AI in Excel: Full Breakdown & Transcript

[00:07] with thousands of rows and simply ask which region is losing revenue or what caused our profit margin to fall instead of manually checking formulas, tables, and columns? Excel has long been used for formulas, pivot tables, charts, and

[00:21] data cleaning, but AI now is changing how we interact with spreadsheets. With increasingly ask questions in natural language, generate formulas, analyze insights faster. This does not mean traditional Excel analytic skills are

[00:36] becoming less important. Analysts still need to understand the data, verify meaningful trends, and make sure recommendations support the actual Excel is not just automation. It's about combining Excel skills, business

[00:50] analytics, and AI to make faster and smarter decisions. So, throughout this of this transformation. So, before we continue, here's a quick opportunity for AI-driven business transformation. Simplilearn in collaboration with IIM

[01:03] certificate program in AI for business strategy. This 10-month live online and professionals with at least 5 years of work experience who want to understand how AI can be applied to real business challenges. Throughout this program, you

[01:17] AI opportunities, build effective AI strategies, and make better business decisions using AI-powered insights. This curriculum covers areas such as applied machine learning, language AI, AI and machine learning, language AI, AI

[01:30] in marketing and finance, intelligent automation, ethical AI, and agentic AI theory, you will also learn through real-world business cases, simulations, practical exercises, projects, and live sessions led by IIM Kozhikode faculty.

[01:44] used tools and platforms such as Power BI, Tableau, Excel, Gemini, and Claude, analytics can support business decision-making. Upon successful certificate from IIM Kozhikode along with executive education alumni status

[01:59] experience. So, whether you're a business leader, manager, consultant, professional looking to take charge of AI initiatives within your organization, strategic skills needed to move forward. Explore the program through the link in

[02:14] today. So, before we get started, let me ask you a quick question. Which tool is mainly used for ETL in Excel? Your options are A, Power Query, B, Slicer, your answers in the comment section below. So, now let's get started.

[02:27] Business analytics is a process of using data to understand business performance collect information about sales, customers, marketing campaigns, inventory, finance, employees, or operations, but simply storing this

[02:40] value. The value appears when the data helps answers a useful questions. For example, a retail company might want to know why sales decrease last month. A which campaign generated the highest return. A finance team may need to

[02:54] Traditionally, analysts use Excel to organize this information, apply formulas, calculate the metrics, create pivot tables, and develop charts. AI adds another layer to this process. Instead of manually deciding every

[03:07] increasingly describe what they want to understand. AI can help identify information, or create an initial analytical view. Microsoft's current with formulas, tables, charts, pivot tables, filtering, formatting, and data

[03:22] insights. The important change is therefore not that Excel stops being a spreadsheet. It is Excel is becoming a more conversational and AI assistant analytical workspace. So, to understand how AI fits into Excel, we first need to

[03:34] analytics. The first is descriptive analytics, example, an Excel sales report might previous quarter. The second is diagnostic analytics, which asks why did

[03:47] it happen? Here, an analyst may filter the data by product, region, customer that the decline mainly occurred in the market. The third is predictive analytics, which asks what is likely to happen next. Historical trends can be

[04:00] customer behavior, or inventory requirements. Finally, prescriptive analytics, what should we do about it? If demand is expected to increase, the inventory or adjust marketing expenditure.

[04:12] progression. What happened? Why it happened? What may happen? What should we do? AI can make movement across these stages faster users can ask questions about data using natural language. But the analyst still needs to decide

[04:26] prediction, or recommendation make business sense. So, good analytics begin with good data. Business rarely has a perfect data set. Information may arrive from Excel files, CSV files, databases, JSON sources,

[04:38] other systems. So, before analysis begin, that information normally needs removing duplicate records, correcting inconsistent text, handling missing numerical formats, sorting the information, filtering unnecessary rows,

[04:53] and validating important fields. Suppose one data set contains the value New York and NY for the same region, a human can understand that these values may analytical calculations may treat them as separate categories.

[05:06] And this is why data cleaning remains important even when AI is available. filtering, finding duplicates, text and named ranges. These create a reliable structure before more advanced

[05:19] analysis begins. The principle is simple. Poor quality data and powerful AI can still produce poor quality conclusions. AI can assist with parts of understand what each field represents, which records should be included, and

[05:32] future of analytics is therefore not data combined with analytical knowledge and AI assistance. One of the biggest changes AI introduces is natural language interaction. Traditionally, if

[05:45] margin, the analyst needed to know the correct formula and cell references. describe the requirement, create a column showing so and so profit margin for every product. Microsoft Excel Copilot can then help create the

[05:59] works. It can also help generate lookups, summarize rows, identify trends and outliers, and explain formulas already present in a workbook. This distance between a business question and an Excel operation.

[06:13] terms of XLOOKUP, conditional formatting, pivot table fields, or want to know which customers generated the highest profit. Natural language AI helps translate that business intention into analytical actions. If the wrong

[06:27] question is poorly defined, even technically corrected formula may produce a misleading answer. So, AI may help generate formulas, but understanding Excel functions remain an important analytical skill. Consider a

[06:39] names, products, regions, quantities, prices, discounts, and revenue. Logical results based on conditions. Lookup functions such as XLOOKUP help connect tables. Statistical functions such as average, count, max, min, other measures

[06:55] help summarize data and identify patterns. Reference functions allow larger data sets. These functions transform raw records into business information. For example, an analyst may calculate whether each transformation

[07:08] formula may retrieve a product category from a reference table. Statistical calculations may identify average order value or compare across regions. AI changes how these formulas can be created and understood. An analyst who

[07:21] does not remember the exact syntax can describe the intended calculation, receive a suggested formula, and ask for an explanation. Microsoft specifically documents Copilot's ability to create formulas, columns, individual formulas,

[07:33] formulas. So, formulas are not disappearing. Instead, AI can make them easier to create, understand, troubleshoot, and apply. When data sets become large, analysts need a faster way to summarize

[07:45] information. This is where pivot tables become valuable. A pivot table can take thousands of transaction records and quickly answer a question such as which which product category is growing fastest, and how did quarterly sales

[07:58] change? Which customer segment produces the highest average order value? formulas for every comparison, analysts can reorganize the same data into different summaries. But modern business data sets often extend beyond a single

[08:12] one table, customer information in another, and product details in a third. And this is where Power Pivot and data modeling become important. Relationships analysts can work across multiple sources without repeatedly copying

[08:25] everything into one worksheet. Microsoft sales co-pilot can create and work with analytical insights. This means analysts increasingly move between two construction and conversational instructions. But the underlying

[08:39] relationships, aggregation, and business logic still matter because AI needs the analysis can happen. One of the most valuable features in modern Excel analytics is Power Query. Power Query supports ETL, which stands for extract,

[08:54] extracted from a source. Then it is transformed by cleaning, reshaping, combining, filtering, or reorganizing it. Finally, the prepared information is analysis. Consider a business that receives 12 sales files, one for every

[09:09] might repeatedly open each file, copy the information, remove unnecessary everything manually. Power Query allows repeatable workflow. Operations such as merge, append, group by, change data

[09:23] type, split columns, filter rows, and remove unnecessary information can become part of the query. Then when new data arrives, the analyst can refresh process. This is important to the future of AI-powered analytics because AI works

[09:36] is consistent. The more routine preparation can be automated, the more time analysts can spend asking better questions, interpreting results, and and AI are not competing technologies. They solve different parts of the same

[09:50] only when people can understand the results. This is where charts, dashboards, KPIs, slicers, and timelines become important. A manager may not want rows. Instead, the manager might need five important answer. What is the total

[10:04] revenue? Are we above or below the target? Which region is performing best? requires immediate attention? A dashboard brings these answers together. recognize. KPIs highlight important measures. Conditional formatting draws

[10:18] attention to your unusual values. Slicers allow users to filter reports by fields such as product, region, or customer. Timelines make date-based analysis interactive. Microsoft now also allows Copilot to assist with charts,

[10:30] pivot tables, highlighting, sorting, filtering, and dashboards. This reduces create an initial analytical view. However, good visualization still require judgment. Adding 20 chart does not automatically create a better

[10:42] metric matters, what comparison the audience needs, and what actions should visual. The analyst provides the business meaning. Traditional reporting often ends with describing the past. Revenue falls $5 million and customer

[10:56] churn increased. One region missed its target. Inventory decreased. But, further. Why did it happen? What might happen next? And what should we do? This Excel-based analytics much more powerful. AI can help users explore

[11:10] large data sets, surface unusual values, summarize patterns, generate formulas, natural language requests. Microsoft specifically supports questions about Copilot in Excel. Imagine a manager asking, "Which products experienced

[11:24] months?" Or, "Show the regions with strong revenue but but falling profit margins." The value is not simply that AI answers the questions faster. The users can participate directly in analytical exploration. But this also

[11:38] creates responsibility. Correlation does not always mean causation. Historical And recommendations need business context. Therefore, AI should accelerate remain responsible for interpreting evidence and making decisions. So, what

[11:52] Excel look like? It is unlikely to be a future where professionals simply stop everything. Instead, the analyst's role is moving higher up the decision-making repetitive analytical work such as generating formulas, reorganizing

[12:07] worksheets, summarizing data, producing pivot tables, creating initial values, and identifying patterns. Microsoft's April 2026 announcement goes further. Copilot's generative capabilities can perform multi-step app-native actions

[12:19] in control of the workbook. That means more time can potentially move towards questions such as, "Is this data reliable? Which KPI usually matters? Why recommendation make business sense? What decision should we make? And how should

[12:33] the result be communicated?" This is why analytical thinking remains critical. The World Economic Forum continues to identify AI and big data among the emphasizing analytical thinking and other human capabilities. The future

[12:46] strengths: business understanding plus Excel analytical skills plus AI capability. And that combination is what can turn Excel from a spreadsheet used intelligent platform for business decision-making. And with that, we have

[13:00] is transforming Excel and the future of business analytics. We saw how Excel is pivot tables, Power Query, and dashboards into a more intelligent analytical workspace where AI can help users explore data, generate analysis,

[13:14] automate repetitive activities, and move from questions to insights faster. But replace analytical thinking. The strongest analyst will be those who understand the data, understand the business problem, know how Excel works,

[13:27] and use AI to make that workflow faster and more effective. So, if you have any comment section below, and our team of experts will be happy to help you as soon as possible. Thank you for watching, and keep learning.