Learn how to raise your odds of success

 
 
 

Troubleshooting Free Trials

Written by Ash Maurya

I got an interesting question the other day on why the call to action after Customer Discovery is “Signup Now” versus “Buy Now”. If you’ve got a problem and solution that resonates with a prospect, why not charge them right away?

First, there is a big difference between describing/demoing a solution and delivering on one. Customers have come to expect a try-before-I-buy period. But more importantly, you can only build a sustainable business on the basis of repeated value creation for the customer which requires time.

The most indicative (and universal) measure of building something people want is retention.

The trial period is where the rubber hits the road. You’ve made promises up to this point and even got some people to buy into them. Congratulations! Now it’s time to start delivering on them. That’s easier said then done. Not because your minimum viable product is less than perfect, but even getting customers to experience your MVP at all can present challenges that have nothing to do with the core product itself.

I’ve written about some of my early woes even getting our product installed. Technical issues aside, we had to deal with several non-obvious usability issues (at the time) that rendered the product DOA – such as presenting the user with a scary unsigned installer, or not auto-launching the product after install. The real challenge though, was that most users didn’t readily report such errors despite there being three on-page ways for getting in touch with us. They simply dismissed the application and moved on. The burden of figuring out what went wrong in all these cases fell on us.

Quantitative metrics can tell you what’s happening but can’t tell you why. For that you need to be able to talk to the customer, witness the problem yourself, or at the least collect more data.

Here are some techniques we use to troubleshoot our trials:

Getting Ready for Trials

Pick the right length for a trial period

The length of the trial period needs to balanced against providing adequate time for the customer to experience your product while still staying engaged in the buying process. This, of course, is product dependent but I feel 14 days or 30 days is about the right time window.

Manuel Rosso, CEO of Food on the Table, made a great point at the last Austin Lean Startup Meetup on problem presentations:

“If you can’t convince a customer after a 20 minute pitch, you don’t have a chance doing so on a landing page.”

I feel the same way about trials. If a customer isn’t convinced after 2 weeks, they won’t buy with a longer trial period either. That doesn’t mean you simply give up on them. More on that down below.

Freemium, done right, is a variation on “free trials”. While not explicitly time based, the user’s need for additional features and/or account limits effectively makes Freemium time based. However, too many freemium products fall into the trap of making this time window too large to where it’s just free forever – the other extreme.

Create an opt-in newsletter

Our newsletter list is currently larger than our active customers list which means that some trial users choose to stay subscribed to our newsletter even after their trials expire. This is a great way to keep them engaged. We see quite a bit of link activity whenever we send out a newsletter which has driven previous trial users back to the product and also generated new referrals as measured by “Send to a friend” links in the newsletter. The last part is actually quite interesting. I know that as a result of this blog, I get quite a few trial users that aren’t really the target audience for our products but they help drive referrals to others that are.

Get a 1-800 number

This is best gesture you can make in showing your commitment towards listening to customers. The fear of getting inundated with calls is largely exaggerated – our 800 number is tied to my phone. Most people do not call when they encounter problems. The majority of the calls I get are pre-signup questions which is a great time to do some unscripted customer development as well as start building a relationship with the prospect. I highly recommend having the founders answer the phone (at least till it’s no longer scalable). I found some of my most vocal customers this way.

Reduce Activation Friction

Get them to trial first, educate later

When it comes to copy on the landing page, I’ve found that less is more. For CloudFire, we have A/B tested a number of variations that have ranged from full-on inline demos to an illustrated list of benefits.

The version that has converted best so far is the one with the least stuff on it:

This again is product dependent but what I found in our case was that trying to educate people on how our product worked differently was a loosing battle. Telling people they could share their photos and videos instantly and without any uploading usually resulted in one of 2 reactions. They either got confused or didn’t believe us. We instead decided to focus on a more general “finished story” benefit and ushered the user to experience the product for themselves.

Here’s a recent customer testimonial we received:

“I had to take a moment & sit in awe of how easy it was to upload photos. rather than the hassle of me going through my photos/folders, clicking on photos, and unclicking duplicates, it was like magic!”

She still used the word “upload” which doesn’t bother me anymore.

Deliver value first, ask to get paid later

Another commonly used tactic for reducing signup friction is deferring the collection of payment information to after the trial. Getting customers to pull out their credit cards is the hardest part of the signup process. When you’re starting out, you have little credibility and trust which translates to “high-risk” for the customer. Getting them to invest attention is challenging enough.

Managing the trial

Always collect contact information upfront

From our activation metrics, we could tell we were loosing a number of people between download and installation but had no visibility into who they were or what was going wrong. We moved up the registration step from post-download to pre-download with no impact on overall conversions. Now we have a way to tie failed installations to an email address and reach out with an offer of help.

Send gentle auto-email reminders

People get distracted and busy during the course of the trial and gentle reminders can help get them back to your application. We’ve had cases of people who got distracted during the download process and forgot about the application until they saw an email from us a day later. The key here is to balance the signal to noise ratio. It is far better to highlight something interesting about the application the user might care about than a blatant “We noticed you haven’t used our application recently”.

Make feedback simple

GetSatisfaction and UserVoice links are fine but not used nearly enough by our early trial users. It’s far better to prompt them with an inline multiple choice question at strategic points in the application flow. One technique I heard recently from Jason Cohen (A Smart Bear) that I haven’t yet tried is to prompt the user with the question “What did you think?” when they exit the application for the first time.

Catch and report unexpected errors

When users run into problems, they don’t turn into testers. They leave. To be able to still learn from their experience, we have had to build code to catch unexpected errors and report them back to us.

Create a dashboard of dashboards

This is the central nervous system with which we manage trials. We have aggregated all the tools we use ranging from Google Website Optimizer to subscription management software to server monitoring under one web application shell that keeps everyone in the company on the same page.

More specific to trials, we have a custom report that shows us the newest users in the last 30 days. From here we can see at a glance where trial users stand and how far into the funnel they are. We can also drill down into their account and get system and application level diagnostic information which has been invaluable for recreating issues. Getting in touch with a trial user (over email) is one click away.

Getting Trial users to talk to you

As I’ve mentioned before, this is usually the hardest part. We get a decent percentage of people respond to our emails and use the 800 number but the majority don’t respond at all. Towards the end of the trial, we send out a final email with an offer for either a 1yr subscription or an Amazon gift card in exchange for 20 mins of their time on the phone.

Keep Testing Qualitatively

For every non-obvious usability issue uncovered using the methods above, there are x that remain undiscovered. There is no better way to uncover these than running usability tests with new users. We run at least 3 usability tests a week.

What do you think?

A trial is a critical window of time when a prospect grants you limited attention and permission to engage them. Properly done they can be a goldmine opportunity for learning, but they can just as easily be fumbled.

Do you have any other techniques you use to run your trials effectively?

  • http://www.marketingtactician.com/ Scott Harris

    Great post, Ash. I especially agree about using retention as a key metric for your business. One of the best things about software as a service is you are better able to analyze your retention and whether something you've changed is improving or diminishing your user's experience.

  • http://twitter.com/tonb Ton Bil

    For those who are into #leanstartup like me, this detailed experience with trials is good info! (unsure of tweet will contain URL, will update)

  • mattharrell

    Great post indeed. Ash, or anyone else, do you know where I can find various examples of pricing strategies that work well for trials and freemium tools? Has anyone done a post or compiled anything that's out there? Any help would be great!

  • http://www.ashmaurya.com/ Ash Maurya

    Matt –

    I did a post earlier in the year on my experiments with pricing: http://venturehacks.com/articles/pricing-experi

    Pricing is more art than science and the right strategy varies by product. There is a link in the post on a great pricing strategies handbook by Neil Davidson.

    Best,

    Ash

  • http://artvankilmer.wordpress.com/ Arthur Klepchukov

    Ash, thanks for the post. I'm glad my email inspired such a solid reply. This really helps with the stage I'm at. I knew that “The most indicative (and universal) measure of building something people want is retention” but had to painfully learn “Deliver value first, ask to get paid later”.

  • http://www.facebook.com/profile.php?id=1251271108 Kaleb Pederson

    > … She still used the word “upload” which doesn’t bother me anymore.

    The above is often key. As businesses, our job isn't to correct the users' grammar nor their understanding of what's really happening behind the scenes — it's to help them accomplish the goal that led them to our business. And, ideally, to do so in a way that's pleasurable and easy for the customer to understand.

    @Ash – Thank you for helping us keep the true goal in mind.

  • Pingback: Software Trials: Fehlervermeidung bei kostenlosen Testphasen - //SEIBERT/MEDIA Weblog()