Why 30 days?
It just occured to me that while I have a 30-day trial of my software available for download, I have no idea why it's a 30-day trial, other than that's the standard length.
I just bought a very nice bicycle, and before buying, I wasn't allowed to use the bike for 30 days, but the salesperson did encourage me to go take each bike I looked at for a ride around the parking lot. Trial time: about 90 seconds.
If you are looking to buy a car, you don't get a 30-day test drive. Some places will let you take it for a spin around the block, and some places will let you take it for 24 hours, but no one gives it to you for 30 days. Trial time: a few minutes to 24 hours
Even buying a computer, you may get to use one in the store, but never do you get to take it home for 30 days. Trial time: When I used to work at OfficeMax, about 15 minutes for the average customer.
So why do we have a 30-day standard for software demos? Try-before-you-buy is obviously a proven idea, but it seems that other industries have discovered that a few minutes or a few hours at most are ideal for trialing a product.
What do you think?
JT
Friday, May 14, 2004
Even 30 days may not be long enough to learn to use (and thus be in a position to evaluate) some new piece of software (for example, Delphi).
Christopher Wells
Friday, May 14, 2004
30 days is if anything too short.
Your analogy with a bicycle is wrong. You make a consccious decision to buy a bicycle. You don't make a conscious decsion to buy software. And of course the bicycle shop loses if you keep the bike for thirty days, whilst you lose nothing if somebody keeps your software.
Thirty days is a minimum for somebody to get the feel of your program. He may only need to use it twice in that time.
What you must do is strike a balance between having the trial period too short, so the user doesn't get to try it out properly and so just uninstalls it, and having it so long that the manages to do what he wants with it and doesn't bother to buy it.
Less than thirty days and you'll be considered cheap, and not even downloaded. It's up to you to decide if you want longer.
Stephen Jones
Friday, May 14, 2004
I agree with that the analogy was poorly placed. Software is not a tangible product, but the result of work, and is often reffered to as IP. Such simmilar items include movies and music.
Using an audio CD as a reference, you often hear a song on the radio many many times, and based on one or two songs you decide to purchase the CD. Often, based on how you enjoy the purchased CD affects future purchases.
The difference is that listening to music is a passive act, where software requires interaction. Thus you often find trial periods and reduced feature sets much like you find in the radio world (a small subset of songs are played on the radio for a short duration). Of course if you like the software, then you might purchase future versions without even demoing.
Thus without a substantial trial period, there is no way to allow users to test software prior to purchasing it.
Elephant
Friday, May 14, 2004
Blame the moon and the fact no one wanted to deal with February.
a cynic writes...
Friday, May 14, 2004
On a less flippant note - a month is just about long enough for someone to come to rely on a program but not long enough for them to finish using it having completed the job they got it for.
a cynic writes...
Friday, May 14, 2004
30 days is often too short for me. You have some time, and go for the download/eval. You get interrupted, and diverted. You have no time left for these frivolous pursuits. By the time you get anywhere close to evaluating (often quite eleborate pieces of kit), those 30 days run out already.
Unless this is something I need one time on a particular project, if I buy it after 30 days I am even more likely to buy iot after 90 days. Unless you have a util for a very specific one time use (e.g. disaster recovery, lagacy file conversion etc.), give your users at least 90 days.
Just me (Sir to you)
Friday, May 14, 2004
On a side note, I've noticed that with the 180 day Windows 2003 evaluation, at the end of the evaluation, it supposedly shuts down, and disables you from doing anything.
Well I haven't tried to log on locally or through terminal services, but all the other services are still working (email, web, ftp). . . Funny, because if it really had died, I am so hooked on it, I would have gone and bought the full deal.
Elephant
Friday, May 14, 2004
Carefull with that Elephant. You did see all the warnings about not being able to upgrade the eval to a licenced system, didn't you? Better get your data out quickly.
Just me (Sir to you)
Friday, May 14, 2004
The data is on a seperate drive. Hopefully it won't corrupt it all when it does finally stop working. I don't care too much about loosing the configuration data. The system is so simple that it won't take more than an hour to redo it.
Philo, when's my MSDN showing?!?!
;-)
Elephant
Friday, May 14, 2004
I used a product not so long ago with this 30 day trial period. The good thing: It didn't count the days since installation; instead it counted the days it was started and used. The other good thing: It allowed online updates even for the trial version. The bad thing: The online update could only be used 3 times - regardless if it was successful or not, regardless if it was even able to connect to the server. No chance to find out the correct proxy configuration...
Holger
Friday, May 14, 2004
Bicycles and cars come with warranties.
Derek
Friday, May 14, 2004
Having read various post-mortem type articles written by shareware authors, they all say that if they look at the download-time to registration-time, the vast majority of people who go on to buy the product decide to do so within a couple of days, not weeks or months.
Most of the feedback you're getting here in this thread is the customer-side, so keep that in mind. Ask potential customers how long the trial period should be and they'll say "a long, long time!". Not surprising...
Mr Fancypants
Friday, May 14, 2004
Agreed.
However if you shorten the trial period you will lose those who think you are being cheap and don't download at all. With ninetly days you will have either a smalll increase in sales or a small decrease.
Nearly everyithing depends on what the app is. As was pointed out in another thread, download to buy ratio can range from 0.5% to 20% depending on the product.
Stephen Jones
Friday, May 14, 2004
"You don't make a conscious decsion to buy software. "
Stephen, How's that?
When I buy a program, it's because I've decided a need a certain type of tool. And, like a shopping for a bicycle, I start looking for what will meet my need.
I've had fewer trial-period induced (unconcious?) decisions to buy software. And those are mostly games.
And many stores have a 30-day return policy, giving effectively a 30-day trial period on material goods.
DaveF
Friday, May 14, 2004
I should have said, "you don't always make a conscious decison to buy software."
It depends on the type of software. For example when I was designing my house I went through all the home archtitect programs before deciding on the one to use. On the other hand with utilities, like for example an Outlook or Photoshop plugin or a back up program, you may well simply be tempted by the ad or recommendation, and decide to try them out to see if the productivity gain is enough to make you want to keep them.
Stephen Jones
Friday, May 14, 2004
"You buy with emotion. You rationalize your purchase with logic."
This applies not just to shareware software, but most everything. It is clearly a generalization which doesn't apply to all products and/or all consumers, but it is true enough in most cases that it is worth keeping in mind when creating a product for sale or marketing that product.
It is common for developers to think "well, I always buy based on logic" and dismiss (emotion-based) marketing as a waste of time, but that is because we have some sort of implicit belief that everyone is like us, which is... wrong.
Mr Fancypants
Friday, May 14, 2004
Elephant: two weeks ago we had a big storm in the Seattle area that knocked out power to tens of thousands of people. Three days after the storm I was in a friend's office. He's a hardware consultant. He related to me the story of a non-profit who had called him: someone else had installed Windows Server 2003 several months back, but that person had installed some kind of trial version. When this non-profit's power came back on several hours later, their server wouldn't start because it was still in its trial mode. Apparently, you can continue to use Microsoft's trialware indefinitely - until you have to restart your server.
Regarding 30-day trials for computers: every Dell comes with that kind of trial period.
We used to give a 60-day trial period for our vertical-market software. We were sure to point out that when we sent the trial pack, the trial period started when we shipped it, not when they installed it. This put a sense of urgency into the trial. Of course, if people needed a bit more time we were able to give it to them. It was a great marketing tool.
Karl Perry
Saturday, May 15, 2004
I installed Win 2k beta 2 back in early 1999, as a DHCP server for a network. It was still running three years later wnen I popped in to see the place.
Foolishly, I decided it was time to reboot the machine. Goodbye network.
Stephen Jones
Monday, May 17, 2004
Some programs are easy to figure out, and within 15 minutes to 2 hours you can see how they work, and if they are good at the task you want them to do. For example, an image browser like AcdSee. You can test its speed and see if it has the options to work the way you want it to in a relatively short amount of time. On the other hand, some programs, like a 3D Cad program or a programming IDE, are complicated enough that unless you are very familiar with similar programs, getting to the stage that you can tell if they are any good could take weeks of USE... but if you are a busy person, and say can only play with your 3D Cad on the weekend, then it could easily be over 30 days before you know if it is better than other programs.
And, there are some programs that are not complicated, but they are extremely similar to their competition. The issue here is not that it takes a month to learn how to use it, but that the difference between it and its competitor is very subtle. And that you may be more familiar with the competitor, so it could take a very long time to tell if the new program is really better, even though less familiar. The best example of this I can think of is word processing. I know lots of people who still use WordPerfect even though they don't use reveal codes- they are familiar with it and haven't seen a reason to change to Word. For many of them, I bet it would take a trial of several months before they were familiar enough with Word to want to switch.
Dan Clarke
Friday, July 23, 2004
Recent Topics
Fog Creek Home
|