parameters: ↳ Variable names (e.g., "budget", "status"). ↳ Data types (number, text, boolean, etc.) ↳ Add descriptions to guide the model's output. Building something interesting with this? Share your next use case below 👇

Is Outlinear AI legit? A Critical Look at its Business Intelligence Capabilities In the dynamic landscape of big data, businesses crave tools to unlock the hidden gems within their information. Outlier AI frequently surfaces as a potential solution, promising to revolutionize decision-making through its business intelligence platform. However, for discerning enterprises, a vital question lingers: is Outlier AI a legitimate and effective tool? Outlier AI: Demystifying the Platform Outlier AI positions itself as an AI-powered platform designed to streamline complex data analysis. Its core functionality lies in automating tasks like data visualization and uncovering critical trends that businesses can leverage to gain a strategic advantage. The Trust Factor: Navigating AI in Today's Business World A recent study by PwC ([link to PWC study on AI decision-making]) revealed that over 70% of executives believe AI will significantly impact their decision-making capabilities. However, concerns regarding potential bias within AI algorithms and data security remain a significant hurdle. Outlier AI Through the User Lens: Insights from Reviews Outlier AI boasts a user satisfaction rating of 84% on platforms like Capterra ([link to Capterra Outlier AI Reviews]). Users commend its intuitive interface and its ability to generate actionable insights. However, some reviewers mentioned encountering an overwhelming amount of information and a potentially complex onboarding process. Is Outlier AI the Perfect Fit for Your Organization? Essential Considerations Before integrating Outlier AI into your business operations, here are some key factors to consider: Company Size: Outlier AI appears to cater more towards larger enterprises with hundreds or thousands of employees. Technical Expertise: While user-friendly, some reviewers suggest that a base level of technical knowledge can be beneficial for optimal utilization. Budgetary Constraints: Outlier AI offers a free version, but its full suite of functionalities likely comes with a price tag. Beyond Legitimacy: Exploring the Competitive Landscape While Outlier AI appears to be a credible business intelligence tool with a strong user base, it's crucial to explore alternative solutions like Tableau or Microsoft Power BI to ensure you find the best fit for your specific needs. If you want to dig deeper, check out that [https://lnkd.in/dEc8yyDT]

𝐆𝐞𝐧𝐀𝐈 𝐯𝐬. 𝐂𝐥𝐚𝐬𝐬𝐢𝐜𝐚𝐥 𝐌𝐋: 𝐖𝐡𝐚𝐭'𝐬 𝐁𝐞𝐬𝐭 𝐟𝐨𝐫 𝐁𝐞𝐠𝐢𝐧𝐧𝐞𝐫𝐬? There's so much buzz about GenAI these days. I mean, everyone's talking about Large Language Models (LLMs) and Stable Diffusion like they're the only thing in AI. If you're new to the scene, you might think that GenAI is the end-all-be-all, and in the process, totally overlook traditional machine learning (ML). But, is GenAI really the answer to every business problem? I had a deep conversation about this with the OG , Prabakaran Chandran and I think it's time we talk about it. First off, let's get something straight. GenAI is fantastic for what it's designed to do—solve fundamental problems that resembles the perception and reasoning abilities of human beings. It's not built for everything, though. There are a lot of areas where classical ML still shines, like recommendation systems, marketing analytics, customer segmentation, and forecasting. So, if you're thinking about where to start, don't dismiss traditional ML too quickly. 𝐒𝐨, 𝐒𝐡𝐨𝐮𝐥𝐝 𝐘𝐨𝐮 𝐅𝐨𝐜𝐮𝐬 𝐨𝐧 𝐆𝐞𝐧𝐀𝐈 𝐨𝐫 𝐓𝐫𝐚𝐝𝐢𝐭𝐢𝐨𝐧𝐚𝐥 𝐌𝐋? Here's the thing: it depends on what you're interested in and what problems you want to solve. If you're into generating stuff, like content or automated tasks, then GenAI is your friend. But if you're more about predictive analytics, customer behavior, or recommendation engines, then traditional ML is where you want to be. 𝐅𝐢𝐧𝐝𝐢𝐧𝐠 𝐘𝐨𝐮𝐫 𝐒𝐭𝐚𝐫𝐭𝐢𝐧𝐠 𝐏𝐨𝐢𝐧𝐭 Okay, so how do you choose? 𝑇𝐴𝐾𝐸 𝐴 𝑣𝑒𝑟𝑡𝑖𝑐𝑎𝑙 [𝐵𝑢𝑠𝑖𝑛𝑒𝑠𝑠 𝐷𝑜𝑚𝑎𝑖𝑛] 𝑤𝑖𝑡ℎ𝑖𝑛 𝑡ℎ𝑖𝑠, 𝑡𝑎𝑘𝑒 𝑎 ℎ𝑜𝑟𝑖𝑧𝑜𝑛𝑡𝑎𝑙 [“𝑈𝑠𝑒𝑐𝑎𝑠𝑒”].  Here's what I'd suggest: 1)𝐾𝑛𝑜𝑤 𝑌𝑜𝑢𝑟 𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦  2)𝑃𝑖𝑐𝑘 𝑎 𝑈𝑠𝑒 𝐶𝑎𝑠𝑒 3)𝐷𝑜𝑛'𝑡 𝐵𝑜𝑥 𝑌𝑜𝑢𝑟𝑠𝑒𝑙𝑓 𝐼𝑛 In the end, there's no one-size-fits-all answer. GenAI and traditional ML are like tools in a toolbox. Depending on the job, you'll use one or the other—or both. So, dive in, explore, and find out what excites you. Either way, you're in for a wild ride in the world of AI and data science. Thanks a lot Prabakaran Chandran and Aditya Sandilya for the meaningful insights #CoffeeWithPrabakaran #GenAI #DataScience

AI as Your Personal Data Analyst Imagine you're a business owner, drowning in a sea of data. Numbers, charts, and spreadsheets everywhere. It's overwhelming, right? Now, imagine having a personal assistant, but for your data. This assistant could understand your questions, dive into your data, and give you clear, actionable insights. That's what ThoughtSpot's new AI tool, Spotter, aims to do. How Does Spotter Work? * Natural Language Questions: Instead of learning complex data queries, you simply ask Spotter questions in plain English. For example, "What were our sales in the last quarter?" or "Why did our customer satisfaction drop last month?" * Data Diving: Spotter then dives into your data, analyzing it to find the answers to your questions. * Clear Insights: It presents the insights in a clear and understandable way, often with visualizations like charts and graphs. Why is This a Big Deal? * Democratizing Data: It makes data accessible to everyone, not just data scientists. * Faster Insights: You get answers to your questions instantly. * Better Decision Making: With clear insights, you can make better, data-driven decisions. The Future of AI and Business Intelligence ThoughtSpot's innovation is part of a larger trend in AI and business intelligence. As AI continues to evolve, we can expect to see more tools like Spotter that make data analysis easier and more powerful. So, the next time you're faced with a mountain of data, remember, you don't have to climb it alone. You can have an AI assistant like Spotter to guide you to the top.

The Top 4 Skills Marketers Will Need in the Future https://ift.tt/5j13wlq There’s no doubt that AI is impacting the role of marketers around the world, but the skills marketers need to thrive in the future go beyond simply leveraging AI tools. Learn the top 4 skills marketers will need in the future in this blog post. via moz (en-US) https://moz.com/blog April 16, 2024

AI analysis can be more effective for decision makers by siloing data into more specialized groups and vetting content for its relevance to a specific organization's needs. Get rid of some of the hay, and the needle is suddenly easier to find.

Many of the best uses for AI are with first-party data. That's why teams getting the most out of AI really prioritize data collection. They understand they need to focus as much on the data development as they do on the tools. 📅 They record their sales meetings and have them automatically transcribed for later analysis/referencing ✏ They load up databases with their marketing collateral for easier content development and referencing in the future. 👩💻 They report on EVERYTHING they do, pulling as much as they can into a central database for analysis This emphasis on data collection isn't new, but with AI, it's much faster to make it actionable. The better we can collect, organize, and make accessible our first-party data, the better results we get with AI. ❗ But... data isn't easily collected from the past, so getting it now is important. ❗ Don't wait until someone asks "what are the most common customer objections" to realize AI could give a data-backed answer to that question. or... "What have we written about [TOPIC]" to realize you could have provided resources/quotes/interviews in 15m... if you had collected that info beforehand.