Stop me if you’ve heard this one:

Your bootstrapped startup is finally off the ground. You’re able to spend $6000/mo on AdWords to drive leads. Sure the conversion rates could be better, and sure it’s not the best ROI on Earth, but on the balance it’s making money.

You don’t have a huge budget, but you can plough some of your winnings back into advertising.

You’ve heard banner ads don’t work well, but they’re cheap, so you start throwing $600/mo into an ad network and trust your web analytics to tell you whether it’s working. The jury is still out, but there’s been only two signups in the first month. Maybe your ads suck?

You’ve heard affiliate programs can work wonders, so you sign up with a cloud affiliate provider and figure you can afford to pay $50 per signup.  A hundred affiliates take the bait, but two months later, half of them haven’t sold anything, and most of the others have sold only once. Two are producing five signups a month, and it’s only cost you $1000 so far, so it’s not all bad, but it’s not moving the needle.

So now what? Should you work on optimizing AdWords since they’re working, or optimizing banner ads since they’re not working?  Should you cut off the affiliate program since it’s a waste of time, or redouble your efforts in affiliate management? Should you optimize these existing channels or try to find a new, more productive channel?

These aren’t hypotheticals — I talked to no fewer than four startups in the past few months (mostly at SxSW) in exactly this position, with these advertising channels, with about the same costs. And my own company WP Engine wasn’t dissimilar two years ago.

So here’s some assorted tips:

“Maximum inventory” is the answer to “Should I optimize?”

A common question is: “Should I spend more time optimizing AdWords? Or find more keywords? Or focus on another marketing campaign completely?”  A common — and unacceptable — answer is “It depends.”

I approach it as a matter of inventory. In advertising, “inventory” means “all of the available advertising space.” In a magazine, that means the total square feet of space allocated to the adverts.

In AdWords, that means the maximum clicks you can get across all relevant keywords.  If you’re the #1 spot for a keyword, you’re already getting the maximum number of clicks for that keyword, and therefore you’re occupying the maximum available inventory. (Of course there’s more inventory for other advertisers, but AdWords allows you only one slot, so this is the maximum available to you.)

So, look at how much additional inventory you think is available for you in AdWords.  If you’re in lower spots for most keywords you care about, or if there are other keywords you’re not yet bidding on, there’s probably a lot more inventory you could be taking. That implies it’s wise to optimize — there’s more sales for you there, and it’s easier and cheaper to optimize an existing campaign than to start up a new  one. You might even experience some cost savings (per signup) as part of your optimization — bonus!

If you think there’s 2x or more inventory you could go get, I say go get it.

However, that’s not the case for us at WP Engine.  At the instant of this writing, we’re the #3 spot on AdWords and the #3 organic spot for “WordPress Hosting” which is one of our most important keywords, and #1 for “Managed WordPress Hosting” which is perhaps our most accurate keyword.  And traffic on related keywords diminishes quickly and therefore improving on those is even less interesting.

We don’t have a lot of inventory left.  We believe less than 2x.  Therefore, even if we spent tons of money on human analysis and were willing to lose money on every sale, it still wouldn’t be an area of significant growth for us.

Once you approach inventory limits, you need to find other campaigns which can 2x your business.

No benefit of the doubt.

When you try a new campaign and it utterly fails, the temptation is to keep spend low and try to optimize. You think:

I might be just a clever turn of phrase or an eye-catching design away from changing that click-through rate from 0.1% to 1.0%, and then I can ramp up spending and have an important new advertising channel.

After all, this same messaging worked great on AdWords, so I know it’s reasonably well-tested. Something small is probably holding me back.

And I recognize my marketing prowess with this company is nascent, so of course campaigns will suck at first but improve with iteration. So let’s iterate!

I’ve found that this line of thought is usually wrong. I don’t have data to give you, it’s just been my experience.

Thinking about the 100+ campaigns I’ve been a part of in the past 11 years at Smart Bear and WP Engine, the ones that were bad out of the gate tended to stay bad.

Maybe it’s because by the time you get to the third campaign your messaging is pretty good, so an early failure means it’s a bad channel rather than a bad message. Maybe it’s because iteration gets you incrementally better but not drastically better and therefore can’t save a fundamentally bad channel. Maybe it’s because a failure is an indication that you don’t understand the channel at all, a condition that simple iteration won’t remedy.

In any case, cut out early and go find another channel.

Website trumps all.

Optimizations you do on your website are more valuable than those you do on individual marketing campaigns. The reason is obvious in retrospect: all marketing campaigns lead there!  So a 10% improvement in bounce-rate off your pricing page means 10% more revenue across all campaigns: paid, organic, and word-of-mouth.

It’s often true too that landing-page optimizations can at least somewhat be shared. The exact text can’t, but often you can cross-apply a layout that increases click-through rate, a form that captures an email instead of signing up directly, or a message to one customer segment that happens to be accessed by multiple marketing campaigns.

Often an early-stage startup has large-effect optimizations lurking all over its website. Maybe the testimonials or demo page is missing, or not compelling, or you realize testimonials need to be on all pages, or a long-scrolling, well-designed home page triples click-through rates, or you didn’t have a strong call to action, or the pricing page was confusing, or your main hook and product description was turning people away.

Only look for large effects.

When you’re small, almost no data you have is statistically significant.  Therefore, most variation you see is due to random fluctuations, not real results, even if you’re using one of those tools which supposedly helps you with the statistics.

I have a 45-minute talk on this topic which is too long to interject here, so you’ll just have to believe me.

The result is that you should only be seeking large effects, and only thinking you’ve “learned something” on that basis.

If you’re getting one sign-up a day, a “10% improvement” still means one sign-up a day.  You need two signups a day.  You’ll know if you get a change that drastic! Those are the only important changes.

Low-budget advice only!

Finally, as the title suggests, this is advice for low budget online marketing. Large budget marketing is completely different. You try as many channels as possible, as fast as possible. Find the ones that work as quickly, then immediately ramp to full inventory. For channels like affiliate programs where you can’t just “write a big check” and make them work, you develop an entire team to extract as much value as possible. The fact that you burn a lot of cash on things that don’t work out is exactly the trade-off you want.

Accelerated (cash) burning for accelerated learning. Good work if you can get it!

But if you’re cash-strapped, use inventory limits to decide whether to flog a channel, get out of bad channels quickly, optimize your website before optimizing channels, and don’t get excited by tiny, immeasurable results.

Please leave more tips in the comments!

In the first year of business, you have literally no data for making decisions and predictions.

Even after the first hundred customers, half of those were serendipitous one-offs, not representative of repeatable, controllable customer acquisition, and the scale of the data isn’t statistically significant.

One of the root questions you have at the start, which is supposed to be data-driven (but you don’t have data) is: What’s the maximum I should bid for CPC (cost-per-click) campaigns like Google AdWords?

The answer for a funded startup is “Bid as much as possible, to get as many customers — and data! — as you can, as quickly as you can, then rapidly iterate from there in the presence of that data.”

Easy for them to say, but what about a bootstrapped, profit-driven business? Here you don’t have the budget to “spend as much as possible,” and you’re keen on getting a reasonable return on investment reasonably quickly, and you can’t just “spend to acquire the data.”

Here’s my way.

(Tune the exact numbers below if you disagree with my assumptions, but the process should be valid for anyone interested in chasing profits with a small budget.)

LTV = MRR x 20

MRR (Monthly Recurring Revenue) is the amount you charge a customer every month.  Actually, it’s the average amount for all customers, which is typically a mixture of different quantities of customers at different tiers, special add-ons, etc..

LTV (Life-Time Value) is the total amount of money you expect to collect from a customer over their entire tenure with your company.  In general you compute this as simply “MRR x [expected months]” meaning the average number of months a customer sticks with you.  Some customers cancel in one month, some cancel in a year, some in five years, and some never cancel!  So it can be difficult to compute LTV accurately for small companies, and impossible to know for young companies (where e.g. five years hasn’t elapsed yet to see exactly how many customer stuck it out that long).

If you do have data, here’s my deep dive on cancellation rates and LTV.

But since you don’t, in my experience (and in a non-scientific survey of some of the 100 startups currently officing at the fabulously Capital Factory co-working space in Austin), a good pre-data rule of thumb is 20 months.

If you have an average customer lifetime smaller than 20 months, that’s a dangerously high cancellation rate for almost any SaaS business, and you need to focus on addressing the business issues before focussing on acquiring more unsatisfied customers. Use surveys and one-on-ones to try to understand whether it’s technical failings, lack of features, missed expectations, bad service, doesn’t hit pain points, or what.

A healthy SaaS company will have a higher number of expected months, but at the start you also will have lots of mis-steps with early-adopters where your product is at its worst — least features, least quality, etc — so it’s good to assume a low LTV instead of inflating it to where it might be in future.

CAC = LTV / 5

CAC (Cost to Acquire a Customer) is your average total cost to get a new customer, which includes (as accountants say) direct costs (AdWords spend, affiliate payouts, the fees your affiliate system charges to process them) and indirect costs (consultants and your own time). So to compute CAC, take your total costs to acquire new customers and divide by the number of customers you acquired.

In general of course CAC needs to be less than LTV, otherwise it costs so much to get the customer that you can never make money. A surprising number of startups have CAC > LTV.  Many justify this either by not correctly computing CAC (e.g. ignoring indirect costs) or saying they’ll “fix that later” by raising prices, finding other channels of revenue.  Others justify by saying they’re doing a “land-grab” for customers, and just having a customer at all has intrinsic value.

Profit-seeking bootstrapped companies cannot afford those delusions. Also you need something far stronger than CAC = LTV, because you need to pay for other business expenses and still produce a profit.  So how big can CAC be before it’s “too big?”

Growing, funded SaaS companies who treat CAC with respect often commonly target CAC = LTV / 3.

Back at my second startup IT WatchDogs, my co-founder Gerry Cullen used to say “A third to built it, a third to get rid of it, and a third to keep,” meaning a third of revenue goes to pay for hardware/inventory/shipping costs of the sale, a third goes to what I’m calling “CAC” here, and a third for the overhead costs, development costs, and profit.

That’s a good model, and I think a bootstrappped company can copy it, but I urge profit-seekers to instead adopt an even more strict model of CAC = LTV / 5.  The reason is that at the start you should be able to find a few efficient ways of acquiring customers, even if those get saturated over time, because it’s exactly what you need early on when you have least-money, least-branding, and least-insight into marketing messages.

CAC = MRR x 4

If you combine the previous two results, you see that the cost to acquire a customer should be no more than four months of revenue.

Another good way to think about it is: “The payback-period for my cost to acquire a customer is four months.” Also, ideally you’re getting the first month of revenue back immediately, so it’s really three months of cash-float.

Companies with large budgets to deploy at scale will often be happy with 6-12 month payback periods; some very high volume businesses like shared hosting will accept 24 or 36 months even! But a bootstrapped company’s cash-flow won’t allow it, even if the math would work in the long run.

Conversion Rate = 1%

Conversion Rate is the percentage of visitors to your website who convert to a paying customer.

This is another step which in practice should be completely data-driven, segmented by customer type and marketing channel, segmented by landing page, A/B tested and iterated, blah blah blah. But since you don’t have data, and you don’t have enough visitors to have real ratios, you have to take a swag at this number.

In that same informal survey I ran, and bolstered by other formal surveys, a huge number of bootstrapped SaaS companies report a 1% conversion rate.

Another way of saying the same thing is “You need 100 visitors to make one sale.”

And since you need to incur no more than CAC dollars in the making of that sale, you need to incur no more than CAC/100 dollars in the making of each of those visitors.

And if you’re running a CPC campaign, that means you can pay up to CAC/100 dollars per click.

And since CAC is MRR x 4, we can substitute and get the end result:

CPC = MRR / 25

So for example if your average customer generates $50/mo, you can spend $2/click.

Indeed, this is a great way to prove one of my main arguments for all bootstrapped companies, which is that you should charge a lot more than you think, in part because it enables you to pay quite a lot per click, which enables a wide number of marketing channels, and out-bidding parsimonious competitors whose paltry LTVs preclude them from competitive marketing spend.

I know we’ll have lots of discussion in the comments, and I hope to see (good-natured) dissension and alternative ideas so that all of us get lots of perspective on how to do this sort of thing.

Seems like every third startup nowadays is using the “Freemium” business model: The lowest service tier is free, and the business is designed to get those users hooked and then upgrade to a paid plan.

It can work wonderfully of course, but usually it crushes and destroys companies, not only because it costs more than anticipated but because the founders didn’t realize the business model itself caused them to make incorrect decisions.

Foibles of Freemium

Let’s dispose of some misconceptions about what Freemium actually does for you and how much it costs.

Freemium is not customer development

Just imagine how much you’ll learn once you have 1,000 real, active users of the system: Everything from behavioral statistics (which features are actually used?) to democratic product development (voting on which features customers would like to see next?).

Trouble is, those freemium users are not like those who will actually give you money. Frequently the features that paying customers want don’t show up on the freebie’s radar.

Think about it: Almost none of the freeloaders will convert to even the smallest tier. If you suddenly started charging only $1/mo for your service, most wouldn’t come aboard. Why is that? Because the need, the interest-level, the value to the end user isn’t pressing enough for even a pittance. But those who will pay $10/mo or $100/mo do have needs, and it’s not just a matter of scale (i.e. because they require more resources), it’s a difference in kind.

Your problem is that freemium users outnumber your paying customers 100-to-1, so in feedback forums they drown out the voices of those who actually matter.

Freemium conversion rates makes marketing expensive

A really good conversion rate for free-to-paid is 4%, like Dropbox. Awesome for them, but normal rates are more like 1%, and that’s if users are reasonably active.

I surveyed a dozen small startups who don’t use freemium, and on average they see a 1% conversion rate from web traffic to a (real, not “free”) purchase. [UPDATE: Andy Brice has much more data which supports this simple assumption] Even assuming you can get a higher website-to-signup rate for a freemium offering (you’d better, right?), you only get paid on on a few percent of those, which means your total conversion rate of web visitors to actual money is 20-100x worse than other startups.

Which means it costs 20-100x as much on marketing campaigns to achieve the equivalent revenue.

This essentially takes paid-advertising and almost all other forms of marketing off the table for driving growth in a freemium business, unless you’re willing to take big losses to get things rolling.

It also means you essentially have to build a viral product, because you can’t afford advertising. Getting true viral behavior is very hard — again most companies who attempt this will fail — and even so you need to seed it at the beginning, so you still have the marketing expense problem.

The few companies who were clear winners here also raised millions in funding, in part to get over this hump.

Freemium tech support is expensive

It’s easy to say “We’ll just direct everyone to forums,” but when people email tech support they want a response.

It’s easy to say “We won’t provide tech support for the free tier — they’ll understand since it’s free,” but if you really do ignore them they’ll be less successful with your tool, which means far less chance of them converting, and less change they’ll evangelize to friends and coworkers.

It’s easy to say “We’ll provide a lesser grade of support for the free tier,” but that means every tech support email and chat session and phone call has to be coupled to an account, so you’re wasting time figuring out what level of support this person “deserves.”

Are you prepared for people who say, “If you help me through this, then if it works I’ll pay.” Are you hard enough to shut the door in their face, even knowing that in fact they probably still won’t pay?

If you care about good support — one of the few true competitive advantages a small startup can have — can you really segment who gets treated well and who gets the cold shoulder? Should you really turn your back on the benefits great tech support brings? Is that conducive to converting free accounts to paid accounts? Is it helping your company’s reputation?

Advantages of Freemium

Obviously freemium also has important benefits which cannot be denied:

• Easy to upsell. They’re already using your tool, so whether it’s by special offer, changing the pay scale, or the user just outgrows the confines of their service tier, there’s many ways a person can start giving you money. That’s something few other business models can boast.
• Stats for selling. It’s awesome homepage marketing to be able to say “Join 1,000 other happy users.” It’s social proof, just like the RSS counter in the corner of the blog implicitly saying “if 40,000 others think it’s worth their time to read this every week, maybe it’s worth yours too.” It’s also useful when bagging bigger customers because it proves your system can scale, both in technology and in training new users with minimum effort.
• Easy to start. Even a “30-day free trial” or “money-back guarantee” is a much bigger barrier than “free.” Getting a web visitor to stop perusing and start using the product is a critical step in any customer acquisition, and you’ve just diminished the barrier as much as possible.
• Not using the competition. Well, they might still be, but at least you have a horse in the race. One more user for you is one fewer user for them.

So in the face of the positives and negatives, how do you decide whether it’s right for you and, if it is, how do you think about it so that you’re reaping the benefits while mitigating the costs?

Charge Freemium to the marketing department

A pattern emerges from those “advantages” bullets: It’s all marketing. It’s lead-gen, reducing barriers to conversion, and competitive advantage.

Retool your expectations of Freemium: It’s a marketing cost. It’s more expensive than you give it credit for, but it could very well be the best marketing strategy available.

So how do you decide whether those costs are worth the benefits? My technique is to “charge the marketing department” exactly like AdWords or any other lead-gen campaign: measuring the total cost of acquiring new paying customers.

The reasoning is: You have a theory that by spending the money to support these freeloaders, you’re in fact building an efficient path to real, paying customers. That goal — revenue — could possibly be attained other ways: AdWords, blogging, or any other marketing technique. So just like any marketing campaign, the marketing department should pay, measure the results, and compare the ROI against other methods.

(And make sure the cost is much less than the total revenue — the equation that most startups fail to achieve, exactly because they don’t honestly consider the total cost to acquire.)

How do you decide how much to charge the marketing department? Suppose you really were charging those customers, but also that you’re only looking to recoup costs and not make a profit. Amortize those costs — servers and tech support — and come up with a correct cost-replacement price for that tier.

That’s what those users should be paying (minus profit), so that’s the amount the marketing department needs to “reimburse” the rest of the company.

Obviously you shouldn’t get carried away with the bookkeeping. But you have no excuse to not be measuring that per-user cost so that you’re running this Freemium program with full knowledge of the cost and comparing it to other forms of marketing.

Further Reading

More great articles on the subject:

• Andy Singleton on how to “make those cheapskates pay.”
• Andrew Chen on the precise financial model of freemium (spreadsheet included).
• Eric Ries on three strategies for “free” that still lead to making money.

Guest post by Ben Yoskovitz, the co-author of Lean Analytics, a new book on how to use analytics successfully in your business. Ben is currently VP Product at GoInstant, which was acquired by Salesforce in 2012. He blogs regularly at Instigator Blog and can be followed @byosko.

Have you pivoted before? Probably.

Was it a lazy pivot? Hopefully not.

Unfortunately, due in part to the popularity of Lean Startup and the concepts it promotes, the pivot has been bastardized to the point where it doesn’t mean much to people anymore. We hear the word “pivot” and groan or shrug, knowing there’s a good chance that people are using the pivot as an excuse.

We know entrepreneurs are delusional. It’s part of our very nature. We surround ourselves with a reality distortion field that allows us to survive the rigors of startup life and believe, in our heart of hearts, that what we’re creating and envisioning is going to succeed. We have to go into the world and convince others of the same thing–customers, investors, partners, employees and so on. In absence of proof, we rely on our delusion and reality distortion field to help propel us forward.

But when your reality distortion field is impenetrable and you’re at the point where you can’t separate fact from fiction, you’ll crash and burn in a horrible way. It happens to most of us. And it’s painful.

Lean Startup pokes a hole in our reality distortion field. It provides a necessary dose of intellectual honesty that enables us to miss the wall (or hit it a bit less painfully) and adapt. To pivot.

So, what is a pivot?

Here’s how I define it: A pivot is a shift in one aspect of your startup’s focus based on validated learning.

If you’re changing your entire business that’s a do-over. Starting over is fine, but it’s not really a pivot (and incidentally, you still won’t succeed unless you’re starting over with some insight that guides you). A pivot is much more narrow than a do-over — and it starts with having learned something. It’s what we call “validated learning”. You pivot when you know where you’re pivoting and why. You’ve gained an insight through your efforts (working through the Lean Startup methodology) that tells you what direction to point yourself.

Validated learning

Validated learning is at the heart of Lean Startup. It starts with a hypothesis, followed by a test to prove or disprove the hypothesis. You use analytics to measure the results of your experiments. Let’s use Backupify as an example.

Backupify provides online backup storage. They may not have been rigorously following Lean Startup from the get-go, but founder Robert May was testing his ideas right from the start. “Initially, we focused on site visitors, because we just wanted to get people to our site,” he said. “Then we focused on trials, because we needed people testing out our product.” Their initial hypothesis could have been, “Do enough people care about online backup storage for us to continue?” The initial website and efforts to attract visitors (and getting them to sign up) was the experiment. In order to justify (to themselves) that it was worth continuing, “enough people” would ideally be defined (10 users? 100 users? 1000 users?) with some target goal.

Robert learned that enough people did care about online backup storage. But as the business matured and they started charging money, they ran into a problem. It turns out that the cost of customer acquisition for consumers was too high compared to the revenue being generated.

“In early 2010 we were paying $243 to acquire a customer, who only paid us $39 per year,” explained Robert. “Those are horrible economics. Most consumer apps get around the high acquisition costs with some sort of virality, but backup isn’t viral. So we had to pivot [from consumer sales] to go after businesses.”

Two things are important about this:

1. Robert could not have learned about the poor economics until he had learned a host of other things (that people cared about online backup storage; that customers stayed engaged for a long enough period of time; etc.) Only once Backupify had figured out how to acquire customers and retain them, could Robert focus on the economics.
2. Robert’s hypothesis that he could build a scalable consumer business was invalidated by the economics. The “validated” part of what Robert learned was pretty clear: the numbers just didn’t add up.

Backupify pivoted to focus on businesses and since then they’ve scaled very successfully. Businesses pay more for (essentially) the same service, and Robert can keep an eye on the core economics by looking at the ratio of Customer Acquisition Cost to Lifetime Value.

Validated learning isn’t always quantitative. Very early on in your startup, it’s most likely going to be qualitative. Qualitative feedback is messy and much harder to interpret, but it’s incredibly important. In my experience, once you’ve spoken to 10-20 people (doing well-structured customer interviews) you see patterns that give you enough insight to make decisions. You can then scale that learning by using surveys and reaching more people in a more quantitative fashion.

The “Lazy Pivot”

If you don’t have validated learning you’re pivoting blindly. That’s a “lazy pivot” and it will likely lead to failure. The lazy pivot thought process goes something like this:

“Well, this thing I’m doing isn’t really working … I’m not sure why … but this other thing looks interesting and cool, so I’ll go do that.”

Ugh. Most probably you haven’t put enough rigorous effort into your existing idea, and it certainly doesn’t sound like you have the necessary insights to move to something else. You’re jumping from one shiny object to another.

Using Lean Analytics to Pivot Successfully

To pivot successfully you need focus and learning. You change one thing about your business –maybe it’s the business model, or pricing, or target market– and see what happens. While writing Lean Analytics, we came up with a cycle that helps explain the process of testing your idea or product, and deciding whether to pivot or not:

You can see from the Lean Analytics cycle that there’s a lot of focused attention and intellectual honesty required to test your assumptions properly and then adapt as necessary. This is a very metrics-focused approach, but you can ask yourself some higher level questions as well.

Evaluating Your Business Model

I like to use a Lean Canvas as a forcing factor to ask myself, “Do I really know what the hell I’m doing?” If you’re not familiar with Lean Canvas, you definitely need to take a look. It’s a 1-page business model tool. If you look at your Lean Canvas (which you should be able to put together in an initial 20-30 minute session), can you honestly say you’ve got enough of the answers to keep going? Do you really understand the problem you’re trying to solve? Do you really know if the solution is right? Do you understand the channels to market? Do you have an unfair advantage?

You have to go through this regular re-evaluation of your business in order to genuinely be able to make good decisions on what to do next.

Some people look at Lean Startup (and by extension Lean Analytics) and believe that it’s an entirely mechanical process, devoid of spirit, and either not effective or so effective that if they just follow the process to-the-letter, they’ll win. Neither is true. Startups aren’t built in factories on a manufacturing line. You can’t ignore your guts and passion.

Follow the Lean Canvas and use it to ask yourself if you’ve got enough information and enough confidence to move to the next step with your startup. If you’re starting right at the beginning with the problem–can you honestly say that you’ve found a problem painful enough that it’s worth solving? Have you done enough, good quality interviews with prospective customers?

One idea we proposed while writing Lean Analytics was scoring problem interviews. Since the interviews are meant to be fairly structured, it’s possible to assign scores to certain questions and get a more quantitative sense of people’s input. This may help you in your quest for validated learning, beyond using qualitative input alone.

Passion Matters

If you don’t have passion for what you’re doing, you’ll fail. The effort to succeed is too great to do it without an insane amount of caring and emotional attachment. While on occasion this is at odds with being process-focused, you have to find a way of combining your passion and guts with intellectual honesty and rigor.

So let’s say you’ve found something interesting –you have an insight based on your current business– that tells you where to pivot to. Before you automatically pivot and keep going you have to ask yourself, “Do I even care?” If you don’t, you have to reconsider whether pivoting makes sense.

I’ve met quite a few people (it’s happened to me too) that get lost in what they’re doing and they invest so much into it that they actually forget why they got into the business in the first place. That’s a scary place to be.

Before pivoting, you have to ask yourself if you’re passionate about the direction you’re heading. If the answer is yes, pivot away! If the answer is no, you need to stop. Maybe you take a step back and re-evaluate, give yourself some time to breathe and think … or maybe it’s time to quit, admit defeat, lick your wounds and come back another day to fight the fight once more.

A Quick Checklist Before Pivoting

Let’s say you’ve got some validated learning and insights in terms of what direction to pivot. And you want to pivot and keep going. What else do you need?

1. Big vision. Prior to starting your company, you should have a big vision of what you’re trying to accomplish. Without that you can easily fall into the trap of building something small (a feature company) that never really has sizable potential. And pivots become meaningless if you don’t have some big, audacious goal ahead of you. A pivot is really about zigzagging towards your big vision.
2. A deep understanding of the problem. Most entrepreneurs I speak with genuinely don’t understand the problem they’re trying to solve and whether or not it’s worth solving. They haven’t dug into the problem enough. Or they’re trying to solve a universal truth. If you don’t understand the problems at their core, you can’t figure out how to pivot properly.
3. Actual (falsifiable) hypotheses. Validated learning isn’t enough. You need falsifiable hypotheses that you can test against, otherwise it’s very difficult to know if your pivot is going well or not.
4. Metrics and lines in the sand. A core concept in Lean Analytics is around picking the right number to track and having a benchmark to compare against. You need to identify your One Metric That Matters (a single metric that tells you how things are going) and draw a line in the sand–a target that you’re aiming for. If you miss the target, you re-evaluate; if you hit the target, you have the confidence (and data!) to move to the next step.

Pivoting is hard. Check that, pivoting properly with the opportunity of succeeding is hard. You can pivot aimlessly all you want, do it lazily and hope for the best, but you’re doing yourself a disservice. Anchor your pivots in honesty and data–give yourself a chance to find a business that’s going to be fantastic.

My fingers trembled as I fed page 34 of 72 into the fax machine, deftly pressing the head of each page into its creaky jaws so that this shitty cheap-o machine wouldn’t snag two pages at once, slantways, obscuring the precious scribblings adorning the footer of each page where it read: “Seller’s Initials: _______”.

This is what the last six years were all for. All the labor. All the risk. The brave face for the troops. The self-inflicted unflagging optimism despite no evidence to support it. All those sleepless nights worried about making payroll. The care and feeding of becoming an expert in something. The hard lessons you have to recover from learning. The experience you get just after you need it. The inner doubt suppressed for the morale of the team. No salaries followed by low salaries. The “eat what we kill” mentality. The scrounging and scrabbling and begging and fighting the assholes for those morsels of revenue, those crumbs of validation.

It’s over. We did it. I did it. American dream? Check.

The 73rd page spat out confirming the successful transfer of the previous 72.

And then…  sadness.

A profound sadness. Not depression — not hopeless or rudderless — but a pure sadness, when your lungs sink into your belly, the punch-in-the-stomach of discovering your dog was hit by a car or that your dad is terminally ill.

“What the fuck?” I thought, “Why am I feeling this? I’m supposed to be feel… happy? I guess? Something other than this.”

Almost all startup founders experience a deep and prolonged sadness after selling their company, even when the sale is an outrageous success. Why?

The answer is important and fundamental for all startup founders, whether or not they ever intend to sell their company.

“I sell my artwork on Etsy. Want to see?”
— barista at an Austin coffee shop

If you ask her, “Who are you?” She would answer: “I’m an artist.”

If you ask her, “What do you do for a living?” She would answer: “I’m a barista, but that’s just my day job. Want to see my artwork?”

Is she a barista because she pours coffee for money? Is she a driver because she drives a car to work? Is she a maid because she cleans her apartment?

No, she’s an artist because that’s what she really is. “Barista” is one of many necessary means to ends, where the “ends” are the basic human needs, followed by creating art.

A startup founder lacks this distinction between personal identity and work identity, and this is the key to the “sale-blues” phenomonon and other behavior.

A startup is the founder’s personal identity. Startups are not something you do to make ends meet or a “necessary evil” en route to what you “really” want to do.

A startups is an obsession. You do it because you couldn’t stop yourself. Because when you were doing anything else, you were thinking about it. That is the mark of “who you are.” Interviewers ask me “Why did you decide to do a second and eventually a fourth startup?” And the answer is “For the same reason that I started the first one — because it’s in my DNA and I have to do it.”

What do you do in your spare time when you have a startup? What spare time? This is all your time. It’s not just the last thing you think about before falling a sleep, it’s the thing that won’t let you sleep. It’s first thing that trickles into your brain in the morning like The Matrix patterns filling the void, and The Matrix is in fact your reality and there you reside.

It’s why you can weather the painful events like the ones whistling through my ears while I fed legalese into a rickety fax machine. You are consumed, this is your life, this is you. There’s no room for anything else.

When you sell your company, others are quick to throw jabs like “So, you sold your baby?”  Which means: “You’re a sell-out.”

It’s sort of like selling your baby. But it’s more like selling yourself. Although of course the business is physically separate from you, and its success is undeniably due to your faithful employees even more than it’s due to you, emotionally the business is not a separate entity.

Speaking of selling babies, this all sounds a lot like “baby blues” — depression caused by elevated levels of the enzyme monoamine oxidase A — that 70% of women experience after giving birth. A third of those women will experience this for up to a year (postpartum depression).

It’s characterized as a feeling of loss and of mourning. Which is seemingly at odds with the arrival of a new life which is just the opposite — gain and celebration. This intellectual dissonance creates secondary emotional effects, specifically the devastating belief that “I must be a bad mom for being sad that I have a new baby.”

Are the baby blues the same emotional effect as selling a company? Maybe not — I already don’t like saying “a startup is like a baby.” Besides, postpartum depression is triggered by the arrival of responsibility and, if you insist on the baby/startup analogy, selling your company is the departure of responsibility.

But one thing that definitely is the same is that “feeling of loss and mourning.” A piece of yourself has been eviscerated, irrevocably.

This is not only a normal feeling, it’s by far the majority case. A long-winded but high-quality study from Columbia Business School (Life after Exit) interviewed 22 entrepreneurs and every one of them experienced this effect. Some refocussed the energy into the Next Thing (almost always new ventures), but most took years to find that thing which could replace not only the excitement but identity, and many still haven’t found that Next Thing at all.

Selling their companies forced them to answer a difficult question: If you could do anything, what would you do? What’s really important to you, as opposed to a job? What do you really want to be when you really grow up?

The problem is, the startup already was that thing!  It’s a grinding, batshit-crazy, risky, irrational, epic adventure. You wouldn’t have done all that in the first place if that wasn’t already the thing you wanted to do. A startup is never the easiest career path.

Does this lead to the conclusion that selling a startup is the wrong choice much of the time? No. It’s true you shouldn’t sell without noticing whether you’re thinking “I can’t wait to start the next thing” or whether you’re thinking “I don’t know what I’ll do with myself.” It’s also true that you must resist the urge to believe that getting millions of dollars will make you fulfilled, or happy. “Money doesn’t make you happy” is cliché because it’s true.

But that doesn’t mean selling is wrong.

Just like a bad relationship that has to be ended even though it’s painful to do it, especially if you genuinely love the other person, just because it’s emotionally painful doesn’t mean you don’t need to do it.

Building a company in year one is completely different than building that company through year five or ten. The CEO’s job description changes over time, and so does the company, whether you sell it or not. Are you emotionally prepared for this as well?

Are you OK with innovation taking a back seat to developing scalable, mature processes? Are you OK releasing control in day-to-day operations to managers, and then releasing control of the managers to your executive team? Are you OK wresting yourself out of TextMate, out of Adwords, off the website, off the live chat, out of the sales calls, trusting your managers and not being that sort of meddling micro-manager boss that you yourself hate? Are you OK shouldering the burden of the livelihoods of dozens of families rather than just “pulling 90 hour weeks” to push some code out the door with a co-founder?

The fact is, startups grow up. They grow into businesses and strap on the shackles of sustainability, risk-avoidance, HR law, strategic planning, executive meetings, and all that. The founder-CEO is still steering the ship but it’s a different sort of ship.

Is this bad? The answer to questions like that is always “it depends?”

In my case, it changed over time. At Smart Bear I didn’t want to lead a huge company. I didn’t want to relinquish Eclipse and my ability to check in code. I didn’t want to manage managers or figure out what changes, strategies, hirings, products, marketing, and sales were needed to make $100m/year. So for me I sold at a good point: before I needed a C_O, but after the company was big enough to garner enough money to cross the Freedom Line.

Now at WP Engine, I have new ambitions and inclinations.  I am now that CEO who manages managers, who sets vision and direction but not day to day operations, who worries about company culture but who doesn’t have SSH access to all the servers, and who is driving towards a company with products and a market and a team which we believe can indeed generate $100m/year.

That’s exciting to me. This is my new challenge. I will always love writing code and getting a company from $0 to $1m/year.

But, today, right now, for reasons unknown even to me, this is who I am.

I went to find the pot of gold
That’s waiting where the rainbow ends.
I searched and searched and searched and searched
And searched and searched, and then—
There it was, deep in the grass,
Under an old and twisty bough.
It’s mine, it’s mine, it’s mine at last….
What do I search for now?
— Shel Silverstein, Where the Sidewalk Ends

They told me I couldn’t sell to The Enterprise with a silly company name like Smart Bear.

They told me I couldn’t sell to The Enterprise without a human sales force using Salesforce.

They told me I couldn’t sell to The Enterprise over GoToMeeting, with a demo and no slides, from a geek with no webinar.

They told me I couldn’t sell to The Enterprise with amateur design and a small-company, human voice.

Oops. I accidentally went and did all that at Smart Bear, and only had millions of dollars of revenue to show for it. Largely from The Enterprise.

Does that mean “They” are wrong? Nope. “They” are in fact describing the way most successful companies sell to The Enterprise. Nothing wrong with “tried and true.”

Unless tried and true doesn’t happen to jibe with you. Unless you can’t bear to hire a traditional “enterprise sales guy.” Unless you can’t stand content-free language and you choose honesty over formal “corporate” language on your website. Unless you prefer excitement over ROIs. Unless you’re happy to compete with big companies instead of becoming one.

I wasn’t trying to be “disruptive” or “innovative” or a “maverick.” I just couldn’t do it. Maybe you’re like that, about something.

You’ll have a harder road if you ignore “Them” because it leads to gut-wrenching awkwardness. But at least you’ll be different. Sometimes just being different is enough, if it’s for good reason.

Who knows. Do it your own way, but you’d better have your eyes open so you can address the special challenges you’re bringing on yourself. Once you buck “Their” wisdom, you can’t rely on “Them” when you get yourself into an unusual pickle. You’ve exited the safe and well-travelled road.

But go ahead and do it. Don’t let the “it’s not done that way” people win on that basis alone.

Maybe it should be done that way.

This is part of an ongoing startup advice series where I answer (anonymized!) questions from readers, like a written version of Smart Bear Live. To get your question answered, email me at asmartbear -at- shortmail -dot- com.

Frustrated Engineer writes:

I’ve been writing code for ten years, recently promoted to a position where I have to talk to our non-technical customers.

They want to use our software or customize things in ways that are technically impossible, but I don’t know how to communicate this to them. They don’t even know how little they know about software, so how do I basically tell them “You can’t understand this. You have to trust me” without sounding like a prick?

I like a doctor analogy.

The doctor says you have a rare liver disease and you’ll need to treat it with a dangerous cocktail of drugs. How do you evaluate whether the doctor is right?

For almost all of us, the answer is: You can’t.

You can’t ask for details, because you’ll get a torrent of nomenclature that’s greek to you (much of which really is Greek, or Latin), body parts you never knew existed, drug names you can’t spell with interactions you’ve never heard of, and references to studies you can’t read.

Even if she explained these details, you can’t evaluate the context — which of those things is dangerous, or unusual? Which are easy or hard to treat? Which are easy to deal with in isolation but complex when they interact? What other things isn’t she saying because in her expert mental process she takes them for granted?

You might try to evaluate the doctor herself, rather than her diagnosis, but here too your lack of expertise leaves you helpless. You can’t grill the doctor on medical knowledge — even incompetent ones know far more than you and can easily overwhelm you with terminology and information. You can’t ask for references because of course she’ll hand you good ones.

In the end there’s nothing you can do, because this is a field that takes years to master, in both education and real-world experience, the complexities and context of which cannot be satisfactorily transmitted to even an intelligent layman.

You’re going to have to trust your doctor.

That’s what it’s like talking to the architect of a software product containing a million lines of code. It’s not that the customer is “stupid,” nor that given enough time, training, and explanation, couldn’t eventually understand it all fully. But sometimes the customer just has to trust the vendor.

One way around this with doctors is to get a second opinion. You still can’t make the call, but at least you can get another expert on the case. Of course if the experts differ, you’re still stuck. Maybe a third opinion?  And even then they might choose the wrong course.

The problem with software is that other experts don’t know the intricacies of the software product in question.  They don’t know what’s in those millions of lines of code, or the trade-offs made by that product’s creators. The vendor might indeed be incompetent, or shrouding themselves in terminology to avoid helping the customer, and a second opinion isn’t an option. That’s unfortunate.

That’s why this comes down to trusting your software vendor, just like your trust your doctor. If you don’t, no one can help you, because you don’t believe what they say and yet cannot evaluate what they say. If that’s case, sounds like you need a new vendor.

So, as the vendor yourself, I would say (and in fact have said) all of the above. Put yourself out there. Make this a moment where you build trust.

All good business is built on that — trust, which is then not abused.

Add your advice to the discussion section!

We looked at all the products in the space and thought, “We can do it better.” So we did.
                       – most founders

It’s a common refrain, and I suppose this is what all founders should think! After all, if your product is inferior in every way to even one incumbent product, why do it at all?

Still, this is always an invalid “idea” for a business.

To see why, consider a home builder who says “I looked at all the other houses in this city and thought, ‘I can do better.’”

Better for whom? A 19-year-old bachelor who never cooks, skipped college, and has no money?  A millionaire with a wine collection?  A family of six?  A family of rats?

So when someone says their product is “better,” my next question is always: ”Better for whom?”

You quickly realize there’s no such thing as “better.”  There’s only such thing as “better for a specific market segment.”

What spurred this blog post was someone who wrote exactly what I quoted above, in the space of project-management (PM) tools. It makes sense — almost everyone dislikes some aspect of their PM tool (which might also explain the proliferation of such tools).

You can already imagine the pitch, a variant of the generic “it’s better” like “All the things you love about Microsoft Project, but none of the things you hate.” Uh-huh. Who’s “you?”

Here’s some examples of pitches which would make sense.  Who knows if they’re good ideas, but at least they are actual ideas, actual proposals, in that they define “better:”

• PM tools assume you already know about project management.  My PM tool is best for people who have been thrust into a PM position but need help on nomenclature, process, best practices, and how to get the most from their tool and their team.
• Modern PM tools are too simplistic because they follow the 37signals and Lean Startup mantra of building the simplest possible thing. But that means they’re no good for large, complex, distributed, multi-user projects. Older PM tools understood that use-case, but they’re clunky install-only implementations where online collaboration, mobile and even the Mac is a bolt-on afterthought.  My PM tool is best for large, complex projects with modern technology requirements.
• PM tools were built in the installed- or web-based-eras. My PM tool was built from the start for the iPad only. It’s the best if you need to manage projects from meetings and on the go.
• PM tools are built for a small, fixed set of workers.  My PM tool is best for distributed, open projects with thousands of members but which still require central management.
• PM tools assume your set of tasks, goals, and projects are relatively fixed.  My PM tool is best for open-ended projects where requirements change constantly and the tool should change with it and even measure the change.

The other thing you should notice about these “better for whom” statements is that they naturally point the way to the next steps in the business:

If you know your audience, you can create targeted advertisements, go to specific conferences, and even cold-call them on LinkedIn. You know what words to put on your homepage. You know which features will especially tickle their fancy. You know what to highlight to differentiate from competition.

Or if you’re still in the ideation stage, it gives you the basis for customer development, both in finding potential customers to call and in what you’re verifying when you get them on the phone.

Surely your product is “better” than the rest… for someone. Who?

An entrepreneur confided in me recently that he didn’t like Lean Startup; did I think that was OK? Or is he not cut out to build a startup?

A founder uses “37signals does it” to justify every decision, even if he can’t explain why, or why it applies to his situation, or whether 37signals did that when they were still his size.

A startup pitch-deck boasts about the mistakes they’ve made because of the lessons they’ve learned. But the lessons haven’t resulted in forward motion. Time to hang up the towel, or does this knowledge and introspection position them for rocketship growth?

It’s easy to confuse philosophy with dictum, musing with mandate.

A compass is not a map. A compass tells you which way is north, not whether you should be heading north.

There’s no map in startups. There’s examples, but for every startup that landed 300 customers from their TechCrunch launch article there’s another which got 0; for every startup that built loyalty on Twitter there’s one with 20% month-over-month revenue growth that has yet to reserve their Twitter handle; for every startup ascribing their success to spectacular design, there’s another successful one who has never hired a designer (and it shows).

The trouble with “rules” in startups, besides the inherent survivor bias, is that by definition the successful ones are anomalies. Statistics show trends, but trends don’t predict outliers.

What does predict an outlier? Nothing. A successful startup looks a lot like an unsuccessful startup at the start. Statistics can’t be used to determine the outcome of the individual case (not just in startups, but anywhere).

Don’t confuse dogma or startup frameworks with laws. You have to find your own way, using all this startup advice for sparks of inspiration, for brainstorming, as a candy store where every item has its own merits but you must pick and choose what to put into your basket.

There’s no map. That’s OK. Keep moving.

When people can devise their own lifestyle, career, and mobility more than at any other time in history, it’s interesting to ask how a startup can support and encourage its employees beyond a paycheck.

When job-hopping is a badge of honor instead of a scar of disloyalty, it’s interesting to ask how a company can retain talent over significant stretches of time.

WHen “friend” means one of 852 on Facebook, it’s interesting to ask what a friend really is.

John Price (CEO Vast) mused on this last one: “A true friend is someone I can telephone at 2am, out of the blue, ask them to bring $20,000, no questions asked…. and they do.”

I vividly recall the moment when I connected on this level with Austin Gunter — our social media keymaster (are you the gatekeeper?), the most online-visible person at WP Engine. I was about to give a night-time talk about making websites fast at a local Austin PHP users group; did anyone at the office want to come? I’ll buy dinner after? Bueller?

They didn’t. But I looked at Austin, and he gave a smirk and nodded. “Yeah, I’m in. I’ll film it too.”

It was during “Final Four Week,” which I didn’t know because I don’t follow basketball. But Austin knew, and had plans to hit a sports bar with friends that evening. But he silently delayed that to see me wax on network latency and reading performance-analysis reports.

It was recorded too.  (See it here if you’re into such things.)  (Does anyone not hate what they look like / sound like on camera?)

To me, that is what a friend is. That’s my equivalent of the $20k at 2am. Supporting your peeps because you know it sucks to do stuff like that alone.

So when Austin said he wanted to move to San Francisco from eponymous Austin (Texas), supporting his move was a no-brainer.

But wait, doesn’t that mean a pay raise to cover cost-of-living? Yes. But wait, isn’t that an HR no-no? Who cares. But wait, what if Austin finds a hot new startup in San Francisco and bails on us?

Well, devotion works both ways.

We always talk about the benefits of startups in terms of “having no boss” and the thrills of the roller-coaster and failing fast and building something that never existed. Not to mention the potential to make more money than we could ever spend.

But that prose is aimed at the founders, not employee #8. That is what the company can do for itself and its founders, not what the company can do for its employees. 

We dwell on how startups enable founders to quit their day job and master their own destiny. But what about everyone else? They quit their day jobs too, perhaps more secure, perhaps with better benefits, perhaps with more family time, perhaps with less stress, perhaps with fewer hours. And yes, employees are getting the checks instead of writing the checks, and that does make all the difference, but still. What about what they can control, how they want to grow and learn, how their personal goals might be fulfilled?

Step one is to create an environment where the culture and comfort and excitement is better than 99.9% of any other job. Good, but you’re not done.

A startup must be an enabler, otherwise you’re just building another big company, exactly like the one you as a founder refused to devote your life to. In 2013, in the tech world, with our opportunities and capabilities, we must do more than just build another big company.

I just got an email from our VP Marketing about an interesting conversation she had with Austin and, now that they’re both in San Francisco, that’s happening all the time. Austin was quick to point out the various benefits of our “social media and community guy” being in the bay area and in close proximity to several of our other team members not to mention the heart of WordPress and a nerve-center of designers and marketing agencies.

I suppose I could “justify” the cost of living salary increase using these incidents as “business cases,” but why should I rationalize? Besides, as a founder I don’t have to justify anything to anyone, right? Isn’t that one of the joys of being a founder in the first place?

The truth is it really is a net positive for the business and for Austin both, because Austin is a critical part of our team, and if he’s happy, the benefits accrue to us as well. And if a startup can’t support that, what’s the point?

So welcome to San Francisco, Austin. We love you. Let’s do this.

Don’t be a stranger.