Video |

Tableau | Data prep | Tool strategy

Tableau Prep input step enhancements | New in Tableau 2023.1

Pre-filtering your dates at the input step in Tableau Prep 2023.1 means a smaller subset and a far more performant flow long term.

Tim Ngwena •20 March 2023 •2 min read •5 min watch •2.5K

Part ofWhat's new in Tableau 2023.1 Tableau Prep Builder tutorial Jan-2020

Watch on YouTube

You can now hold shift to select multiple fields at once in the input step and right-click to remove them
Removing fields applies a change to the step that you can view or undo via the changes column
You can pre-filter at the input step using either a calculation or relative dates by right-clicking a date field
Filtering dates at input rather than in a later clean step gives you a smaller data subset and a more performant flow long term
Salesforce data, with its hundreds of date fields, particularly benefits from input-step date filtering

0:00In Tableau 23.1 Tableau have improved a few

0:03capabilities with the input step in Tableau

0:05prep.

0:06If you want to follow along just open the

0:08sample flow and you open up Tableau prep

0:09and it will load

0:10up this particular flow. If I go to the

0:12orders input here at the top and I just

0:15bring this up,

0:16you might have noticed in the last few

0:17release that Tableau prep has been sort of

0:19getting a

0:19little bit of an enhancement in terms of

0:21the user interface, a few changes here and

0:23there.

0:24And but fundamentally in this release one

0:26of the things you can now do, this used to

0:28drive me

0:29absolutely crazy. You can now hold shift

0:32and select multiple items at once. So I don

0:36't know

0:36what I actually did there. I did something

0:38very different there. Let me just try that

0:39again. Let's

0:40select all today, hold shift, select

0:42customer id and it did. I did something

0:44then it selected

0:45the whole column. So maybe that's a small

0:47bug but nonetheless you can select a bunch

0:50of fields,

0:50right click and you can remove them. And

0:52notice when you remove them it crosses them

0:54out and if

0:56you're not familiar with Tableau prep,

0:57anytime you do anything like that it

0:59actually applies a change

1:00to the step. So you can see here there's a

1:02change. Let's go ahead and take a look at

1:04that and when we

1:05select that you can see the items have been

1:07removed. And if you want to undo that you

1:10can

1:10of course just go here and hit the x and it

1:12will basically remove those changes. The

1:14other way you

1:15could do that, let's say you select four

1:17and you only want to remove one. Let's go

1:19ahead and remove

1:19all four again. If I move my face to the

1:22left, if you go to this changes column what

1:24you can actually

1:25do is you can actually undo this change.

1:27But what this change does is it remembers

1:29everything that

1:30was done in that step. So if we right click

1:32and remove, notice it removes all the other

1:34four as

1:34well. That's not what I was expecting.

1:37Honestly I'd expect to just remove that one

1:39item from that

1:40change if that makes sense. But nonetheless

1:42maybe that's something coming in the future

1:44.

1:44So it's a really nice quality of life

1:45improvement. The other improvement in this

1:48release is with

1:49date. Actually if you go to the order date,

1:53what you can do is you can right click and

1:56you can

1:56filter. And there's two ways you can filter

1:58. You can do a calculation which allows you

2:00to type

2:00whatever calculation you want. And then if

2:03you right click order date, go back to

2:05filter, you can

2:05actually use relative dates. This is

2:07allowing you to essentially pre-filter

2:09based on a bunch of dates

2:11at input. Now the reason this is is because

2:13in essence up until now Tableau Pro, when

2:16you connect

2:16to any data source, let's say you connect

2:18to Salesforce. Salesforce is the big talk

2:19of the

2:20moment. Now Salesforce has literally

2:22benefits from this release in so many ways.

2:24If you've ever

2:25connected to the opportunity table in Sales

2:27force, it's just full of absolutely hundreds

2:30of useless

2:31fields. Some fields will never get used

2:32because an organization doesn't actually

2:34use them.

2:35And the other thing is there's just

2:36hundreds of dates. So like whenever you

2:38look at any Salesforce

2:39data set, there's just dates everywhere.

2:41There's history day, change day,

2:43opportunity creation date,

2:44last edit date, last modified date, last

2:47access date, just date galore. So depending

2:51on your

2:51criteria, what you want to be able to do is

2:53to filter to a specific time range. And

2:56what that

2:56does is that later on in the flow when you

2:58're working with your data, because you

3:00filtered at

3:01the start, you're working with a much

3:03smaller subset. So what this change has the

3:05effect of

3:06doing is it makes it easier for you to have

3:08a more performant flow long term because

3:11you're already

3:12filtering everything out at the very

3:13beginning, which means that Tableau prep

3:15doesn't consider it

3:16in the flow as it's processing your data

3:18set up until this change, people would have

3:20to essentially

3:22put in a clean step, which is what you've

3:24got here, the dual the data prep in the

3:26clean step,

3:26and then from then on, they'd go but at

3:28that point, the data is already all coming

3:30into the

3:31flow and it's already sort of slowing you

3:33down. So quick one 75% of people watching

3:36the video right

3:37now, at least in the last 90 days on

3:39subscribed over the last year, it's 80%

3:42even bigger. So I'd

3:44really appreciate your help hitting that

3:45sub button. So we can push forward.

3:47Unfortunately,

3:48the subscriber count is the measure that

3:50some organizations look at. And those

3:52organizations

3:53are going to help bring this content to

3:55another level. So I really want your help

3:57to help push

3:57this channel to the next level. Let's carry

4:00on. This is a really nice quality of life

4:02improvements

4:02specifically for the input step. I'm now

4:05going to make another video about the

4:07column row order. I'm

4:08going to do that as a separate video

4:10because it's it's a bit of a like a

4:12philosophical feature as

4:13well as this the new feature we should talk

4:15about. So go ahead and check out that video

4:17. And yeah,

4:18check out all the other videos that I've

4:20made on 23.1 and many other releases over

4:22the last

4:22couple of years. I'll catch you in the next

4:24one.

4:25Transcribed by https://otter.ai

4:35[ Silence ]

Rollup

Tableau Prep 2023.1 adds two quality-of-life improvements to the input step: multi-select field removal and the ability to pre-filter dates before the flow even starts, which keeps your data smaller and your flow more performant.

Demonstrated on the sample flow's orders input step in Tableau Prep, with Salesforce data used as the motivating example since it typically has hundreds of date fields.

The Breakdown

Shift-click to multi-select fields 0:24

You can now hold shift to select multiple fields at once at the input step, then right-click to remove them all together, rather than removing fields one at a time as before.
Removed fields show as struck-through 0:40

Once removed, fields appear crossed out in the input step, giving you a clear visual record of what's been excluded.
Changes are tracked and undoable 1:15

Any removal applies a change you can inspect in the changes column and undo by clicking the x. Note that undoing works at the level of the whole change, not per individual field — removing four fields as one action means you undo all four together, not just one.
Pre-filter dates at the input step 1:44

Right-click a date field at input and choose filter to either write a calculation or use relative dates. This lets you cut down date-based data before it ever enters the flow, rather than filtering later in a clean step.
Why this matters for messy sources like Salesforce 2:20

Sources like Salesforce's opportunity table are full of near-duplicate date fields (creation date, last modified date, last access date, and so on). Filtering to the date range you actually need at input avoids carrying all that noise through the flow.
Early filtering means a smaller, faster flow 2:56

Filtering at input rather than downstream means Prep works with a smaller subset from the very start, so every subsequent step processes less data — giving you a more performant flow long term instead of loading everything in and filtering only after a clean step.

Worth Knowing

Undoing a multi-field removal change reverts all fields removed in that action at once, not one field at a time — Tim notes this felt like unexpected behaviour and may change in future.
Shift-select can sometimes misfire and select an entire column instead of the intended fields, so double-check your selection before removing.

Use It When

Reach for input-step date filtering whenever you're connecting to a wide, date-heavy source like Salesforce and only need a specific time range — filtering there instead of in a later clean step keeps the whole flow lighter and faster.

How this Rollup was made provenance & method

A Rollup is drafted by AI from the video's transcript, then reviewed and edited by Tim. Everything used to produce this one is listed below — the model, the exact prompt, and the source video — so the process is transparent and reproducible.

Transcription: On-device — NVIDIA Parakeet v3 for recent videos, OpenAI Whisper large-v3 for earlier ones. The transcript never leaves the machine or gets published.
Drafting: Claude Sonnet 5 in the cloud, from that transcript.
Prompt: The exact Rollup prompt (v2) — the full system prompt, unedited.
Source video: Watch on YouTube
Drafted: 5 July 2026 at 09:38
Reviewed & edited: 5 July 2026 at 09:41 · by Tim Ngwena

Model + prompt + video is everything you'd need to recreate a Rollup like this yourself. The one thing we don't share is the transcript.

Rights. The video and its transcript are the property of TN Media Ltd. Unauthorised use or download is prohibited. © TN Media Ltd.