7

There seems to be a growing trend with online applications to save data/progress automatically. In many cases where an interruption to the connection or service is quite likely, it is definitely a good idea in case the user is absorbed in the task and forgets to do so. Google Docs is a good example of a web application saving after every user action. Microsoft OneNote is another example of a desktop application doing taking the same approach.

The only reason I can think of not to save automatically is if it acts as a reference point for undo actions, such as desktop applications that the user may want to roll back changes for.

Is there are good reason why Save buttons are still used rather than status messages to inform the user when a save action has been performed?

  • Could it be that users are used to the idea and just need to change their behaviour gradually?
  • Does it have to do with performance or implementation issues that are due to technical or technology limitations?
  • Are there any usability rules or guidelines that it is breaking and no alternatives can be provided?
Michael Lai
  • 27,124
  • 17
  • 89
  • 186

5 Answers5

6

It really depends whether we're referring to Save as simply an action which persists the users data without any wider impact, or whether this could also apply to other scenarios.

If my Save action has an impact on something wider (e.g. applying some settings that cause significant changes to a system), then losing manual control over this action could be undesirable for the user.

However, I agree that in the majority of scenarios, having automatic and unobtrusive persistence of my data is a benefit.

Mr Ed
  • 61
  • 1
  • +1 - This answer takes into account auto save may not be the best solution; It depends entirely on the context of what you are doing. – Wander Jun 30 '14 at 11:20
3

Usage of auto save button is great but it should not be forces to user. It works like a charms in google doc, email etc scenario. But I'd never implement it with forms because

1: It takes control from user

2: Auto saving data will fire lots of server calls

3: With auto saving forms a false alarm might send to user that some one is storing data without submitting like credit card etc.

Again context does matter.

UXbychoice
  • 1,850
  • 1
  • 15
  • 23
2

The auto save is a great idea. Agree with you that Google docs and Gmail using the auto save are the best examples to save data/progress automaticaly. But this may not be ideal for every application. Imagine trying to play a game which autosaves every couple of minutes, it would be annoying. This would be more annoying if the saving time is more. Google docs or Gmail takes very less time to perform the auto save and hence gives a good UX.

Save button would be more suitable when you are performing a progressive task with minimal intervention of automatic tasks like saving. As said earlier, a game progress kind of task is where a save button is ideal. Performance issues definitely creep in when the data to be saved is high in volume and this can in turn contribute to bad ux again.

Context of application is definitely the key to decide to have auto save enabled or for the save button to be used. A better option would be allow the user to switch on/off the auto save option. When switched off the Save button would be available.

Vinay
  • 579
  • 3
  • 8
  • The trend in games these days are to have checkpoints that automatically triggers a save action. So I would argue that even games applications are following this pattern. – Michael Lai Jun 30 '14 at 06:07
  • @Micheal Lai, I agree. But user arriving at a checkpoint in a game tells him there will be a progression save happening. Thats one of the key ideas of having a checkpoint in a game apart from other gains. But what if the save happened at unannounced or unexpected places in game progression. – Vinay Jun 30 '14 at 06:11
1

This reminds me of a discussion a while back about the current save icon, the floppy disk. The OP thought the floppy disk was outdated and youngsters didn't recognize the metaphor. The discussion consisted of people trying to come up with an alternative.

My opinion on that matter was to remove the save button all together. Documents and such should be saved automatically. We have the technology!

To my opinion autosave is just the next step in UX. You automate an important task so users don't have to worry about saving. Never again will students writing their thesis have to fear about their computer crashing with their document unsaved.

When done right, almost invisible, I think autosave is the only way to go nowadays.

Paul van den Dool
  • 9,384
  • 1
  • 24
  • 38
  • The OP is talking specifically about web applications. What does autosave even mean in that context? Suppose the app is running in a browser and the user is typing into a textarea. Every 10 seconds the app sends the text to a server. Why? because (1) the browser might terminate (2) the browser is too stupid to implement autosave. In my opinion it is absurd for me, an app developer, to do something that the browser can implement far more effectively and efficiently. Yes the app should implement publish, or even auto-publish (as in google docs) but that's a different question. – John Henckel Aug 12 '15 at 14:34
1

I think it's your second choice explanation for why everyone doesn't save automatically. . . "It's technical or technology limitations of the specific job and code language and architecure already in place. "

Saved buttons are still used because lots of back end programers haven't learned to program auto saves well and haven't had experience with new ways of saving information that don't require clicking a button and reloading a page. Also, frequently back-end code can be years old from when the back end languages didn't encourage or support saving very well.

Steve Whetstone
  • 149
  • 3