Data tables - what's column should be first?

Question

When creating a table with a lot of rows and a number of columns, what is the recommendation for which column should be first and most important?

An example scenario for the data table would be a table displaying the status of a lot of different servers a user has. Different columns would the "Status" of the server (online, offline, has a problem), "Server Name" which would be the identifying column, and a few other fields associated with describing the server such as its IP address, date installed, description, etc. There will also be an "Actions" column that will be a drop-down menu of actions the user can take such as [ Edit | Restart | Delete ].

I'm debating if the first column should be the STATUS column or the SERVER NAME.

So, the proposed column ordering would be:

[Checkbox Column] | Status | Server Name | [Actions Menu] | Other Field | Other Field | Other Field

Or, does it make sense to have Server Name first followed by Status?

Answer 1

It depends on the weight you want each to have. The ones you put further to the left, the more likely and easily people will see them as we read left to right.

In your particular case, I'd likely do status and then the server name because if a server is down I want the admin to immediately notice. I'd likely have it color coded as well, green or black for doing alright, red for down or needs attention. Finding the server name once they know the status is easy.

Answer 2

Here's a rule of thumb on data tables. For people who read from left to right, you assign priority from left to right because that's how they read. (Note: it'll be reverse for right to left languages e.g. Arabic)

Columns used for ID & scanning gets higher priority

Because we read and scan information using the "F-pattern", you want the identifying columns to be on the left. If this is a basic information look-up table (there's no particular piece of info that user will be scanning for to determine action). Then it's usually the Name or ID column that's placed on the left.

Sometimes, user are looking for a specific piece of information to identify rows of interest. Like in your example with Server Status, the server admin needs to know right away if there's issues with a server and the status allows them to track that. I'm just going to call these as scanning columns. Scanning columns takes precedence over ID columns and so if you have them, they'll be shown ahead.

Lastly you have the selection columns. These are checkboxes, radio buttons and actions (buttons, dropdown menus). These typically goes on the two sides of the table so it's easy for users to spot. If selection is important, you may put checkbox/radio button to the left. The less important items tends to be capped on the right.

Answer 3

This problem has many parallels with database administration interfaces, where there are often many columns and rows.

Both php myadmin and navicat solve this conundrum simply by allowing the user to configure column order to suit their own particular use case.

Your domain is similar, you have technically adept users; detailed and complex technical data; and a myriad of different potential use cases (and potentially quite a few more columns), so I would advise allowing the user to configure column order for their own convenience.

Answer 4

On the one hand server name/ip is the ID of the server and on the other it looks like the most important property is the server status. Not a clear answer here I think but just note you can combine both attributes to a single column. This is usually done by putting the primary information on top and the secondary below it, with smaller font and lighter color.

For example:

^{download bmml source – Wireframes created with Balsamiq Mockups}