Numbers crunching & success: How to create a successful startup growth plan using data PlatoBlockchain Data Intelligence. Vertical Search. Ai.

Numbers crunching & success: How to create a successful startup growth plan using data

Editor’s note: Joe Procopio is the Chief Product Officer at Get Spiffy and the founder of teachingstartup.com. Joe has a long entrepreneurial history in the Triangle that includes Automated Insights, ExitEvent, and Intrepid Media.He writes an exclusive column about entrepreneurship for WRAL TechWire. His columns are published on Mondays as part of TechWire’s Startup Monday package.

+++

RESEARCH TRIANGLE PARK – Growing your business is not rocket science. For most startups, the difference between success and failure is the difference between fumbling around in the dark and following a well-lit path towards growth.

No one is going to light that path for you.

In over 20 years of building startups using data to create effective, repeatable strategies for growth, I’ve learned that each path is unique to each business. The biggest mistake any entrepreneur can make is abandoning what they know to be true about their own business to follow someone else’s can’t-miss growth plan.

Joe Procopio (Photo courtesy of Joe Procopio)

If you can build your startup into a viable business, you have everything you need to create traction and scale that business.

Here’s how to do that.

If you want to scale, let data be your light in the dark

I’ve seen it a million times: A founder will build a startup to some initial success point and then freeze — unsure of exactly why their customers are so enamored with their product or service.

Last week, I wrote a post outlining the most critical mistakes startup founders and leaders make when faced with the task of scaling their initial success. Most of the time, those founders and leaders have the right idea — using data as a guide to determine the direction and magnitude of their next move. The problem is almost always in the execution:

  • Keeping too tight a grip on that initial success and letting new opportunities slip away.
  • Listening to the wrong signals and chasing unproven theories.
  • Letting an abundance of optimism or pessimism cloud the decision-making process.

Anyone can tell you that you should be using data as your light in the dark for growth. So how do you make sure that you’re using it properly? I’ll restate the DON’Ts I wrote about in the previous post and give you actionable strategies to execute instead.

DON’T Do This: Ride any wave too long

The biggest mistake a startup founder or leader can make is to analyze all the data around the company’s initial success, look at only the positives, and decide to stay the course. Nothing lasts forever, all good things must come to an end, and if your business is growing, there is no upper limit on where your numbers should be.

Do This: Always be experimenting

You should be in a constant state of controlled experimentation with your product, your positioning, your market fit, your pitch, and your messaging. You don’t need wholesale changes with every new version or change, but you do need to take several steps into the darkness to see if you’re going to stub your toe, so to speak.

A reader asked: How much time should I devote to creating reportable data from an MVP? My answer is “All of it,” or at least as much time as you can. An MVP without a tracking mechanism on every interaction, from initial discovery of the business to closing the sale, is just a very expensive way to fumble around in the dark.

It doesn’t matter if you’re selling SaaS software or gardening tools. Every touchpoint in the discovery, transaction, and usage of that product should be tracked, automatically or manually, including when the interaction happened, how it happened, what the result or next step was, and what that result or next step means to revenue and costs.

You should track every data point and let the results sort themselves out. I can’t tell you how many times I’ve asked a founder if they were tracking a data point and the answer was no and the reason was they didn’t feel like they needed it.

If there’s one thing I’ve learned about product-market fit, it’s that you don’t know whether or not a data point is important until you can empirically prove that it isn’t. You can’t prove that until you track it. The only caveat I’d add is you have to draw the line with effort. If a data point is too costly to track, you may need to guess.

Finally, I’ll add that you should balance how many experiments you’re doing at once. I’d recommend always doing more than one experiment at a time, because when you’re trying to scale, time is always short. But one thing to consider is making sure that the impact of one experiment doesn’t cloud the results of another.

For example, if you’re adding a new feature, be careful how drastic a change you then make to your messaging. If your new feature is awesome and your new messaging sucks, you’ve just given yourself a false negative.

DON’T Do This: Kill the cash cow

Of course, the opposite of analysis paralysis is a wholesale shift that abandons the gains from the initial success in the name of growth.

A classic example is the startup that attracts millions of customers for a free product (say, content), and then sees dollar signs if they charge those “customers” a small price for the same product (say, $1 a month). Two things usually happen and they both come as a surprise:

  1. The vast majority of those “customers” won’t convert.
  2. The cost to serve the new paying customers turns out to be far more than the revenue they generate.

Do This: Look for green shoots

Massive oak trees don’t appear overnight. They start with green shoots. When you make any change to how your business operates, you’re going to foster some negativity in your current customer base. Instead of clear-cutting your forest and being shocked when nothing grows back, replant a single tree first and monitor how the new growth happens.

Those measurements should always be based around revenue and retention. When you make changes to your product or service, you’re trying to increase your revenue and keep both your new and old customers longer.

When you run your experiments, hypothesize the expected results. In other words, if you make a change, that change should result in X% new customers paying Y% more in Z% less time. Then hypothesize the impact to your existing base: We plan to lose X% of our customers and those customers should be no more than Y% valuable to us.

Abandon failed experiments quickly. You don’t have to cut them off without warning, but be able to undo them, bring them back in house, and tweak them until you fix those percentages. This is especially true when you’re losing more customers than you expected or losing customers that were more valuable to you than you anticipated.

DON’T Do This: Give up the macro for the micro

Just because an idea doesn’t work doesn’t mean it was a bad idea. Small changes in your data, good or bad, don’t require sweeping action. To get metaphorical again, you don’t build a skyscraper on a foundation that hasn’t set — and you don’t tear down a skyscraper because the roof is leaking.

Do This: Act on patterns, not on data points

To scale, you need to define your success as revenue minus cost and repeat and expand. To grow, you need to define your success as lifetime value of a customer (LTV) minus cost to acquire a customer (CAC), and expand.

One bad data point, one bad customer, one failed relationship, can throw your trend line off, but it might not dictate the trend itself. Same on the other side. One great customer doesn’t mean the experiment worked.

So when we talk about risk in entrepreneurism, risk isn’t making the next bold move, or pivoting in a direction no one is expecting — that’s gambling. Risk is deciding when a pattern is emerging based on a limited number of data points.

To answer another question that came out of the last post: How do you successfully utilize anecdotal evidence when it doesn’t pass any sort of significance test?

This is the difference between being a good entrepreneur and a bad entrepreneur. And that comes down to risk and mitigation. You, as the owner of the idea and the leader of the execution, have to make that risk/reward decision in a timely manner, based on the patterns that you recognize.

Any entrepreneur can sell a good product. Not too many entrepreneurs can recognize a great product.

Getting to conclusive data is the hardest part of data-driven growth. But once you get there, it’s almost automatic. Once you have confidence that you can get $X amount of LTV for $Y amount of CAC, that’s when you push the accelerator.

Filling the gap between confidence and conclusive is what makes a great entrepreneur.

+++

Hey! If you found this post actionable or insightful, please consider signing up for my weekly newsletter at joeprocopio.com so you don’t miss any new posts. It’s short and to the point. Or if you’d like more tactical startup advice direct to your inbox, get a free trial of Teaching Startup.

More from Joe Procopio:

Data crunching & your startup: Don’t let it destroy the dream

Time Stamp:

More from WRAL Techwire