The Tableau Desktop Specialist is a hands-on exam wearing the clothes of a knowledge test. Salesforce, which owns Tableau, renamed the credential to Tableau Desktop Foundations in 2025, but the content, the question style and the 60-minute length are unchanged, so everything here still applies. The thing that makes this exam different from most entry-level certifications is that almost every question maps to an action you would take in the Tableau Desktop interface: dropping a field on a shelf, choosing a chart in Show Me, writing a quick calculation, building a dashboard. That is why the single most reliable predictor of passing is hands-on time in the free Tableau Public tool, not hours of video.
This guide is a full self-study course. It walks through each of the four weighted domains in depth, explains the handful of concepts that genuinely confuse people (the dimension/measure and discrete/continuous split, and the filter order of operations), then turns all of it into a week-by-week plan, a final-week routine, and a description of exam day. It is original teaching material and study guidance only. It contains no real or simulated exam questions, and you should always confirm the current name, weights and format against Tableau’s own exam guide before you book.
Chapter 1: Exam overview and how to use this guide
What the Desktop Specialist actually measures
The Desktop Specialist certifies foundational, working literacy in Tableau Desktop: that you can connect to data and prepare it, build and analyse visualisations, share what you find, and use the core Tableau vocabulary correctly. It is deliberately broad and shallow rather than deep. You are not expected to be a calculation wizard or to know every chart type, but you are expected to be comfortable doing the everyday things an analyst does in the product. Tableau’s own guidance is that the exam suits people with foundational skills and at least roughly three months of applying them in the product, and that recommendation is the single best clue about how to prepare: spend the time in the tool.
The exam is organised into four domains with official weights: Connecting to and Preparing Data at 25%, Exploring and Analyzing Data at 35%, Sharing Insights at 25%, and Understanding Tableau Concepts at 15%. Those weights are the most important planning fact in this guide. More than a third of the exam sits in Exploring and Analyzing, so that is where the bulk of your building time belongs, but notice that Sharing Insights is a full quarter and is the domain people most often under-rate.
Format and the practical mindset
The exam is 45 questions in 60 minutes, of which 40 are scored and 5 are unscored trial questions that do not count, and the time limit includes a few minutes for reviewing the non-disclosure agreement at the start. Questions are multiple choice and multiple response (some ask you to select more than one correct answer), delivered through Pearson VUE either at a test centre or online with a proctor. The passing standard is a scaled score of 750 out of 1000, and because the score is scaled rather than a raw percentage, your goal is broad competence across all four domains rather than perfection in one. Unusually for a vendor certification, the credential does not expire, so there is no renewal to plan for.
The practical mindset matters here. A large share of questions describe a small task (“which shelf would you use to…”, “what does this produce…”) that is trivial if you have done it and a guess if you have not. Read every question as “have I actually done this in Tableau?” and let that drive your study toward building rather than reading.
How to use this course
Read the chapters in order at least once. The Understanding Tableau Concepts material in Chapter 5 is small in exam weight but it is the foundation the other three domains rest on, so it is taught alongside them rather than saved for last. Treat the bold terms as a checklist: by the end you should be able to explain each one in a sentence and, more importantly, show it in Tableau Public. The final three chapters turn the content into a schedule, a final-week routine and an exam-day plan. Short worked examples appear where an idea is easy to misread, but none of these are exam questions; they are teaching illustrations.
Chapter 2: Connecting to and Preparing Data (25%)
This domain is the start of every Tableau project: getting data into the tool and shaping it so it is ready to analyse. A quarter of the exam lives here, and most of it is mechanical once you have done it a few times, so the goal of this chapter is to make the moving parts familiar.
What this domain covers
You connect Tableau to a data source, which can be a file (such as an Excel workbook, a CSV or a .hyper extract), a server (such as a SQL database or a cloud source), or one of the many built-in connectors. You land on the data source page, the screen where you arrange tables, set data types, rename and hide fields, and preview rows before building anything. From here you decide how the data is combined and whether Tableau queries it live or works from a saved snapshot.
Live connections versus extracts
One early decision is whether to use a live connection or an extract. A live connection queries the source database directly each time the view updates, so the data is always current but performance depends on that source. An extract is a compressed snapshot of the data saved in Tableau’s own .hyper format; it is fast and works offline, but it is only as fresh as the last time you refreshed it. The teaching point the exam cares about is the trade-off: live for up-to-the-moment data against a capable source, extract for speed, portability and working away from the source.
Joins, unions and blends: the distinction that matters most
Combining data is the part of this domain that generates the most confusion, and the exam tests the distinction directly, so be precise about three different operations.
A join combines tables at the row level on a shared key, widening each row with matching columns from another table, exactly like a SQL join. You choose a join type (inner, left, right or full outer) depending on which unmatched rows you want to keep. Joins are used when the tables live in the same data source and share a key, for example an Orders table joined to a Customers table on a customer ID.
A union stacks rows from tables that have the same columns, growing the data downward rather than widening it. Unions are the right tool for combining like-shaped files, the classic case being twelve monthly sales files with identical columns that you want to treat as one long table.
Data blending is different again. It links two separate data sources on a shared field at the level of the view, rather than merging them into one source. Blending is what you reach for when the data lives in different places that cannot easily be joined, for example sales figures in a database blended with targets in a spreadsheet on a common Region field. A useful way to hold the three in mind: join widens rows in one source, union stacks rows in one source, and blend links across two sources in the view.
Data types and preparing fields
On the data source page you also set each field’s data type (string, number, date, date and time, boolean, or geographic), because Tableau behaves differently depending on type, most visibly with dates and with geographic roles that enable mapping. Light preparation also happens here: renaming fields to something readable, hiding fields you will not use, splitting a column (for example separating a full name), and creating simple aliases. Getting the types and names right at this stage prevents a lot of confusion later.
Common pitfalls in this domain
The recurring mistake is mixing up the three combine operations, especially reaching for a join when the situation calls for a union or a blend. Another is ignoring data types and being surprised when a date will not behave or a map will not draw because the geographic role is not set. As a teaching example to make the join/union split concrete: if you have one wide table of customers and one wide table of orders that share an ID, that is a join; if you have the same orders split across several monthly files, that is a union. Build both at least once in Tableau Public and the distinction stops being abstract.
Chapter 3: Exploring and Analyzing Data (35%)
This is the largest domain and the heart of the exam, and it is where most of your hands-on practice should go. It covers building visualisations, controlling how they look, filtering and sorting, grouping data, and adding calculations. The unifying idea is that in Tableau you build a view by placing fields onto shelves and cards, and the tool draws the result.
Building views: shelves, the Marks card and Show Me
The structure of any view comes from the Rows and Columns shelves: fields you place there become the axes and headers of the chart. The Marks card then controls how the data is drawn, with properties for colour, size, label, detail, shape and tooltip, and a mark-type selector (bar, line, circle, and so on). Learning what each Marks property does is high-value, because it explains how a plain chart becomes an informative one. Show Me is a helper panel that suggests chart types based on the fields you have selected and will build a starter view for you; it is useful for learning which fields make which charts, though experienced users often build directly on the shelves for more control.
You should be comfortable producing the common chart types by hand: bar charts, line charts, scatter plots, highlight tables, maps, and a dual-axis chart (two measures sharing or aligning axes, a frequent source of “how do I do that” questions). The exam does not need every chart, but it expects the everyday ones to be second nature.
Filtering and the order of operations
Filtering limits what a view shows, and Tableau offers several filter types that, crucially, apply in a fixed order. From highest to lowest precedence the sequence is: extract filters, then data source filters, then context filters, then dimension filters, then measure filters, and finally table calculation filters. This order is one of the few genuinely conceptual things in the exam, because it explains results that otherwise look wrong. A context filter, for instance, runs before ordinary dimension filters, which is why it changes what a “top N” filter is computed against. As a teaching example: if you filter to the top 10 products and also filter to one region, putting the region filter into context makes the top 10 calculate within that region rather than across everything. Understanding why is exactly what the exam rewards.
Sorting, groups and sets
Sorting arranges the members of a dimension, either manually, alphabetically, or by a measure (for example sorting products by total sales). Groups combine selected members of a field into a single category, which is how you tidy up messy data or roll several items into an “Other” bucket. Sets are custom subsets of the data that can be either fixed or dynamic (defined by a condition such as “customers with sales above a threshold”); sets can be shown in a view, combined, and used inside calculations, which makes them more powerful than groups. The distinction worth holding is that a group renames-and-buckets members, while a set is a flexible, often conditional, membership that you can reuse.
Calculations: row-level, aggregate, table and LOD
Calculations are where this domain gets its depth, and the exam expects familiarity with several kinds.
A basic, row-level calculation operates on each row before aggregation, such as [Sales] - [Cost] to produce profit per row. An aggregate calculation works on summarised values, such as SUM([Sales]) / SUM([Quantity]) for an average price; the key idea is that the aggregation is part of the formula. A table calculation computes across the values already displayed in the view, such as a running total, a percent of total, a rank, or a moving average; because they depend on what is in the view, they are governed by a direction (for example “table across” or “down”) that decides how they are computed. Finally, a Level of Detail (LOD) expression lets you compute at a specific granularity regardless of the view, written with FIXED, INCLUDE or EXCLUDE, as in { FIXED [Region] : SUM([Sales]) } to get sales per region even when the view shows something more detailed. For this exam you need to recognise what each type is for and read a simple example, not to write advanced LOD logic.
Maps and a note on practice
Tableau builds maps automatically when a field has a geographic role (such as Country, State or City), plotting marks on a map and letting you layer measures via colour and size. The practical lesson across this whole domain is the same: build, do not just read. Twenty small vizzes you constructed yourself, including one dual-axis chart, one running-total table calculation, one set, and one map, will teach you more than a hundred you only watched.
Chapter 4: Sharing Insights (25%)
A full quarter of the exam is about communicating findings, not just producing them, and it is the domain candidates most often under-prepare because it feels less technical than calculations. It covers assembling dashboards, building stories, formatting for clarity, and publishing or sharing the result.
Dashboards
A dashboard is a single canvas that combines multiple worksheets into one view, along with interactive elements. You add sheets, arrange them with tiled or floating layout, and bring in objects such as text, images, web pages and containers. The interactive power comes from filters and dashboard actions: a filter action lets a selection in one sheet filter another, a highlight action emphasises related marks, and a URL action opens a link. Using one worksheet as a filter for the others, often by turning a sheet into a filter or adding a filter action, is a common task, so build a dashboard where clicking a bar filters a second chart at least once. Knowing the difference between tiled and floating, and what the Marks-to-dashboard-action flow does, covers most of what this part asks.
Stories
A story is an ordered sequence of story points, each capturing a particular worksheet or dashboard in a particular state, used to walk an audience through a narrative or an analysis step by step. The distinction the exam wants is simple: a dashboard shows things together on one screen for exploration, while a story sequences them to make a point. Build a short story of two or three points so the navigator and the captions are familiar.
Formatting and publishing
Formatting covers the everyday polish: fonts, colours, number and date formats, titles, tooltips, legends and reference lines. You should know where formatting is controlled (at the worksheet level, and per-field) and that tooltips can be customised to show exactly the detail you want on hover. Publishing and sharing is the final step: you can publish a workbook to Tableau Public (the free, public option), or to Tableau Server or Tableau Cloud in an organisation, and you can export images or PDFs. For study purposes, publishing to Tableau Public is ideal because it is free and it doubles as building a portfolio you can show an employer.
The pitfall to avoid here
The mistake is skimming this domain because it is “just dashboards”. At a quarter of the exam it carries as much weight as the entire data-preparation domain, and the questions reward someone who has actually assembled a dashboard with a filter action and built a short story. Spend real time here, not just a final skim.
Chapter 5: Understanding Tableau Concepts (15%)
This domain is the smallest by weight but the most important for understanding everything else, which is why it is worth studying early even though it appears last in the exam blueprint. It is the vocabulary and the mental model of how Tableau works.
Dimensions versus measures
The first core distinction is dimensions versus measures, and it is about a field’s role. Dimensions are fields that slice the data into categories, such as Region, Category or Order Date; they define the level of detail of the view. Measures are numeric fields that get aggregated, such as Sales or Quantity, which Tableau by default sums, averages or counts. By convention Tableau colours dimensions blue and measures green in the Data pane, but the defining idea is the role: a dimension cuts the data, a measure is the number being summarised. Almost every view is some combination of dimensions setting the structure and measures filling it in.
Discrete versus continuous
The second distinction, and the one most often confused with the first, is discrete versus continuous, which is about how a field is plotted, signalled by the pill colour on the shelf. A discrete field (a blue pill) produces distinct headers and a finite set of separate values. A continuous field (a green pill) produces an axis and an unbroken range. The reason people confuse this with dimensions and measures is that the defaults line up (dimensions are often discrete, measures often continuous), but they are independent: a date can be treated as discrete (separate month headers) or continuous (a time axis), and a measure can be made discrete. The clean way to keep them apart is to remember that dimension/measure is about the field’s job, while discrete/continuous is about the blue-versus-green pill and whether you get headers or an axis.
Aggregation and how Tableau is organised
Aggregation is the act of rolling detailed rows up into a summary, with functions such as SUM, AVG, COUNT and MIN/MAX; understanding that Tableau aggregates measures by default explains a lot of its behaviour. You should also know how a Tableau workbook is organised: a workbook contains worksheets (individual views), dashboards (combinations of worksheets) and stories (sequences), with the Data pane on the left holding dimensions and measures and the shelves and cards arranged around the canvas. This structural vocabulary is what makes the rest of the interface legible.
Why this small domain pays off
Although it is only 15% of the exam directly, getting these concepts solid quietly raises your score across the other three domains, because filtering, calculations, building views and combining data all assume you understand roles, pill colours and aggregation. If you can explain dimensions versus measures and discrete versus continuous in your own words and demonstrate each in Tableau Public, the rest of the exam reads much more easily.
Chapter 6: Study plan and timeline
With the content understood, the remaining work is pacing it so that the hands-on building, which is what actually moves your score, never gets squeezed out. Two things drive the plan: the domain weights and the fact that this is a practical exam best learned in the tool.
Set up to practise from day one
Before anything else, download the free Tableau Public desktop application and grab a sample dataset (the bundled Superstore data is ideal, or any public CSV). You do not need a paid licence to prepare for this exam. From the first session, the rule is to rebuild every concept you read about rather than only watching it, because passive video is the slowest route to a hands-on credential.
Allocate time by weight, but protect Sharing Insights
Spend your hours roughly in proportion to the domain weights, with the most going to Exploring and Analyzing (35%), a solid block each on Connecting and Preparing (25%) and Sharing Insights (25%), and a smaller but early pass on Understanding Tableau Concepts (15%). The one adjustment to pure proportion is to consciously protect time for Sharing Insights, since it is the domain most people neglect despite it being a quarter of the marks.
Choose a timeline
A balanced plan for someone newer to Tableau runs about four to six weeks at roughly five hours a week: a week or so connecting and preparing data, two weeks or more in the Exploring and Analyzing domain building charts, filters, sets and calculations, a week assembling dashboards and a story, and a final week consolidating the concepts and doing a timed review. Total beginners can stretch this to eight weeks at about four hours a week, building Tableau vocabulary and hands-on skill from scratch one domain at a time. People who already use Tableau day to day can compress to a fast track of around fifteen hours total, reviewing the four domains, drilling the calculation types and the filter order, and building one dashboard end to end. To turn whichever timeline you pick into dated weeks for your own start date, use the free study-plan generator.
Build judgement, not recall, throughout
Throughout the plan, after each topic, rebuild it in Tableau Public the same day rather than saving practice for the end. The exam rewards the person who has connected data, built the common charts, written a basic and a table calculation, assembled a dashboard with a filter action, and published it. If you are weighing Tableau against the Microsoft alternative before committing, the PL-300 vs Tableau Desktop Specialist comparison covers how to choose by the platform your target employers use.
Chapter 7: Final preparation and revision
In the closing week, the work shifts from learning new material to consolidating and confirming readiness, and from building freely to building under a clock.
Consolidate the sticky points
Revisit the three things that trip people up. First, re-read and re-demonstrate the dimension/measure versus discrete/continuous distinction until you can show both in the tool without hesitating. Second, walk back through the filter order of operations and prove to yourself with a context filter why a top-N result changes. Third, run once more through the join, union and blend distinction by building a small example of each. These three areas account for a large share of the questions that separate a pass from a near miss.
Do a timed pass and a self-check
Do at least one timed run across all four domains so the 60-minute pace feels normal; with 45 questions in an hour you have a little over a minute each, which is comfortable if you are not stuck. Use the mastery checklist as a final gate: can you connect a file or server source and prepare it with a join or union, build the common chart types and use Show Me sensibly, explain the two key distinctions, write a basic calculation and a percent-of-total table calculation, and assemble and publish a dashboard and a story? If any item is shaky, that is your revision target.
Use only legitimate materials
Prepare only from Tableau’s own training, the official exam guide, and your own practice in Tableau Public. Avoid sites offering recycled live questions; they breach Tableau and Salesforce certification policy and copyright, and they teach recall instead of the practical skill the exam actually measures. Building twenty real vizzes is both the honest route and the faster one.
Chapter 8: Exam day and format
What to expect on the day
On the day, the exam is 45 questions in 60 minutes (40 scored, 5 unscored), taken either at a Pearson VUE test centre or online with a proctor. If you test online, the check-in process opens about 30 minutes before your scheduled time, so start early to clear the system checks and the identity and workspace verification without stress, and make sure your computer, network and room meet the technical requirements in advance. The first few minutes include reviewing the non-disclosure agreement, which is part of the 60.
Pacing and question style
The questions are multiple choice and multiple response, so read carefully whether a question wants one answer or several, and select all that apply when asked. Most questions are small, concrete and practical: which shelf or card to use, what a given configuration produces, how to combine two data sources, what a calculation type does. Pace yourself steadily rather than lingering; if a question is genuinely uncertain, make your best choice and move on, because the comfortable time budget rewards momentum over agonising. You need a scaled score of 750 out of 1000, and because the score is scaled across the whole exam, broad competence beats deep mastery of one domain.
After the exam and what it is worth
Your result is shown after you finish, in the score-reports area of your Pearson VUE account. A pass earns the Desktop Specialist (now Desktop Foundations) credential, which does not expire, so there is nothing to renew. Treat it as a stepping stone: it proves working Tableau literacy on a CV and is a sound first credential, but for analytics roles the role-based Tableau Certified Data Analyst is the natural next step, and if your target employers run Power BI instead, the Microsoft PL-300 maps to that tool. Confirm the current exam name, domain weights and format on Tableau’s certification page before you book, since Salesforce periodically updates the programme.