SmoothSpan Blog

I just read a notice in Buffer, the software I use for Social Media sharing, announcing that as of July 2017, LinkedIn would be closing down API access to LinkedIn Groups.

Time to give up on LinkedIn for marketing.

Here’s the problem:

I use LinkedIn by cross posting my latest content there.  It’s easy to do because I can automate it via Buffer (or Hootsuite, WordPress Plugins, and a variety of other methods). It’s a good thing too, because overall, LinkedIn is not a very effective social platform.

In fact, for the last 30 days of traffic on my business site (CNCCookbook.com), Social Media has only been responsible for about 3.5%.  And of the Social Media, here’s the breakdown:

Pinterest:  6597

Facebook: 2504

LinkedIn: 1365

Reddit: 583

Twitter: 146

These are pretty tiny amounts of traffic.  By comparison, my post popular article got 31,741 views.  That’s just for one article and I have several thousand on the site (I’m a content marketer!).

Over on my entrepreneurship blog (BobWarfield.com), I have a series I call the “Top 15 Marketing Masters.”  It’s an analytical benchmarking series where I look at the Top 15 marketers in aggregate and also individually.  I have chosen them based on the traffic they generate to their own websites because after all, if a market can’t market themselves, what kind of marketer are they anyway?

On average, the Top 15 get just 7.83% of their traffic from Social Media.  Surprised?  Despite all the hype around Social Media, you shouldn’t be.  After all, it’s dangerous to bet your business on someone else’s Walled Garden.  You can think you’re a rancher and wind up a cow in no time.  These Top Marketers know that.  In fact, the highest share of Social Media traffic is 19% for Gary Vaynerchuk.

If we drill down and look at the relative importance of the different social platforms to these Top 15, here’s what it looks like:

Facebook: 54.58%

Twitter: 21.47%

LinkedIn: 12.32%

YouTube: 5.4%

Other: 6.22%

They’re not doing much better than I am with LinkedIn.  So, unless they’re large enough that’s it’s worth it to have someone manually posting to LinkedIn Groups, many of them (as well as countless other businesses) will just give up on LinkedIn.

Why is LinkedIn closing the API?

I’m sure they have their reasons, but this sort of change is exactly why you never want to bet your business on someone else’s Walled Garden.  I have a whole post that tells exactly why this is a problem and how to avoid it here:

[ Beware of Web Services: Are You the Cow or the Rancher? ]

Is there anything we can do about this?

Well, it’s possible some outfits will figure out how to hack LinkedIn and post automatically anyway.  That’s an arms race as LinkedIn works to block them.  In the end, not reliable and probably not worth the trouble.

I sent LinkedIn Feedback telling them exactly what I thought of it, and you could too.

[ Here’s a link to a page telling you exactly how to give them feedback.  ]

If your business is in the same boat I am–will quit using LinkedIn if you can’t automate–tell them exactly that.  Maybe they’ll relent.

I doubt it.  Walled Gardens seldom get better, they just get worse.

Recently, I needed to purchase an SSL Certificate to get a Norton Seal on my web site.  Such seals signify the site has been swept for malware and generally give visitors a sense of confidence that the site is legit and safe.

Originally, the seal could be purchased separately (and more cheaply) from the SSL Certificate, but Symantec changed all that about a year ago.  If you run an e-commerce site like my CNCCookbook, such seals are a tax–you can’t live without them as they can significantly affect your conversion rates.  I ran a test a few years ago where the seal swung conversion by 30%!

So, when my old seal ran out at the end of 2016, it was time to renew with an SSL Certificate.  Seems like a tried and true enough sort of thing, so I signed up.  The site assured me all would be operation within 2 business days.  Here we are business day 5 into the process, and still no seal.

The notes said they’d be in touch to confirm I was really associated with the site, but I could find no evidence of either email or a phone call.  I figured I’d better get onto their Customer Service system and ask whether they needed anything from me.

That’s where the trouble started.

I didn’t have an account on their support portal.  Like so many companies, this requires a separate sign up from being a customer.  Talk about not having a 360 degree view of the customer.  Seems inexcusable for a company the size of Symantec, but unfortunately, they’re probably also at that size where they start thinking customers are just a burden to be minimized once they’ve got your money.

Anyway, I signed up for the account so I could submit a ticket and my first challenge was that submitting a ticket requires me to speak some Asian language–Chinese or Japanese, I presume:

Apparently Symantec expects me to speak an unknown language to file a ticket…

Notice there is no obvious way to change the language on this page, and there’s no way to tell what the prompt is even asking for.  Some number, I’d guess.  Perhaps it is a support ticket ID?  I’ll probably never know.

And why ask in whatever language this is?  Are they trying to guess the language based on my IP?  If so, everyone else seems to think I’m in Oakland, California.  So far, I am seriously not impressed with Symantec’s Customer Service chops, unless, of course, their goal is to prevent me filing the ticket.

I’m not so easily deterred though.  They did screw up their evil plan by providing a “Submit Case” link.  So I clicked that.  Things are a little better on the new page:

Let’s re-enter all the information they already have!

Well this is awesome, they want me to re-enter all the information they already have because I’m a customer.  Nice job, guys!  I see you’re still trying to understand the whole CRM thing.  Maybe you should get some help with that?

Of course, you have little choice but to comply.  So I entered mass amounts of information, and eventually I’m down to telling it what product and version I have.  Well, the product is “SSL Certificate” and they have no version number so far as I know.

Guess what you get back from that?

Say what?

Say what, fools?  SSL Certificates don’t have a version number?!??

Eventually, I discovered you can circumvent this if you submit a “Non-Technical Support Issue.”  But, Heaven help you if you can’t produce your order # or customer serial # if you go down that path.

Now I wait to see if any meaningful help comes back from this black box.

Symantec: world’s worst support?

They’re definitely on my list for it at this time.  I will be contacting the reseller too in order to see if they can expedite.  I’ll report back on that as a postscript to this article.

Every 2-3 years, I manage to wear out a keyboard.  With nearly 3000 articles written for my CNCCookbook blog alone, 5 other blogs, and countless emails responded to, I guess that goes with the territory.

Normally, I don’t think too much about what keyboard to buy, I just head over to Staples and get whatever they have.  This time I decided to order online, and while I was at it, I checked around for some reviews.  Was there some ultimate keyboard that wasn’t too expensive I could try?

It turns out there is an ultimate keyboard, and if your day involves lots of typing, you need to check one out.  I hate to be suspenseful, but I put a whole keyboard review over on CNCCookbook, and even with this premium keyboard, I’m not going to retype it all again here.

Click the link, head over, and check it out.

I just put the finishing touches on one of my biggest content marketing efforts to date, an ambitious article called, “Beginner’s Step-By-Step Guide to Making CNC Parts.”

All told, it took almost two months of part-time effort and was as much work and words as a small novella.  I don’t expect to win any literary prizes with it, but I do expect it to help a good many CNC (Computer Numeric Control, the field my business, CNCCookbook, is serving) Beginners to launch their journey into the world of Robotic Machine Tools that make things for you.

Writing the article has left me feeling reflective about the CNCCookbook journey.  It’s become one of the biggest if not the biggest CNC-related blog on the Internet.  I’ve accomplished marketing goals all by myself that a lot of top marketing people would love to recreate.

CNCCookbook has been a magic business for me in the magical world of CNC.  We live in an age of 3D Printing, which gets most of the Hype, but also of CNC in general.  Computer-controlled machine tools that are even more sophisticated than those that put Men on the Moon are available not just to businesses, but also to hobbyists and small businesses operating out of their garages.

I recently interviewed Zach Kaplan, the founder of Inventables, for the CNCCookbook blog. In the inverview, Zach remarks that there are some 300,000 manufacturers in the US today, but he thinks within 10 years there will be over 3 million manufacturers.  This amazing growth will be fueled by the power of these entry-level CNC machines such as the X-Carve CNC Router that Zach’s company, Inventables, sells.

I think Kaplan is probably conservative, and that we’ll get to that 3 million manufacturer mark much sooner than 10 years.  We live in an unprecedented time of opportunity with desktop CNC to help us make products and the Internet to help us market them.

I’ve interviewed many small CNC businesses that got started from nothing and are doing very well.  A great example would be the little Iowa company of one Brad Martin that makes bottle openers in the shape of grenades.  It’s called Tactical Keychains and affords Martin a nice living where he is growing steadily and is his own boss.

X-Carve is VC-Funded, while its chief competitor, Carbide3D (another outfit I’ve interviewed from time to time) was crowdfunded via Kickstarter. My own company, CNCCookbook, was created with no external capital, just my own sweat equity.  Those are really the rungs on the evolutionary funding ladder–VC, Crowdfunding, and now Bootstrapping.

Lately, I’ve noticed motion away from conventional VC by a number of savvy entrepreneurs.  They’ve realized that the economics are not that great when they take Venture Money.  Certainly nowhere near as good as the economics for the VC’s themselves.

Companies like Atlassian, Github, SurveyMonkey, and Mailchimp have shown that you can grow a company quite large without ever taking any capital.  VC Jason Lemkin writes that the cost is 4 more years to reach a given size.

Personally, I think 4 years is a pretty small price to pay for the lower risk and superior economics that are possible when you bootstrap.  I’ve had plenty of experience with Venture Money–CNCCookbook is my 8th Startup and VC’s were involved with the other 7.  I wish I’d embarked on CNCCookbook and the Boostrapping path 10 years ago.  I’d be that much further ahead on a journey that looks like it has no ceiling.

Today, I take home more cash than I have taken home from any of the VC companies.  That doesn’t count stock option money, but it’s still pretty darned good when you consider I’ve been an executive for two companies that made it to pretty decent sized public firms.  I was able to do this all by myself–I have a few part-time employees, but have gotten here largely through my own efforts.

I owe my success to my ability to write software, but just as much if not more to my ability to do Content Marketing.  It’s been my magic bullet, and it works something like this:

1. I give away valuable content about CNC, the market I’ve chosen to be in.
2. People find the content via Social Media, Search Engines, and Referrals as other sites link to CNCCookbook.
3. They visit, consume the content, enjoy it, and pass on the word.  I can’t claim it’s viral, but it’s pretty darned good.
4. As they become regular readers, they’re exposed to content about the kinds of CNC problems my software solves.  It’s fairly low-key, and I try to avoid ever being spammy. Eventually, those customers that have the same problems take a free trial of our software.  If they like it, they buy it, and I get to repeat the cycle for others.

If you’re wondering about the details of all that, as well as how I got CNCCookbook to over 4.5 million visitors a year, I am writing about it over on my Entrepreneurship Blog, BobWarfield.com.  I’ll walk through the entire system I’ve developed for Content Marketing CNCCookbook there over time along with a lot of other useful information for Entrepreneurs and Small Business Owners.

There’s a bonus too: if you sign up for the newsletter, you’ll get a free online course called Work Smarter and Get Things Done.  It teaches all the productivity hacks I use and even includes a piece of free software I wrote to help implement my productivity system.

Get Me My Free Work Smarter Training!

Should you take 1 eight-year startup job or 4 three-year jobs?

Jason Lemkin asks and answers an interesting question for career seekers in VC Startups:

Is software a great industry to job hop?

His conclusion is, “No,” largely because he makes the case that you won’t make nearly as much money.  His concerns are:

1. You don’t vest much.  He estimates 15% a year allowing for future grants.
2. You may not be able to afford to buy the shares when you leave if the strike price is too high.
3. If you get RSUs instead of options, you may get nothing when you leave.

I’ll address #1 in a minute, but let’s talk about #2 and #3.

Not being able to afford to buy the shares implies you’re joining the startup relatively late. The average startup requires 7 years start to finish to reach liquidity.  How late does the startup have to be before you can’t afford to buy your shares?  Tough to say.  An early unicorn would certainly do it.  But in most cases, my guess is you’re at least halfway through the 7 years.  So you can no longer buy shares at say the 3 year mark.

Okay, but is that a problem if you join at year 3 and allocate 2 years to figure out whether it is worth staying the remaining 2?  Unless things are still seriously hot after your 2 years are up (year 5), it seems to me you can still look at it as a 2 year stint and maybe just buy fewer shares.  If the thing really is a Unicorn, you’ll need fewer.  In other words, reason #2 is not as compelling as it seems–you don’t have to settle for nothing, you will get a chance to look at a few more cards in the poker hand during the 2 years, and you may simply wind up with a smaller return if you mistakenly leave early because you didn’t buy all of your vested shares.

Issue #3 I look at totally differently.  Do you want to join a startup that offers RSU’s (Restricted Stock Units) instead of a options?  Do you want to join a startup that gives you nothing if you leave after having given them 2 years?  Personally, I just wouldn’t sit down to play at such a table.  The house has things rigged too thoroughly in their favor.  I’m sure there are endless anecdotes about companies with RSU’s that made fortunes for their employees, but remember-the odds are already heavily stacked against you (shortly we’ll see just how much).  Do you really want this additionally risk?  And are companies that operate this way more or less successful?  Are they more or less likely to be good to their employees, people like you?  Color me skeptical.

Now let’s get back to concern #1–“You didn’t vest enough to make real money.”  You only got 15% of what you could’ve gotten instead of 100%.  Let’s look at it somewhat differently.  First, your resume looks terrible if you have a new job every year, so you need to stay for 2 years, not 1.  That gets you 30% rather than 15%, a big step up.

Second, concern #1 is expressed as though making the money is a sure thing, and it most certainly is not.  I prefer to look at it this way:

Should you take one 8-year job or four 2-year jobs?  Which one has the highest likelihood of putting you in the money?

If we look at it that way, and make a few assumptions, it’s possible to model the two scenarios in Excel.  In fact, we can handily run a Monte Carlo simulation and see what results we get.

Here are the assumptions I used for the simulation:

– You can either stay in 1 job for 8 years (about what it takes to go from 0 to liquidity in round numbers) or you can take 4 jobs and stay 2 years each.

– Your chance of picking a winner is 1/8. 1 in 8 deals wins.  Overall 3 out of 4 VC deals fail to return the investor’s money.  It’s a sure bet that a fair number that return the money still won’t return anything to you as the VC’s have all sorts of preferential terms in the deal.  So let’s just say it is 1 in 8.

– If the 1 job guy gets a win, he gets 100%. If the 4 job guy gets a win, he gets 30% (2 years at Jason Lemkin’s 15% a year figure).

Now we do 5000 iterations of that in an Excel spreadsheet for a Monte Carlo simulation. Here are the results:

– The 1 job guy only has a 14% chance of getting his 100% of shares to return. I wonder how many would sign up for a startup if they soberly concluded those were the odds?

– The 4 job guy has a 42% chance of getting his 30% of shares in the money.

Whoa!

Way better odds for the portfolio effect of taking 4 jobs with 2-year stints. So now the decision is a utility curve issue. Say we’re talking $10 million. Do you want a 14% chance at $10 million or a 42% chance at $3.3 million?

Jason has said that absolutely the most important thing is to make your first few million–he should be voting for the 42% chance at $3.3 million if you’ve never made your first few million.  He then advocates swinging for the fence to make 10x more, after you have your first few in the bank.

I’m not even sure that’s the right strategy, it is again, a utility curve thing.  What’s your goal?  How much will you sacrifice to get to that goal?  How much is enough money?

How lucky do you feel?

Or, look at it this way:

You only have so many deals, so many throws of the dice, in your entire career.  How many are left, particularly when you consider that many feel startups are a young person’s game?

I don’t feel startups have to be a young person’s game, BTW, but I do think you have to find some way of achieving a portfolio effect to maximize your likelihood of success.  Otherwise, you’re working your tail off and taking substandard pay mostly to help your investor’s win big due to their portfolio effect while the greatest likelihood is you’ll make absolutely nothing.

PS:  Now you’re wonder if your odds will be better because you’re smarter and this deal you’re looking at is just so good. Let’s say your odds are the same as the VC’s–1 in 4 deals will hit instead of 1 in 8.

What does doubling the odds do for you in the Monte Carlo simulation?

Recall we originally had a 14% chance of making 100% and a 42% chance of making 33%.  If we double the odds per deal, we now have a 26% chance of making 100% and a 69% chance of making 33%.

I don’t know about you, but having the odds favor me on each 2 year stint to the tune of 69% sounds awesome.  Heck, I might even succeed at more than one of the 4 stints, which would get my 33% up to 66% or maybe even more.

Tomasz Tunguz (Redpoint VC) and Lloyd Tabb have got it wrong–way wrong.  Tunguz recently published an article based on a conversation he’d had with Tabb that suggests early and mid-stage software companies can’t benefit from A/B Testing because they don’t see enough web traffic to make the results statistically significant.  They suggest that instead, they should make decisions based on qualitative data:

… interviewing users about the whys underpinning their points of view on price, reviewing the video of people exploring the product, and opinions about design. It’s the qualitative data, the acumen of an brilliant designer, the insight of a skilled product manager, the empathy of a master marketer.

Ouch!  Back to anecdotal evidence and marketing decision making by the most important person in the room.  Back to the bad  old days, in other words.  There’s nothing wrong with doing those things they suggest, but before you bet your company on the results, you must A/B test them.  These are just inputs to decide what to test, in other words.

Back to anecdotal evidence and the bad old ways of marketing…

Before we throw the AB Testing baby out with the bathwater, let’s take a closer look at what’s possible.  The Chief Witness for Tunguz and Tabb is Optimizely’s Sample Size Calculator:

Optimizely’s Sample Size Calculator…

It’s a great tool that I use all the time, BTW.  They’ve selected the default view, which suggests that if you have a baseline conversion rate of 3%, and you want to see a minimum 20% detectable effect with 90% confidence, you will need 12,000 visitors to the page.

There are two key questions to explore before we can agree or disagree with the proposition in an informed manner:

1. Are these the right inputs for Sample Size Calculator?
2. Given the right inputs, is the sample size too large for most startups to attain?

For the first question, I submit that the defaults are actually not very relevant at all. Requiring 90% confidence or be willing to accept anecdotal evidence is pretty silly.

Heck, I run my own bootstrapped startup, it’s entirely my capital that’s at risk (I’ve accepted no outside investment), and I would be thrilled to ring up 70% confidence interval tests all day long.

As it turns out, Optimizely will only let us go to 80% confidence, but Google’s A/B Testing will tell us it’s evaluation of the confidence regardless of level.  I will add that the statistical confidence is also not the only factor we should consider.  It’s important to make sure you really have a representative sample.  For example, test results may vary by day of the week, so I never accept a test that’s run for less than a week, even if the confidence is 90% or more.  In fact, I typically prefer 2 weeks as a minimum.

Cutting the Optimizely confidence down to 80% gets us down to a sample size of 11,000. Let’s next consider the baseline conversion rate.  3% is not an especially good benchmark for a product landing page.  Groove.com surveyed SaaS companies and came back with a visitor-to-trial conversion rate that averaged 8.4%.

If we plug in an 8% conversion at 80% confidence, the sample size plummets to 3,300 visitors before we can measure a 20% detectable effect.  We’ve cut it almost 4x, but we’re not quite done.  What about that 20%?  Is it not worth conducting A/B tests unless they result in 20% differences?

Here I’ll turn to my own experience AB Testing for my own company, CNCCookbook. In the last 8 months I’ve conducted 55 A/B Tests.  The average change between the baseline and the variant I measured was 30%.  Are you surprised?  I was VERY surprised at how much impact even seemingly little things could have.  FWIW, 44% of my tests yielded a positive improvement, 29% showed the idea failed, and 49% of the tests failed to reach statistical significance.  I have no idea how that compares to the scores for other marketers, but I am very happy with the results.

If we plug that 30% number in, we get to a sample size of 1,300 visitors.  Applying my rule that I usually test for 2 weeks, we need to come up with less than 100 visits a day to the web page we’re testing.

Is that bar too high for startups to clear?  It shouldn’t be if the marketers are doing their job right.  I’m a one-man bootstrapped company and my CNCCookbook site sees about 15,000 views a day to the site.  I get about 250 a day to the home page and about 450 a day to my product home page.  As I write this, Google Real Time Analytics cheerfully informs me there are about 50 people running around on my site.

Clearly I can do very statistically significant A/B Testing and it has benefited me quite a lot. I get over 6000 visits in 2 weeks so I can measure as little as a 15% change in that time, and even less if I am willing to let the tests run longer.  Incidentally, don’t overlook the value of a test that ISN’T significant.  That test is telling you at the very least that even if it is bad, it is no worse than the statistically measurable results.  So, if we can test to a 20% detectable effect, adopting the wrong variant will do no more harm than 20%.  Sometimes when we need to move ahead boldly, knowing we can do no more harm than that is good enough.

Granted, I’ve had this company for a few years, but if I can get this far by myself, a VC-funded startup should be able to do at least as well and much faster.  They have to in order to have much hope for a Unicorn-valuation.  Tabb’s company, Looker, the one that presumably prompted the discussion, looks like it should have a little less than half the organic search traffic I get based on SEMRush results.  Clearly, Looker should be able to benefit tremendously from A/B Testing if it chooses to.

So, VC Board Members–expect quantifiable results from your portfolio companies and don’t take sample size whining for an answer.  Entrepreneurs, saddle up and ride this A/B Testing horse–it’s a powerful tool that can really move the needle.

My best advice for startups right at the beginning, BTW, is start building your audience BEFORE you build your product.  I call it achieving Content-Audience fit, I’ve been writing about it for years, and it is absolutely the very first thing a founding team should do when they get together.  Achieving it provides a number of powerful validations for your team, but more importantly, it validates there is a reachable audience, and in reaching it, you gain a powerful tool for shaping your journey to Product-Market Fit.  Not to mention, you set yourself up to achieve enough traffic to do meaningful A/B Testing just that much sooner.

Stealth Mode is harmful in this respect–it delays your access to Content-Audience fit for no meaningful benefit.  So what if the world knows what broad market you’re working in or even what broad problems you write about?  You don’t have to tell them anything about your product or how it helps solve those problems.

No more excuses–get on with  your marketing people, and do some rigorous AB Testing of it!

Every now and again, a good rant can be cleansing.  I mean that in the best possible way.

Today, for example, I had a good rant after reading that Fred Wilson, a Wealthy and Influential Venture Capitalist in New York, was “Bothered” by the “Losing Jobs to China Discussion.”

Fred’s not long enough on facts to do much more than be troubled and hand wave away the discussion.  In his mind, there won’t be any manufacturing jobs because automation is destroying them so quickly anyway.

It’s really the Robots that took all the jobs…

Fred doesn’t think it’s worth bothering with the Manufacturing Sector because soon there won’t be any jobs left after automation anyway.  Far better to make everyone an IT guy or some such.  We’re in a transition we should double down on be happy with.  Something like what he says here:

The US and a number of other countries around the world are building new information based economies. That is the long term winning strategy.

So while we can critique our leaders (business and political) for giving up on the manufacturing sector a bit too early, I think the US has largely played this game correctly and will be much better off than the parts of the world that have taken the low cost manufacturing jobs from us.

The thing is, most all of this is a lot of Lies, Half Truths, Myths, and general Bollocks that got started by people who would benefit from offshoring manufacturing and is maintained as a cherished belief as so many myths are just because it’s been a self-fulfilling prophecy. In other words, if we destroyed our manufacturing economy it must be because our manufacturing economy was doomed and not worth saving in the first place.

Take the Robot argument.  It’s uber-popular in VC circles because people like Andrew McAfee have made careers out of pushing this thesis.  Yet, if we actually look at the numbers (which I do in detail in the article below), it’s very hard to make the case that Robots have taken more than maybe 20% of the jobs away.  That’s a far cry from eliminating an entire market segment or deciding they’ll never be able to produce enough jobs to be worth considering.

The reality is the whole thing was manipulated by a variety of parties, is based on a large number of non-truths, and is relatively easy to reverse.  Moreover, it would be extremely valuable to reverse it.

For all the detail, check out my longer post with facts and figures over at CNCCookbook, my own Manufacturing-related company.

It may be that article hit a nerve, because Fred’s site deleted a comment wherein I referred to it as “spam,” LOL.

I just read a post on the KissMetrics Blog by Cody Lister entitled, “More Trial Users is Not the Answer For Your Startup’s Growth.”  It’s not a terrible post.  In fact, some aspects are pretty decent.  Lister basically wants startups to focus on better engagement and the onboarding experience and less on just running as many people as possible through the trial process.  He wants you to be sure you’ve got product market fit before you try to scale out with as many trials as possible.  I have no problem with the latter, BTW, but it is unrelated to the areas I do have a problem with.

Unfortunately, perhaps due to the act of trying to make the point as persuasively as possible, he strays into at least one area where I think his advice is dead wrong.  It’s way too Black and White, and whenever someone gives me a Black and White answer, I instinctively look for the exceptions to the rule.  Let’s put aside that in fact more free trials will help you to grow, it simply may not be the optimal thing for you to be focused on right now (or it may be, it all depends).

Instead, let’s drill down on the area that really got me thinking it was bad advice from a marketer who should know better.  Lister states:

Eliminate Or Reduce Free Trials

What if one day, your team just decided to shut down your free trial accounts that were past 14 days since their sign up date? Would you suddenly go out of business?

No, you’d save money from server costs and force people to make a decision.

It’s only when your free trials run out that you know whether the end user found your product worth paying for.

You need to figure out how to improve the engagement of your existing trial users to convert them to paid users.

ConvertKit and Edgar, which today generate millions of ARR, never offered free trials.

I often come across startups that give away free trials for 30 to 60 days. I just don’t get it.

I could not disagree more with his advice to eliminate or limit free trials to 14-days.  He states it as an absolute to the point that, “he just doesn’t get” why anyone would be stupid enough to offer a 30 or 60 day trial.

He gives only two odd exceptions to his rule:

• A B2B SaaS offering costing more than $200 a month.  No explanation whatsoever why the arbitrary figure of $200 was chosen.
• An offering where personal data had to be entered and value received increases proportionally to the amount of data entered.  He argues this creates switching costs, which is worthwhile, but actually misses the point.  What he misses is not only does it create switching costs, but the more data in services like DropBox, the more likely the user is to experience the “Aha” moment that closes the sale.  Switching costs come later, after the user is satisfied and someone else wants to woo them away.

Let’s dig into it with a couple of real world examples that I think will help explain the real reasons why you need to think about your Free Trial in terms of the user experience and not in terms of arbitrary advice from marketers.

Ironically, one of the reasons I tried but did not adopt KissMetrics (the very blog where this is posted) was because I could not tell within 14 days whether it would deliver value. In fact, KissMetrics is a wonderful illustration of the problem with this one-size-fits-all advice.

It’s biggest benefit is a better understanding of your sales funnel. So ask yourself, “How far does a user travel in the funnel in 14 days?” Further, how much of the 14 days is needed to get things set up and to accumulate enough people travelling through the funnel to make things even interesting?

You can now see where I’m going.  It might very easily take more than 14 days to get to that “Aha” moment where I see the value in a product like KissMetrics and I’m ready to pay up for it.  In fact, for my company, it really was longer than 14 days.  This was exacerbated by various aspects of the KissMetrics user experience.  It took the service time to accumulate enough data points to show me any funnel reports.  It took me time to understand the service well enough to get my funnels set up properly.  And it took time given my web site’s traffic to accumulate enough data points to see any kind of picture clearly.  BTW, it’s no small web site, I get 2 million uniques a year.  Pretty good for a small business.

I believe 30 days would’ve worked nicely for my case, but alas, I only had a 14-day trial to work with.  So I moved on.

Let’s try another recent personal example: Drip, the marketing automation app.  I wrote about my experience with them recently.  They had a 21-day trial.  During that time I was trying to:

• Learn a complex new application
• Tie in my mature and complex email best practices
• Develop a new lead nurturing automation campaign far enough to evaluate the product

I felt it was reasonable that my “Aha” moments for Drip would include:

1. Verifying it could do what my existing provider, Mailchimp, was already doing for my business.
2. Verify that it could so something that Mailchimp couldn’t via its increased automation features. After all, Drip was going to be more expensive–it should show me some magic relative to Mailchimp during the trial.

As I documented in my write up, I was unable to accomplish these tasks within 14-days despite trying like crazy to get them done.  I had a mixture of problems ranging from product bugs to unclear UX to my own stupid noobie user mistakes.  I could not even get my email newsletter out, despite trying hard for 2 weeks running, so I couldn’t even verify Drip worked as well as Mailchimp, let alone see the impact of its new features.

I wound up sending Drip’s Founder an offer–extend the free trial and work with me until we can make my Drip experience a happy one.  In exchange, I’d buy the product and write about my experiences in places like this blog.  He declined, saying many of his competitors didn’t offer a free trial at all.

Here’s the thing:

If you’re going to offer a free trial, you really should make sure it is long enough that your users can reach the “Aha” moment where they’ve confidently demonstrated your product’s value and it’s an easy choice to reach into the pocketbook and become a paying customer.  Ignore all the other rules of them because reaching the “Aha” moment is the only thing that matters for your Free Trial.  That is its singular purpose.

If you’re not going to do that, why have a free trial at all?  I can’t imagine a reason unless it’s just part of the old bait and switch–get them to commit a little, even just give us their email, and each thing they give up will make the next thing that much easier.  That’s a well-understood marketing concept, and it even works to an extent, but is that really the way to build your successful business?

I can’t believe marketers think so, at least not the good marketers.  Please tell me you’re not in that camp.

Length of trial is something that should be tested, preferably AB tested if you can arrange it. Don’t get too greedy and eliminate your trials before your customers can experience the “Aha” moment that guarantees they will love the product. If you can make that happen in 14 days, great, but don’t just assume that’s the case.  Give them whatever time they need. Even offer to extend the free trial for ANOTHER 30 days if they’re not done evaluating.

You’d be surprised what treating your customers as human beings rather than inventory will do for you.