7

We plan to release an .exe file to public. The options would be to download the .exe file from the websites direct server link, so that we can provide direct download support with respective to download managers access too.

However our marketing team recommended to provide an small installer application through which the exe file can be downloaded through the installer,they suggested this way so that they can earn revenue by showing some advertisement while installer downloads and installs actual exe file.

Which is best way to provide an download option from a website; the above two methods were common in internet, any alternate to the above two methods with respective to user experience?

The first method is better in my opinion:

  • direct download from server - backup of exe file, it can be carried away easily without downloading next time

  • download manager access

The second method would be better since it provides auto installation.

Which method is better with respective to user experience? Any alternative to above two methods also welcome.

4
  • 16
    Is there a scarier phrase for a UX person than: "However our marketing team recommended..."?
    – JonW
    Nov 6, 2013 at 12:17
  • 3
    I'd say it depends who is using going to be installing the app, end users, IT guys, power users etc? Nov 6, 2013 at 12:19
  • 1
    Personally, I far prefer an installer for a download of any reasonable size. There is a lot more flexibility with an installer, and it can remove itself once it has successfully installed the app. That being said, ads in an installer are annoying and don't put them there. Nov 6, 2013 at 18:02
  • 1
    Rephrase the question: "Is it a good user experience to show our users even more ads they didn't ask for?" :)
    – DA01
    Nov 6, 2013 at 22:27

4 Answers 4

15

The User:

  1. Spends half an hour searching for the product.
  2. Finally finds it. Downloads it. Low bandwidth: it takes 10 minutes.
  3. Installs/starts it. ("Oh! I downloaded a downloader, not the application! Where is the difference anyway?")
  4. The downloader downloads. Low bandwidth: it takes another 40 minutes.
  5. The application installs itself.

Result: The user had to wait 10 minutes longer for the app an in addition, there's a downloader on the user's computer.

As a user, I probably don't get the difference or I'm afraid of doing something wrong, so I keep both programs. I want to start the program, so I click on the shortcut on my desktop. The downloader starts again.

"You have already downloaded the App".

What? Why does the program not start? It's the same as on the webpage. This should work! Oh, there's another icon with a slightly different name. This works.

So, while your marketing guys are right - they can show advertisement - I really doubt that it is worth the bad user experience.

All in all, it depends on your target group. Are they used to this procedure? Is it easy for them to get rid of unnecessary applications?

By the way: It is also possible to show the ads within the normal installer.

EDIT: From "The Inmates Are Running the Asylum" (Cooper, 2004)

There is no reason that any non-Web program—regardless of its technical details—can't have a completely invisible installation process. [...] The only reason why nonbrowser programs require installation is that this is the way programmers have always done things. Putting a bunch of questions in the install program made their programming job easier. Early browsers didn't have facilities for asking those questions, so programmers merely shrugged their shoulders and stopped asking them. If further proof were needed, programmers hardly even noticed the setback, while for many users it made the Web the easiest platform they had ever used

1
  • 4
    Additionally, a single download makes it much easier to see the total file size/how long it will take to download. Downloading a small installer hides the final download time, especially if your downloader doesn't reveal the final size either.
    – Ben Brocka
    Nov 6, 2013 at 15:51
5

When sales gets in the way of good UX, you know you're in trouble!

It's a common problem the digital world over. UX can't rule the roost entirely, as Marketing have probably already sold the idea with the expectation of bringing in extra revenue. Which, to be fair to them - from a purely sales point of view - is pretty opportunistic, and fair play.

From a purely UX point of view, there's nothing more annoying than an advert in between your initial action, and your expected end action.

The actual answer is tricky. The download of the exe itself requires a certain level of computer knowledge from your customer to then know what to do with an exe file. As I don't know your target market, this may or may not cause problems.

The auto-install is a blanket 'download for dummies' approach, which may alienate your more tech savvy customers.

I wonder if giving the option of both might be an option? Not sure how you would maintain the updating of each, but giving a direct link to the exe for experienced users would provide a better experience for them as they are in control of installing, but also giving the option of an auto-install for less savvy customers would provide an easier experience for them.

Marketing are happy as you are still entertaining the extra revenue idea, and the less tech savvy people may be more accepting of interstitial adverts if they don't have the technical know-how.

You could then keep track over a period of time on numbers downloaded and decide either way. But either way, your customers have made the decision themselves to go down that route, and people are generally more accepting of their own actions, rather than actions forced on them.

3

Ads in the installer/downloader look crummy (unless they are ads for your own very related services/applications, e.g. an ad for your paid version in the installer of the free one is perfectly acceptable).

Then there is the question of the overhead generated by a downloader/installer. For a small payload, there is little time to show ads and it has a high relative overhead: so a bad idea. For a huge payload, you need a well-implemented downloader: taking care of download aborts, allowing to save the intermediate file(s), ... If the downloader is of bad quality, your application will not be used in administrated settings (schools, companies, ...) if the admin has to download several hundred megabytes for every client computer.

But the most important point from a practical point of view: what kind of revenue do they expect to get? A user typically only runs the installer/downloader once and I suspect only a few users keep watching a long-running download/install all the time. You won't get much money that way, if any. You are more likely to lose customers even before they try your application. From a financial point of view, you're better off including ads on the website or including adware in the software itself as you get many more impressions that way. From UX point of view, neither is advisable.

IMHO, including the ads has more disadvantages than advantages, the most important one is that it may scare away customers.

1

The best use for download managers is when the download process itself is different, such as using BitTorrent to save bandwidth costs. When that is not the case, there are several disadvantages of using a download manager:

  • can't install on computers without internet access;
  • lots of engineering effort wasted on reimplementing and maintaining functions the browser already has;
  • the user ends up with an extra executable, causing confusion;
  • download size/duration is initially hidden (thanks Ben);

It has all the disadvantages of a regular installer, plus the added complexity and failure modes of downloading and running an extra executable.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.