* The Business of Software

A former community discussing the business of software, from the smallest shareware operation to Microsoft. A part of Joel on Software.

We're closed, folks!


» Business of Software FAQ
» The Business of Software Conference (held every fall, usually in Boston)
» Forum guidelines (Please read before posting!)


Andy Brice
Successful Software

Doug Nebeker ("Doug")

Jonathan Matthews
Creator of DeepTrawl, CloudTrawl, and LeapDoc

Nicholas Hebb
BreezeTree Software

Bob Walsh
host, Startup Success Podcast author of The Web Startup Success Guide and Micro-ISV: From Vision To Reality

Patrick McKenzie
Bingo Card Creator

More or Less Reminders During the Trial Period?

A couple of weeks ago I released a new version of my main product with (what I think) is a more elegant purchase reminder system for users during the trial period.  Basically it prompts the the user on software start up to purchase and allows them to dismiss the reminder for 7 days.  In seven days it reminds them again and so-on.  When they reach the last 7 days of the 30 day trial period the reminder pops up and can be dismissed for 24 hours.  When the trial period is up the reminder appears whenever the software is started.

Previously the software reminded the user each and every time they started the software. 

So, I am wondering, what's your belief on which of those alternatives is likely to lead to more conversions?  I realize some of you are just going to say "test test test" but I'd like some educated opinions.  Just as a matter of interest (and I realize it's probably just a statistical glitch) conversion in the last week has been.....well, frankly, terrible.  I can't really afford to sit on this for a month or more so if the general consensus is that the new system is much worse then I'll just roll back to the old method.
Mark Nemtsas Send private email
Monday, July 22, 2013
The system you describe sounds reasonable to me.
Scott Send private email
Monday, July 22, 2013
I find your reminder scheme fine. I would just add any kind of incentive on the last reminder to get a conversion.
Jose Send private email
Monday, July 22, 2013
"conversion in the last week has been.....well, frankly, terrible"
The thing about reminders is that 95% of your trial users need more than one reminder.  Please don't give up on your system and increase the frequency.  If no one converts after 3 or more reminders, you have other problems.

By the way, in most things 3 reminders seems to be the sweet spot, as long as you space them out so your prospect/delinquent account has time to feel guilty about not doing anything.  Be generous with your time allowance after the first reminder, and cut that allowance in half after the second reminder.  For the third reminder, give less than 24 hours before you send a notice that the trial is over/the account has been closed.  Then give them a week to crawl back asking for grace, but don't send out any notices.  After that, strike them off your contact list.  Begging doesn't help sell or collect anything.
Howard Ness Send private email
Monday, July 22, 2013
Thanks for your replies so far.  My expectation when I implemented this new reminder system was that I'd push the purchasing decision back in the trial period.  But hopefully with the better (and less annoying) reminders the number of conversions would actually increase.  Right now the most reminders any person in the trial period has seen (assuming they dismiss it for 7 days) is 2.  I expect the next 7-14 days should see people purchasing the software back at the levels I was seeing previously.  That's if I can hold my nerve.

As a footnote, am I the only one who finds playing with this aspect of a reasonably successful product entirely nerve wracking?  The system I had in place previously worked fine, but it was ugly and I knew rather annoying.  The thought of changing it was truly scary.  I tested the heck out of this new system in dozens of different VM's and played with the system date endlessly to make sure it all worked smoothly.  It's certainly one of the most thoroughly tested minor releases I've ever had.  I'll go back to chewing my fingernails and wait for more responses (and some sales of course).
Mark Nemtsas Send private email
Tuesday, July 23, 2013
Yeah I worry a bit when I try new things too, but once you get some results in it's less nerve wracking as you put it.

Trying different things is a good approach.

Be aware that with small sample sizes results are highly unreliable though.
Scott Send private email
Tuesday, July 23, 2013
The problem with startup reminders is that they get in the way of what you are trying to do. Some might say this is a good thing. I'm not so sure.

As an example, I rarely update FileZilla because it prompts me just when I'm trying to do something else. If it left a non-intrusive, non-modal update window to the side or behind the main window then I would process it when convenient.
Adrian Lock Send private email
Tuesday, July 23, 2013
How do you keep track of the date, though, unless you're checking online?  Any offline date can be bypassed and reset by running the exe under Sandboxie or such.
PSB136 Send private email
Tuesday, July 23, 2013
A colleague once told me about his brother's shareware that showed, instead of a reminder, a Happy Day discount to all trial users on one randomly selected day of each month. IIRC half of all sales were on that day.

I also recall reading about a more sophisticated strategy you could implement on top of your three reminders scheme:

On the first reminder, display a 20% coupon that is valid for seven days. Then on the second reminder, display a 15% coupon, and on the third one, a 10% coupon. The study said most people would either use the first or second coupon, or not buy at all, and that the overall revenue increased too, but I've forgotten by how much, sorry.
Dmitry Leskov Send private email
Tuesday, July 23, 2013
I like that idea
Bring back anon Send private email
Tuesday, July 23, 2013
"Any offline date can be bypassed and reset by running the exe under Sandboxie or such."

The answer to this is not to worry about it. What percentage of users are technically minded and going to go to this trouble over and over? 1%?
Scott Send private email
Tuesday, July 23, 2013
The vast majority of sales from trials happen in the first few days. Choosing not to remind the user that your software is not free in those early days is a huge mistake.

For the majority of people who try software (excluding techies like us, who really, really don't count), a reminder screen when the software launches is not a problem. Might be a problem if the nag screen was attached to freeware (winzip), but when you're deciding whether to buy or not, a reminder that that decision needs to be made is anything but a bad thing.

Our 'reminder' screen has very visible links to the online help, video tutorials, reviews page, and support email. Whenever someone clicks one of those links, we see from the URL parameters how many days remain in their trial. On average, sales happen about 3 days in.

The best change we made to improve trial conversions was to switch from a 10 consecutive day trial to a 10 days of use trial. Trial periods longer than 10 days are unnecessary, as they push the buying decision so far back that many users have forgotten about your software by the time the trial expires.

Guys, most customers decide on Day 1 to buy or not to buy your software. Don't make it any harder for them to give you their money.
Marlee Ammon Send private email
Wednesday, July 24, 2013
I think I disagree with Marlee slightly. If I am just trying out a new piece of software that requires daily usage, I'm going to take it through its paces. If, however, I am using it for a singl epiece of functionality, I will buy it once I test that the functionality works as I need it to.
Bring back anon Send private email
Wednesday, July 24, 2013
I *might* say yes right away.  I definitely can say no right away.

If something survives enough keep-or-toss questions that I have gotten something useful done with it, I will probably get it.

I usually have alternatives so I can afford to be selective.  I want to eliminate obvious mismatches as soon as possible.  It takes longer to determine that something is a good match.


Gene Wirchenko
Gene Wirchenko Send private email
Wednesday, July 24, 2013
By my experience most people either buy after the 1st reminder or do not buy at all. After some tests on our software we introduced more strict trial limitations. It's sounds odd but it worked and led to more sales.
Marshall Trevis Send private email
Friday, July 26, 2013
Marshall, you might already have people having enough data to decide to buy.  I can assure you that if the trial does not give me enough opportunity to check out the program, I will toss it fast.

I assume this means that there is a fine balance in there somewhere.


Gene Wirchenko
Gene Wirchenko Send private email
Friday, July 26, 2013
There have been a large number of discussions here over the years on trial strategies.

Often there is strong advocacy for the method a person is already using, but it seems obvious that not all strategies are going to be optimal for all sorts of software, or for all target customers.

To reboot a bit, perhaps we could start over by seeing if we can agree on any fundamental principles:

I would propose that the process fails in cases where a user deletes the trial. Generally in these cases the user deletes the trial because he found another solution, such as:

1. solving his problem by hand, or
2. finding a free program, or
3. perhaps he even moves on to a competing solution with similar features which he pays for

In all of these cases, expiring a trial leads to the user either:

1. buying the program, or
2. looking for another solution (such as in the previous list), or
3. giving up.

Discussion usually revolves around an assumption that #1, buying the program, is going to be the result of expiring a trial.

I suspect that experimental data shows this is not the case in most expirations.

It would greatly behoove authors to understand what actually happened after the trial expired. I postulate that #2 and #3 are actually the most common cases, in which case the process has failed.

Solutions other than completely expiring the trial are thus worth considering and testing.

Some solutions that have been proposed and which are used by known successful companies are:

1. Allowing renewals of the trial by requesting one from the company, inconveniencing the user somewhat but not so much they delete the program.

2. Trial expiration results in lost functionality, such as saving files, or limiting the number of records that can be stored. But opening is still supported so users are able to access the files they made during the trial, they just can't save any further work with them. This one is extremely sticky and ensures that the user will never delete the crippled trial if they need it to open and view their own creations.

In my opinion and observed experience, many purchasers are cautious daters who take a very long time to make a decision. Some may have gotten burned on bad software purchases in the past. I have had sales take place several years after initial customer contact. When the door stays open, the customer thinks about you and your product from time to time. The chance they will purchase increases with the more time they spend which means more trust.

What doesn't work is unlimited function unlimited time non-nagging trials. Some sort of reminder or function limitation is needed to get them to periodically think about their obligation to purchase as they continue to make use of the product, and the advantages it will bring, such as being able to save files again.

Having feature limited software sitting on their hard drive and being regularly used is ADVERTISING.
Scott Send private email
Saturday, July 27, 2013
On the advertising, many times I have seen positive reviews and recommendations made about my software by people who are currently enjoying a long term disabled function trial.

These recommendations not only improve my google ranking, but they absolutely lead to sales.
Scott Send private email
Saturday, July 27, 2013
Personally I think the trial reminders are waste of time. Super waste of time is to invest in smart reminders. If software is worthy and needed, users will buy it and you will get happy customers anyway. If software is not worthy or not that needed, the reminders will be laughed at. So what are these reminders good for? To make sure that forgetful users don't forget to pay you? Come on.. Just make a note that "this is a trial software" when launched for the first time, and "thank you for trying" when expired. This is contrary to the used-car-salesmen-approach, you must put a call for action like BUY NOW in blasting red on Sundays after lunch and on Mondays before breakfast.. And that will improve the conversions they say. See if common sense allows this :)
Dima Send private email
Saturday, July 27, 2013
I believe that Scott has it on most trials not leading to purchases.

If someone checks out just two packages, then 50% of the trials do not lead to a purchase.  If he checks more, then it is even worse.

If I am looking for software to deal with a certain area, unless I have a firm recommendation, I will check more than one package (assuming they exist).  Even with a firm recommendation, I may check others.


Gene Wirchenko
Gene Wirchenko Send private email
Monday, July 29, 2013
This is why I always say "good software sells itself" (and get laughed at).  If someone wants your app, they will buy it.  Simple as that.  No amount of reminding will make bad software sell.
PSB136 Send private email
Friday, August 09, 2013
A a follow up to this as it's now over 30 days since I implemented the new reminder system.  Conversion is back to usual levels and possibly a little higher (about 15-20%) but it's too soon to say if this is an on-going thing or not or just another blip.
Mark Nemtsas Send private email
Sunday, August 18, 2013

This topic is archived. No further replies will be accepted.

Other recent topics Other recent topics
Powered by FogBugz