Quick Answer: what Excel for data analysis means when a report is due
Five days, about an hour a day, ending in one report you can hand over and refresh next month. Excel for data analysis means four jobs in order: clean a messy export, join it to another export, aggregate it into the numbers people asked for, and refresh it next month without rebuilding it. Everything else is decoration.
One number from our own database, queried on 2026-09-25. Learners here have built 1,558 paths. Of those, 318 are for Python and 158 are for SQL. Three are for Excel, plus a fourth that is an internal test path of ours. Across 5,713 video slots in all those paths, 39 carry Excel in the video title, and of those 39, 8 mention VLOOKUP or XLOOKUP, 5 mention Power Query, and 3 mention a pivot table. That is an observation about our own users, not a claim about the world. But it matches how this skill arrives. People build a learning path for the language they want to learn. Almost nobody sets out to learn the spreadsheet already open on their second monitor, and then the recurring report lands on them with their name on it.
So this is not a channel list. It is five days with one job each, built from free videos, aimed at one deliverable: a report you can defend in a meeting and rerun next month without starting over.
The one decision that shapes the week: one-off or recurring
Before Day 1, answer one question. Does this report come back next month, or is it a single ask that dies when you send it? A one-off ends on Day 3, with a pivot table and a chart. A recurring report has to end on Day 4, in Power Query, with a refresh button. That one fact decides the order.
Most people get this wrong in the same direction. The first one feels like a one-off, so they clean the export by hand in a long careful afternoon, build the pivot, send it, and feel fine. Four weeks later the next export arrives with 800 more rows, a renamed column and two new regions, and the afternoon happens again. By month four you have spent more time than Power Query would have cost, and you still have nothing to hand a colleague when you are away.
So assume it comes back. Almost all of them do. The plan below is built so that every manual thing you learn on Day 1 has a recorded equivalent on Day 4, and you learn it by hand first on purpose. Skip to Power Query on Monday and you will follow a tutorial without knowing why any step is there.
One practical note. Get last month's real export onto your machine today, plus a copy of the report the last person sent. Every exercise below runs on your file, not a sample file.
Day 1: clean the export once, properly
By the end of today the ugly export is a proper table. One header row, one row per record, dates that sort as dates, numbers that add up, no trailing spaces, no duplicate rows, and no column that holds two different things. Every other day of this week assumes Day 1 happened.
Five things do most of the work. Turn the range into an Excel Table with Ctrl+T, so ranges grow with the data and formulas stop pointing at the wrong rows. Split columns that hold two facts, with Text to Columns. Remove Duplicates, after deciding which columns define a duplicate, because that decision is the whole job. TRIM and CLEAN for the invisible spaces that stop two identical-looking values matching. Then fix types: a number stored as text sits on the left of its cell, and that left-alignment tell will save you more grief this week than any formula.
Best for: Alex The Analyst (~1.4 million subscribers). Teaches this material to analysts rather than to accountants, so cleaning is framed as the thing that happens before analysis, not as housekeeping. Start with: Alex The Analyst, "Cleaning Data in Excel | Excel Tutorials for Beginners", https://www.youtube.com/watch?v=_jmiEGZ6PIY
Best for: MyOnlineTrainingHub (~880K subscribers). No padding, with a practice file, for a fast second pass over the same ground. Start with: MyOnlineTrainingHub, "How to Clean Data in Excel in 10 Minutes (Free File)", https://www.youtube.com/watch?v=oT4emh72fuA
Best for: Simon Sez IT (~730K subscribers). Slower and classroom-paced, if a quick pass left you unsure why a step was done. Start with: Simon Sez IT, "Excel Crash Course - Data Cleaning in Excel - Microsoft Excel Tutorial", https://www.youtube.com/watch?v=EwmuaqnoaKs
On your own file today: take last month's real export and write down every problem before fixing anything: merged cells, totals rows inside the data, dates in two formats, a region column with trailing spaces. That list becomes your Day 4 script. Then fix them by hand and save the result separately. Keep the original untouched.
Day 2: lookups and conditional sums
By the end of today you can answer a question that needs two files. XLOOKUP pulls a value from one table into another by a matching key. SUMIFS and COUNTIFS add up or count only the rows that meet conditions you set. Those two ideas turn two exports into one number.
XLOOKUP is the modern replacement for VLOOKUP, and worth learning even though every older colleague writes the old one. It looks left as happily as right, it does not break when someone inserts a column, and it has a built-in answer for when nothing matches, which stops the report filling with errors in front of your boss. If your version of Excel lacks it, use INDEX and MATCH until the upgrade lands.
SUMIFS is the one people underestimate. Most of what a month-end report reports is a conditional sum: revenue for this region, this month, excluding refunds. Write the conditions as cell references rather than typed-in text and one formula serves a whole grid.
Best for: Kevin Stratvert (~4.4 million subscribers). Explains software to non-specialists for a living, at a slower pace than anything else on this list. Start with: Kevin Stratvert, "How to Use XLOOKUP in Excel - Step-by-Step Tutorial for Beginners", https://www.youtube.com/watch?v=y5H6Yi3V4cw
Best for: Kenji Explains (~970K subscribers). Fast, finance-flavoured, and covers the arguments people skip, including the not-found case. Start with: Kenji Explains, "The Ultimate XLOOKUP Tutorial (The Best Excel Formula)", https://www.youtube.com/watch?v=Gfu26nNvrsk
Best for: Excel Campus (~710K subscribers). A short refresher if you half know this and want the syntax confirmed. Start with: Excel Campus, "Xlookup In Excel: Explained In 3 Minutes", https://www.youtube.com/watch?v=txK0C6SkaMg
Best for: Leila Gharani (~3.1 million subscribers). The clearest free treatment of multiple-criteria conditional sums, which is the half of Day 2 most tutorials rush. Start with: Leila Gharani, "How to Use SUMIFS, COUNTIFS and AVERAGEIFS in Excel (Multiple Criteria)", https://www.youtube.com/watch?v=AZuBNWMh7VM
On your own file today: find the second file your report needs, the lookup table nobody documented. A region mapping, a product list, a target sheet. Join it to your cleaned export with XLOOKUP and count how many rows fail to match. That mismatch count is the most useful number you will produce all week.
Day 3: pivot tables
By the end of today you can go from ten thousand cleaned rows to the summary the report reports, in about a minute, without writing a formula. A pivot table groups rows by the fields you drag into it and aggregates the rest. It is the fastest route from data to answer in the whole application.
The mental model is small. Rows are what you group by. Values are what you measure. Columns split that measure across a second grouping. Filters cut the whole thing down first. Everything else in the interface is a variation on those four boxes.
Two habits separate a pivot you can trust from one you cannot. Build it from an Excel Table, not a fixed range, so next month's extra rows are included instead of silently ignored. And check the aggregation every time: Excel will count when you expected it to sum, usually because one cell is text, and a count looks plausible enough to ship.
Best for: Leila Gharani (~3.1 million subscribers). The tightest conceptual explanation of what a pivot table is doing, which matters more than the clicking. Start with: Leila Gharani, "Excel Pivot Table EXPLAINED in 10 Minutes (Productivity tips included!)", https://www.youtube.com/watch?v=UsdedFoTA68
Best for: Alex The Analyst (~1.4 million subscribers) and Kevin Stratvert (~4.4 million subscribers). The same material aimed at someone producing an analysis, and a slower step-by-step version if the ten-minute one moved faster than your hands. Start with: Alex The Analyst, "Pivot Tables in Excel | Excel Tutorials for Beginners", https://www.youtube.com/watch?v=lH7HfwUFnYA , or Kevin Stratvert, "Pivot Table Excel | Step-by-Step Tutorial", https://www.youtube.com/watch?v=dvbLrwD2SpA
Best for: MyOnlineTrainingHub (~880K subscribers). The current-year version, which matters because the pivot table interface has actually moved. Start with: MyOnlineTrainingHub, "Excel PivotTables Made Easy in 2026 (+ What's Changed!)", https://www.youtube.com/watch?v=K3KqOTdypyw
Best for: Simon Sez IT (~730K subscribers) and Excel Campus (~710K subscribers). Two longer options for a second sitting, one a crash course, one a dashboards series opener. Start with: Simon Sez IT, "Pivot Table Tutorial - Learn PivotTables in 1 Hour - Excel Crash Course", https://www.youtube.com/watch?v=2RKw-HGCLB0 , or Excel Campus, "Introduction to Pivot Tables, Charts, and Dashboards in Excel (Part 1)", https://www.youtube.com/watch?v=9NUjHBNWe9M
On your own file today: rebuild last month's report as a pivot table, then compare your numbers to the ones actually sent. If they differ, find out why before you go home. The reason is almost always a filter, a date range or a duplicate.
Day 4: Power Query, the day that pays for the week
By the end of today next month is a refresh, not a rebuild. Power Query records the cleaning steps you did by hand on Day 1 as a named, editable sequence, then replays them on a new file. It ships inside Excel under Data, Get and Transform. This is the day that turns a report into a process.
The idea is simpler than the name. You point Power Query at a file or folder, and everything you do in its window, removing a column, splitting text, changing a type, filtering out the totals row, is recorded as a readable step in a list. Load the result into the workbook. Next month, drop in the new export and press Refresh. Your list of Day 1 problems is literally the script.
Two things are worth getting right on the first build. Set data types explicitly rather than accepting the guess, because a type guessed from the first two hundred rows will fail on a later file. And do not rename source columns in Excel after the query is written, because the query will look for the old name and stop.
The last piece is refresh order, and nobody warns you about it. If your pivot tables sit on top of a query, they can refresh against yesterday's data unless the query finishes first. That looks like a wrong number rather than a bug, which is worse.
Best for: Chandoo (~890K subscribers). The best structured entry point, with a sample file so you build something on the first pass. Start with: Chandoo, "How to use Power Query in Excel - Perfect Guide for Beginners (with FREE Sample Data)", https://www.youtube.com/watch?v=UAFExySaSPY
Best for: Simon Sez IT (~730K subscribers). The clearest statement, in plain language, of why you are doing this at all. Start with: Simon Sez IT, "Stop Cleaning Excel Data by Hand: Power Query for Beginners", https://www.youtube.com/watch?v=BrLQmJ1Vqk4
Best for: TrumpExcel (~530K subscribers). A numbered series opener, if you prefer short sequential lessons to one long video. Start with: TrumpExcel, "01 - Introduction to Excel Power Query (Get &Transform in Excel)", https://www.youtube.com/watch?v=yguT5TaqeXc
Best for: Leila Gharani (~3.1 million subscribers). The longer automation build, with files to download, once the basics land. Start with: Leila Gharani, "Learn to Automate Everything with Power Query in Excel (Download Files)", https://www.youtube.com/watch?v=x7mzOYEn0XA
Best for: Excel Off The Grid (~74K subscribers). The smallest channel here and the most specific, on the refresh-order problem above. Start with: Excel Off The Grid, "Make Power Query refresh before PivotTables | Reduce refresh to one click | Excel Off The Grid", https://www.youtube.com/watch?v=HrR-FwEXjGM
On your own file today: rebuild Day 1 in Power Query, step for step, from your list of problems. Then test it: take a different month's export, point the query at it, refresh. If the numbers come out right without you touching a cell, the week has paid for itself.
Day 5: present it and check it before anyone else does
By the end of today the report is readable by someone who was not in your head, and you have checked it yourself. One chart that makes one point. A pivot formatted so a manager can scan it. And a pass over your own numbers looking for the errors that survive a clean build and still embarrass you in a meeting.
Checking comes first, because it is the part people skip. Four checks catch most of it. Does your pivot's grand total match the row count and sum of the source? Do the mismatched rows from Day 2 still exist, and do you know where they went? Are blanks in a grouping column quietly forming their own category? Does last month's number, recomputed by your new process, match what was sent? If not, you have found an old error or made a new one, and you need to know which before Monday.
Then presentation, and the bar is lower than you think. One chart, one point, a title that states the point in words. A pivot with number formatting applied, columns wide enough to read, no grid of raw decimals. Nobody has ever complained that a report was too plain.
Best for: Leila Gharani (~3.1 million subscribers). Both halves of today, and the checking video is the one most analysts have never seen and should have. Start with: Leila Gharani, "Every Analyst Should Run This Check on Their Excel Report", https://www.youtube.com/watch?v=v85SsD3HMMk , then "Stop Sending Ugly Pivot Tables to Your Boss", https://www.youtube.com/watch?v=7eqGp_K4hIE
Best for: Kenji Explains (~970K subscribers). Optional, and only if the report is meant to be explored rather than read once. Start with: Kenji Explains, "Make an Interactive Excel Dashboard From Scratch (Full Tutorial)", https://www.youtube.com/watch?v=3GceuJe_PYc
On your own file today: run the four checks, write the one sentence the chart is meant to prove, and hand the file to a colleague who does not work on this report. Ask them what the number means. If they cannot tell you in ten seconds, the problem is the layout, not them.
What to skip this week
Skip VBA and macros, Power Pivot and DAX, shortcut memorisation, array formula gymnastics, most of Excel's function library, and chart decoration. All six are real and some are genuinely useful later. None of them gets a defensible report out of your hands by Friday, and every hour spent on them is an hour not spent on Day 4.
VBA and macros. VBA is Excel's built-in programming language and a macro is recorded VBA. It was the automation answer for twenty years and it is not the first answer now. Power Query does the data preparation half better, and a colleague can read a query's step list where they usually cannot read VBA. It stops being skippable the day you inherit a workbook that already runs on macros, because then you maintain it whether you chose it or not.
Power Pivot and DAX. The data model plus its formula language, for relationships between many tables and measures a pivot table cannot express. Real power, real learning curve. It stops being skippable when one figure has to be calculated across several tables at once, or when the source data outgrows a worksheet.
Memorising shortcuts. Shortcut lists feel productive and change nothing about whether your report is right. Learn the four you use hourly, Ctrl+T, Ctrl+Shift+L, Ctrl+Arrow and F4, and let the rest arrive on their own. It stops being skippable if you are graded on speed in a live interview test.
Array formula gymnastics. Dynamic arrays and clever nested one-cell solutions are impressive and fragile. A formula nobody else can read is a liability on a report that outlives you. Prefer a helper column. It stops being skippable when you hit a problem that genuinely has no other answer, which is rarer than the tutorials suggest.
Most of the function library. Excel ships hundreds of functions and this week needs about twelve: XLOOKUP, SUMIFS, COUNTIFS, IF, IFERROR, TRIM, TEXT, LEFT, RIGHT, EOMONTH and a couple of date functions. Search for the rest when a specific problem demands one, which is exactly when it stops being skippable.
Chart decoration. Gradients, shadows, secondary axes, pie charts with eight slices. Decoration is what people do instead of deciding what the chart is for. It stops being skippable when the audience is external and a designer hands you a brand template, in which case follow the template and stop.
Two of these connect to other tracks. If your report keeps pulling from a database export somebody else runs, our SQL reporting plan for CRM managers covers getting the export yourself. If dashboard requests keep coming, the Day 4 skills carry over, and our roundup of YouTube channels for Power BI certification is the step after this week.
If you only have one sitting
If you have one long session instead of five days, take a single crash course covering cleaning, lookups and pivot tables in one pass, then do Day 4 separately as soon as you can. One sitting gets you a report you can send. It does not get you a report that survives next month, because Power Query will not fit inside it.
Best for: Alex The Analyst (~1.4 million subscribers). The compressed version of Days 1 to 3 in one video, aimed squarely at analysts. Start with: Alex The Analyst, "Learn Excel in Under 3 Hours | Pivot Tables, Lookups, Data Cleaning", https://www.youtube.com/watch?v=zXnQNytHCPM
Best for: excelisfun (~1.4 million subscribers). A full free data analysis course posted publicly, one video at a time, over years. Far more than a week's material, and the most thorough free thing on this page, so save it for month two. Start with: the playlist itself, excelisfun, "Excel & Power BI Basics Data Analysis Free Course (E-DAB)", https://www.youtube.com/playlist?list=PLrRPvpgDmw0lPPRiJO5dCUratRGpGx3aT . The channel's own announcement video, "Free Excel Data Analysis Basics Course at YouTube (E-DAB)", https://www.youtube.com/watch?v=r31X8YCF9bM , is a ninety-second signpost to that playlist rather than a lesson, so start at the playlist and treat the announcement as the table of contents.
One disclosure. Most of these creators sell paid courses on their own websites: Leila Gharani, Chandoo, MyOnlineTrainingHub, Excel Campus, Alex The Analyst, Kenji Explains and Simon Sez IT all do, and several of them open the very videos linked above with a pitch for one. Every video named in this plan is free on YouTube and nothing here depends on buying anything. Treat those paid products as an optional extra offered separately, not as part of the free channel content recommended above.
Frequently asked questions
Six questions that come up in the gap between being handed a recurring report and admitting nobody ever taught you the tool you use every day.
What does Excel for data analysis actually mean?
Four jobs, done in order. Clean a messy export so every column holds one consistent type. Join it to a second export so two files answer one question. Aggregate it into the numbers people asked for. Then refresh it next month without rebuilding anything. Formulas, pivot tables and Power Query are just the tools for those four.
Can you learn Excel for data analysis in one week?
Enough to ship one report, yes. Five days at about an hour each covers cleaning, lookups, conditional sums, pivot tables, Power Query and a sanity check. It will not make you fast, and it will not cover modelling. It will get a defensible month-end report out of your hands on time.
Do you need Power Query, or are pivot tables enough?
Pivot tables are enough for a one-off. Power Query is what makes a recurring report survive. If the same export lands every month, the cleaning steps you did by hand on Day 1 become a recorded sequence that replays on the new file. That is the difference between an hour and a morning.
Should you learn VBA or macros for a monthly report?
No, not this week. VBA is Excel's programming language and macros are recorded VBA, and almost everything a report needs from them is now done better by Power Query, which is easier to read and easier to hand over. Learn VBA when you inherit a file that already runs on it.
Is Excel still worth learning in 2026, or should you go straight to Power BI?
Learn Excel first, because your colleagues open Excel. Power BI is the better tool for dashboards that many people read, and the Power Query skills from Day 4 transfer to it directly. But a report nobody can open is not a report, and the file on the shared drive is still a workbook.
Does LearnPath give a certificate for finishing an Excel learning path?
Yes. LearnPath builds a path of free videos, generates a quiz from each transcript, and issues a certificate when the path is finished. Three learners here have ever built an Excel path, so treat that certificate as proof you finished your own plan, not as a credential a hiring manager knows.
Proving you did it
LearnPath turns a topic into an ordered path of free YouTube videos, generates a quiz from each video's transcript so you find out whether the watching stuck, and issues a certificate when the path is finished. For Excel, that certificate proves you did the work in the plan. What proves you can do the job is a report that ran again the following month without you rebuilding it.
Be clear about what the certificate is not. It is not accredited, no employer has heard of it, and it does not replace the Microsoft exams if your organisation asks for one. It is a record that you followed a sequence and passed the quizzes on the way. Worth something when nobody is checking your work, worth nothing if you present it as more.
The deadline does more work than the certificate, which is why this plan has days instead of a video count. Three learner-built Excel paths out of 1,558 is not a gap in ambition, it is a gap in timing: nobody schedules the spreadsheet, so the spreadsheet gets learned in a panic the week the report is due. If Excel is your first tool rather than your missing one, our guide on becoming a data analyst with no experience covers the order everything else goes in.
The proof at the end of this week is not a badge. It is one file, opened by someone else, with numbers that reconcile to last month, that you refreshed in front of them in ten seconds. Build a work-ready Excel path before your next month-end.