Video |

Tableau | Data visualisation

How to use the Makeline function in Tableau Desktop 2019.2 and newer

MAKELINE doesn't just join two dots, it bends to the curvature of the earth so your flight routes actually look right.

Tim Ngwena •10 January 2020 •1 min read •5 min watch •2.8K

Part ofWhat's new in Tableau 2019.2 Tableau Functions

Watch on YouTube

MAKELINE() takes just two arguments, a start point and an end point separated by a comma, to draw a line between them
With very large data sets you should narrow your data down first, as drawing millions of lines while generating points twice per record is extremely demanding on Tableau
Use the new 2019.2 map layers, such as streets and terrain, to add context so journeys are easier to interpret
MAKELINE respects the curvature of the earth, so long routes like flights bend to reflect the true shortest path rather than a straight line on a flat projection
Combining MAKELINE with parameter actions lets you change the origin point dynamically and replot lines to all destinations

0:00Hey, it's Tim here. In the previous video,

0:05I showed you how to make points in Tableau

0:072019.2. In this video, I'm going to show

0:10you how to make a line between those points

0:12in

0:122019.2. I'm still using the taxi data that

0:16I was using before from the ExaSoul

0:17community

0:18project from a while back. And to use this

0:21new function, you can just start by

0:23creating

0:23a field. So let's just type in make line.

0:27That's the name of the function. And like

0:30I called that before, if you don't know how

0:32to use a function, you've always got this

0:34sort of menu here on the right hand side.

0:35So if you don't see that, if you see sort

0:37of just like a normal window size like this

0:39, you just hit this arrow here on the right

0:41hand side and it expands sort of all the

0:43functions for you. You can order them or

0:45select them

0:46by sort of function. And you can see here

0:48we've got these two spatial functions, make

0:50line and make point. Make line just needs a

0:54start and an end point. So basically, we'll

0:57just take our pickup point and our drop off

1:01point. And it just needs a comma in between

1:05those two. And now our calculation is valid

1:08. Let's just call this a line and hit OK.

1:12Now

1:12when you're doing this, if you're working

1:14with a really large data set, I just

1:15encourage

1:15you just to just think for one second, this

1:19taxi data that I'm using, it's 13 million

1:22rows. Okay, so imagine drawing 13 million

1:26lines, whilst also generating points twice

1:29for each of those records. And that's a lot

1:32of sort of transformation for Tabbo. So

1:34that's

1:35going to take a long while and I don't have

1:37a powerful laptop. So let's just narrow

1:39this

1:39data set down a little bit. Remember before

1:42what I did was I brought the company name

1:44and actually focused on Donald Barnes. So

1:46let's just keep that in our data set, then

1:49remove the company and the number of

1:50records. And now we're down to 92 records,

1:53it's going

1:53to be much, much easier to work with. Now

1:55we created the line already. So let's just

1:58double click on that. And there we go, we

2:00have the lines. And the lines go from the

2:02start to the finish. And actually, if you

2:04're doing this, you probably want to add

2:06more

2:07context so people can kind of see what's

2:08going on. So let's go up to the map layers

2:10and use

2:11some of the new maps that we've got in 2019

2:14. So here's some streets. And we're going to

2:16bring the streets in and the terrain so we

2:18can start to understand how this works

2:20across

2:21the city. Okay. Now, this is a really,

2:23really powerful feature. If you're just

2:25trying to

2:26visualize sort of journeys, it works

2:28especially well, across maybe smaller data

2:31sets where

2:32you have maybe 20 or 30 routes that you'd

2:34like to draw on. And it's just a really,

2:36really

2:36nice way of sort of starting to make

2:38mapping a little bit more interesting. So

2:41we've done,

2:42we've plotted some points, and we've drawn

2:44the line from really short distances. But

2:47let's imagine that you want to plot airline

2:49routes. So let's just zoom back out. Let's

2:52get the whole map into scope here. And let

2:55's assume that we wanted to say plot the

2:58flight

2:58route between New York and London. Now, if

3:01you drew a straight line across, that

3:04wouldn't

3:04actually be the most direct route because

3:07of the projection and the curvature of the

3:09earth. A flat map doesn't quite reflect

3:11that correctly. So actually, the route you

3:13take

3:13would be more a little bit like this. And

3:15so the make line function actually behaves

3:19true to that sort of phenomenon. So let's

3:21have a look at an example. Again, I've gone

3:24to Mark Reid's profile, and he's drawn this

3:26really nice example. In fact, he's got a

3:27great

3:28workbook showing lots of features in 2019.2

3:31. I encourage you to go and take a look on

3:32his

3:33profile. And the great thing about this map

3:37is it uses London as the origin in this

3:40example

3:41because it's selected here. And he's

3:42actually using the parameter action as well

3:44in this

3:45video. So if I hit Granada, it changes the

3:49start point for the visualization, and then

3:53plots the line to all the finish points, if

3:57that makes sense. Okay, so this is a really

4:00,

4:00really good example of how that works. And

4:03notice how the curvature actually reflects

4:06the correct sort of route or the straight

4:09est line you could take if you envision the

4:11curvature

4:12of the earth. For all the flat earthers, I

4:14'm sorry this doesn't conform to your view

4:16of

4:16the world, but this is just how it is. So

4:18if I take New York, for example, if I was

4:22going from Bali to New York, I wouldn't fly

4:25across the ocean. Actually, the quickest

4:27way

4:28is to fly over the top of the earth, over

4:30the top, literally over the top, go all the

4:33way north, and then back down the other

4:36side. That's actually the shortest route

4:39you take.

4:40Depending on where you are, this sort of

4:42effect can be really, really extreme. You

4:44see, if

4:45you look at Sydney and you go into Sao

4:47Paulo, Sao Paulo to Basel, it tends to work

4:50the further

4:50or the more hemispherical the locations are

4:54. It's just really important thing to bear

4:56in

4:56mind with the make line function. It's

4:58actually observing the shape of the earth

5:01and the curvature

5:02in a really sort of powerful way actually

5:04that makes plotting things like flights

5:06more

5:07realistic so people can actually get a

5:09genuine sense of what's actually going on.

5:11If you

5:11enjoyed this video, subscribe. If not, drop

5:14a comment below and let us know what kind

5:16of content you'd like to see.

5:17[ Silence ]

Rollup

MAKELINE draws a line between two spatial points with a single simple calculation, and it correctly bends long routes to follow the curvature of the earth rather than plotting a flat, misleading straight line.

Follows on from a video on creating points with MAKEPOINT in Tableau 2019.2. Demonstrated using taxi pickup/drop-off data, then extended conceptually to flight routes.

The Breakdown

Write the MAKELINE calculation 0:00

Create a calculated field using MAKELINE, passing in a start point and an end point separated by a comma; if you're unsure of the syntax, the function menu on the right lists spatial functions and their arguments. You need existing point fields (e.g. built with MAKEPOINT) before this will validate.
Narrow large data sets first 1:13

Before plotting, think about scale: generating points twice per record and then drawing a line for each is heavy processing, so filtering a multi-million-row data set down to a manageable subset (e.g. a single entity of interest) makes the build far more workable.
Plot lines and add map layers for context 1:54

Double-click the line field to render the start-to-finish paths, then use the map layers panel to add streets, terrain and other reference layers so viewers can orient the routes within their geography. This works especially well for smaller sets of routes, such as 20-30 journeys, where individual paths stay legible.
Understand the curvature effect on long routes 2:41

For long-distance routes like flights, a straight line on a flat map projection isn't the true shortest path; MAKELINE reflects the earth's curvature, so a route such as New York to London bows northward the way an actual flight path would.
Combine with parameter actions for dynamic origins 3:22

As an example built by Marc Reid, pairing MAKELINE with a parameter action lets a user click a new origin point and have the visualization replot lines from that point to all destinations, turning a static route map into an interactive one.
Recognise how extreme this gets at longer distances 4:23

The curvature effect intensifies for near-polar or highly hemispherical journeys — a Bali-to-New York route actually goes over the top of the globe rather than straight across the ocean, and similar effects show up between points like Sydney and São Paulo.

Worth Knowing

Drawing lines across very large data sets is computationally demanding since points are generated twice per record before lines are drawn, so filter down first.
MAKELINE only takes two arguments — start point and end point — there's no extra configuration.
The curvature behaviour isn't a visual style choice, it reflects genuine shortest-path geography, which can look surprising until you understand why.

Use It When

Reach for this when you need to visualise journeys or connections between two locations, from short local trips to long-haul flight routes, and want the paths to be geographically honest rather than flat straight lines.

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.