← All Kits · Certification Kit · All guides
Tableau Desktop Foundations Exam: 40 Questions, and 20 Right Passes It
By the end of this page you'll know exactly what the Salesforce Certified Tableau Desktop Foundations exam asks, how it's scored, which version of Tableau it tests, and which of its four sections is worth more than double another. Every figure comes from Salesforce's own exam guide, checked on 9 August 2026. It takes about twelve minutes.
Here's what to do with it today. Before you open another lesson, put the four section weights next to your calendar and split your remaining study hours in the same proportion. Section 2 is 37% of the exam, which works out to about 15 of the 40 scored questions. You need 20 correct to pass. That one section is three quarters of a pass on its own.
The short version: 40 scored questions, 70 minutes, and 48% passes. The four sections are not the same size, and the published weights tell you where your hours belong.
The sizes are the whole reason to read this, so they get the picture.
- The format, in numbers you can plan around
- Where the marks are, converted into questions
- What a 48% pass mark actually buys you
- The version it tests, and why that saves you time
- Details people miss
- Why studying to the weights works
- Building your own study order
- The whole thing on one screen
1. The format, in numbers you can plan around
Before the table: 70 minutes and 40 questions. How long is that per question, and does it feel tight to you?
It's about a minute and three quarters each, and for most people it isn't tight at all. Here is the full format.
| Thing | Value |
|---|---|
| Scored questions | 40 |
| Unscored questions | Up to 5 more, mixed in unmarked |
| Question types | Multiple choice and multiple select |
| Time | 70 minutes |
| Passing score | 48% on the English version, 55% on the Japanese version |
| Registration fee | US$75, plus local tax |
| Retake fee | Free |
| Prerequisite | None |
| Tableau version tested | 2022.3 |
Two of those rows do real work on exam day.
The unscored questions are invisible. You could see 45 questions and only 40 of them count, and nothing on screen tells you which five are which. So a question that feels strange or unusually hard may simply not be scored. Answer it and move on. There's no way to spot them and no benefit to trying.
Every question is multiple choice or multiple select. There is nothing to build. You will not be dropped into a live copy of Tableau and asked to make a chart. That changes how you should practice, and it's the single biggest planning fact on this page. Reading and recognizing beats building, for this exam only.
On timing, work the arithmetic yourself rather than trusting the feeling. 70 minutes is 4,200 seconds. Divide by 45, the worst case where every unscored question shows up, and you get 93 seconds a question. Divide by 40 and you get 105 seconds. Either way you have well over a minute for a question with four short options.
2. Where the marks are, converted into questions
Before the numbers: the four sections are weighted 23, 37, 25 and 15 percent. Which one would you study first?
Percentages are hard to plan against, so turn them into questions. Multiply each weight by the 40 scored questions.
| Section | Weight | Arithmetic | Scored questions |
|---|---|---|---|
| 1. Connecting to and Preparing Data | 23% | 0.23 × 40 = 9.2 | about 9 |
| 2. Exploring and Analyzing Data | 37% | 0.37 × 40 = 14.8 | about 15 |
| 3. Sharing Insights | 25% | 0.25 × 40 = 10 | about 10 |
| 4. Understanding Tableau Concepts | 15% | 0.15 × 40 = 6 | about 6 |
Those four add to 40, which is the check that the conversion is right. Salesforce doesn't promise an exact count per section, so treat these as the shape rather than a guarantee.
Now the sentence worth carrying into your study plan. Section 2 is about 15 questions and you need 20 to pass, so section 2 alone is three quarters of a pass. Section 4 is about 6 questions, which is less than a third of one.
Say out loud what that means for an hour of your time, before reading on. An hour spent on exploring and analyzing is worth roughly two and a half hours spent on Tableau concepts, because 37 divided by 15 is about 2.5. This is the hinge of the whole page, and it's the reason to read a weighting table at all.
What section 2 covers is the daily work of the tool: filtering, sorting, grouping, sets, hierarchies, calculations, table calculations, and mapping. If you want a single place to start, start there. That is also why four of the six Tableau guides on this site sit inside that section.
3. What a 48% pass mark actually buys you
Before the answer: 48% of 40 questions. How many can you get wrong and still pass?
Twenty. The arithmetic is 0.48 × 40 = 19.2, and since there is no such thing as a fifth of a question, you need 20 correct. That leaves 20 you can miss.
That is a genuinely forgiving mark, and it's worth knowing for one reason only: it should stop you postponing the booking. People delay a certification for months waiting to feel ready, and 48% is not a mark that requires feeling ready. It requires knowing half the material.
What it should not do is turn into a plan to skip a section. Twenty wrong sounds like a wide margin until you notice that multiple-select questions usually need every box right to score, and that the questions you feel confident about are the ones you're most likely to misread. Aim well above the line, and treat the 48% as permission to book rather than permission to prepare less.
The retake is free, which changes the risk further. A first attempt costs $75 and a second costs nothing, so the real cost of being wrong about your readiness is time, not money.
4. The version it tests, and why that saves you time
Before the explanation: the exam guide says it currently tests on Tableau 2022.3. Why would that be good news?
Because it puts a fence around what you have to learn. Anything Tableau shipped after 2022.3 is outside the exam, so a newer feature you read about in a release note or a recent video is not something you need for this.
That matters more than it sounds. Tableau ships several releases a year, and the tutorials you find by searching are written against whatever version was current when the author wrote them. A 2026 walkthrough of a 2025 feature is good learning and irrelevant preparation.
Two practical consequences. First, if a menu in your copy of Tableau doesn't match a screenshot in your study material, the exam is more likely to follow the older wording. Second, when your notes and a video disagree, Tableau's own help documentation is the tiebreaker, not the video.
Picture your own study list for a moment. How many items on it are features you've only ever seen in a recent release note? Those are the ones to move to the bottom.
5. Details people miss
Before the list: someone tells you a wrong answer costs you marks. Are they right?
No. Nothing in the published guide describes a penalty for a wrong answer, and the score is a percentage of correct responses. So there is never a reason to leave a question blank. Answer every one, including the ones you're guessing at.
Five more worth knowing.
The exam was renamed. This is the exam that used to be called Tableau Desktop Specialist. Older study material, forum posts and course titles still use the old name, and they're usually talking about the same thing. Check the publication date and the section weights before trusting any of it.
Multiple select tells you how many to pick. These questions say how many answers to choose. Read that number before reading the options, because choosing two when it asked for three is a wrong answer that felt right.
The passing score differs by language. 48% on the English version, 55% on the Japanese. If you're reading a study group's advice, check which version they sat.
The section names are the syllabus. "Connecting to and Preparing Data" is not a category invented for the score report. It's a list of the things you'll be asked about, and the full exam guide breaks each section into numbered objectives. Reading those objective lists is the cheapest hour of preparation available.
Certification maintenance is published separately and it changes. The exam guide itself doesn't state an expiry date for this credential, and Salesforce keeps maintenance requirements on their own help pages, which were revised for the Winter '26 release. Do not assume a credential lasts forever because a blog said so. Check the maintenance page after you pass, and diarise it.
Why studying to the weights works
Splitting your hours by the published weights is not just tidy. It's the only way to make an hour of study comparable to another hour, because an exam that weights its sections unequally is telling you the exchange rate in advance. Most people never look it up, and then spread their time evenly, which quietly overpays for the smallest section.
Two findings from the memory research change how the hours themselves should be spent, and both are about how you practice rather than what you read.
The first is that testing yourself beats re-reading. Roediger and Karpicke had students study passages and then either restudy them or take a test on them. On a retention test a week later, the students who had been tested remembered substantially more than the students who had restudied, even though restudying felt more productive at the time (Roediger & Karpicke, 2006, Psychological Science, 17(3), 249–255). For a multiple-choice exam that is close to a free result: practice questions are both the preparation and the format.
The second is that the same total hours produce more if you spread them out. Cepeda and colleagues reviewed 254 studies covering more than 14,000 participants and found spaced practice reliably beat massed practice on later recall (Cepeda, Pashler, Vul, Wixted, & Rohrer, 2006, Psychological Bulletin, 132(3), 354–380). Six hours across six evenings beats six hours on a Sunday, at no extra cost.
Put those together with the weights and you get the whole method. Split the hours by section weight, spread them across days, and spend them answering questions rather than reading about answers.
On the exam facts themselves, Salesforce's published exam guide is the authority and it is the only source used above. Where a course description, a video or a summary disagrees with it, that page wins. This matters more than usual here, because the most commonly repeated third-party figures for this exam are wrong on the question count, the time limit and the passing score at the same time.
Building your own study order
Think about the last certification or test you prepared for. Did you know the section weights before you started, or did you find them afterwards on the score report?
Do this in order. It takes about twenty minutes and it's the only planning this exam needs.
- Read the four section names and their numbered objectives in the Salesforce exam guide. That list is the syllabus, and reading it is faster than any summary of it.
- Split your remaining hours 23 / 37 / 25 / 15. If you have twenty hours left, that's about 5, 7, 5 and 3. Write those four numbers down where you'll see them.
- Spend the first block on section 2, the exploring and analyzing work. It's the biggest section and it's also the part that makes the tool useful after the exam, so none of that time is exam-only.
- Answer questions rather than read notes, from the first evening. Getting one wrong early is the cheapest way to find out what you don't know.
- Spread the blocks across days, not into one weekend. Same hours, more retained, and it costs nothing to schedule.
- Book the exam before you feel ready. 48% and a free retake is a low-risk first attempt, and a booked date is what turns a study plan into study.
If you have paper nearby, there's one sketch worth five minutes and it's optional. Draw the four sections as four bars at their real weights, then draw your last week of study as four bars beside them. Whichever pair is furthest apart is your next study session.
The Tableau Desktop Foundations kit drills all four sections as multiple choice and multiple select, in the same shape as the exam, and links back to the guide here when you miss one.
Open the certification kit →Want the concepts first? Tableau concepts, start to finish covers dimensions, measures and aggregation in the order they build on each other.
The whole thing on one screen
This is the retrieval sheet. Cover the right column, work down the left, and say each answer out loud before you check it.
| Question | Answer |
|---|---|
| Scored questions | 40 |
| Unscored questions | Up to 5, mixed in, unmarked |
| Time | 70 minutes |
| Seconds per question | 93 at 45 questions, 105 at 40 |
| Passing score | 48% English, 55% Japanese |
| Questions needed to pass | 20 of 40 |
| Questions you can miss | 20 |
| Fee | US$75, retake free |
| Prerequisite | None |
| Version tested | Tableau 2022.3 |
| Question format | Multiple choice and multiple select. Nothing to build. |
| Section 1, Connecting and Preparing | 23%, about 9 questions |
| Section 2, Exploring and Analyzing | 37%, about 15 questions. The biggest. |
| Section 3, Sharing Insights | 25%, about 10 questions |
| Section 4, Tableau Concepts | 15%, about 6 questions. The smallest. |
| Former name | Tableau Desktop Specialist |
| Penalty for a wrong answer | None published. Never leave one blank. |
One last thought, and I'd like other people's answers. The fact that changed my own preparation was that nothing on this exam is built, only recognized, which makes practice questions worth more than practice workbooks. What's the fact about an exam you wish you'd known before you started studying for it, instead of afterwards?
Tableau for Analysts is 110 pages on how Tableau actually thinks: dimensions against measures, relationships against joins, sets, and level of detail. Enough that you stop dragging fields until it looks right.
Tableau for Analysts, $19 →References
- Salesforce. Salesforce Certified Tableau Desktop Foundations Exam Guide. Help article 005298988, help.salesforce.com. Checked 9 August 2026. The authority for every exam figure on this page.
- Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249–255.
- Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006). Distributed practice in verbal recall tasks: A review and quantitative synthesis. Psychological Bulletin, 132(3), 354–380.
The Tableau Kit walks through dimensions, measures, aggregation and dashboard building, with worked examples and flash cards. Or drag fields onto shelves yourself in Viz Drill, one chart at a time.
Open the Tableau Kit →