If I had a nickel for every time I accidentally proved someone’s ‘gut feeling’ wrong in the first minute of a discovery call… I’d have many a sad nickel in my possession.
But it’s never been my intention to strip people of their confidence. Instead, it’s my main intention to help organizations create a culture of confidence through data-driven decision making. It starts with giving the good ol’ gut feeling a sizable gut-check.
“We don’t need data, we already know which aircraft require the most repairs and maintenance.”
You’re going to want to finish reading this example, especially after realizing that the “Head of Repairs” for a medium-sized airline made that comment to me, directly. This isn’t anecdotal folks… there are data-dismissive folks putting you in metal tubes and rocketing you through the air.
Let’s begin.
I recall a sales call not long ago with a medium-sized airline which needed to meet their FAA mandate by providing the agency with critical details on their repairs. Their project was simple, the use case straightforward, and we were nearly set to move into signing and kicking off their project.
There was one obstacle - the “Head of Repairs” did not believe in the value of such a platform to help their team make data-driven decisions. Their mechanics “...already knew which aircraft had the most repairs, where the issue areas were, and having better data wouldn’t change that”.
So as any concerned individual would do, I had them send over 5 different files loosely pertaining to their last 6-months of repair and instantly added the datasets into our partner DOMO instance. I began dashboarding using their data during the call.
Once I had the data in front of me, I asked some questions.
These were the same questions the business had mentioned the FAA wanted answers to on a consistent basis. And as you might have guessed, the “Head of Repairs” got every single question wrong - not completely wrong, but definitely wrong enough to pique the interest of a government body overseeing aircraft and passenger safety.
Is now a good time to remind you, the reader, that many of these aircraft are flying over your city as we speak? Nah, you probably don’t need that reminder. Everything’s fine without data… everything’s fine.
Jokes aside, there is often a lot of value to be found in tribal knowledge, especially in companies that are still in the process of transitioning to be more data-driven. But over-reliance on tribal knowledge results in an increasingly tribal organizational culture that becomes less and less data-supported over time. That’s a trend we want to reverse, or ideally, prevent.
No! The problem isn’t that people have gut feelings. In fact those gut feelings are often crucial!
The problem is that more often than not we see businesses depend on the gut feelings of their experienced individual(s) in order to get work done. Over time, fundamental knowledge of how the business operates becomes known only to those few, who “work their magic” without anyone else knowing how. Eventually those people retire, quit, (or are fired!) and the business is stranded without the “magic insight” they unwisely depended on for the livelihoods of everyone working there. The business almost literally has to reinvent the wheel.
Imagine being the hiring manager trying to replace that person.
“Hey Jenny, we need you to find someone that has 10+ years of experience, can operate independently, has a real ‘self-starter’ vibe… oh and they must have *magical power*. What’s with the incredulous face? Oh just shortlist anyone with the name ‘Merlin’! Great, thanks Jenny!”
You might think that’s a silly example, but I promise you that the HR managers reading the above are high-fiving each other and yelling “SOMEONE GETS ME, THEY FINALLY GET ME!”
Gut feelings as a problem statement
Problem: In most cases, misuse of, or over-reliance on gut feelings will eventually cause businesses to fall short of their goals or fail completely. But gut feelings are useful sometimes! How do we use them constructively?
Solution: Gut feelings are best used as a starting grid, not the finish line. Gut feelings should be used as part of the scientific method - the hypothesis that acts as a cue to begin data exploration, discovery, and planning.
Well, it isn’t complicated… but it does take more effort than people are typically used to because it’s so easy to “misuse” those gut feelings and allow them to take you down the path of least resistance. As an old mentor of mine once said: “Nothing’s easier than doing nothing.”
And relying on gut feelings alone is as close as you get to doing nothing.
So, let’s talk about what it looks like to put in the work.
No one has a crystal ball… yet. It doesn’t matter how much experience, clout, ego, etc that you carry with you. Everything you feel is, at best, an educated guess. Instead of making bold claims that set the foundation for extreme confirmation bias and sunken cost behaviors, try setting up your gut feeling as a hypothesis.
Bad: “We will have 5M USD in sales this year.”
Good: “If we increase our marketing spend by 200% and our cold calling efforts by 100%, we will increase our sales to 5M USD this year from a starting point of 1.2M USD in Q1.”
The anatomy of the statement matters here.
You need the following ingredients:
This might feel remedial, but it’s worth reminding you, the reader, that a hypothesis needs to have supporting data to be proven - Grade 6 science fair, anyone? It should be familiar! We’re using data to do business scientifically.
So what kind of data can we use to support our hypothesis?
We connect the ingredients from above.
Part 1: A proposed action that is attributable to the desired outcomes
This is arguably the most difficult part of this whole process, but it is inarguably the most important.
If a member of an organization is going to propose some ‘great new idea’, they will need to propose how they will attribute the proposed action to the successes they posit will come in part 2. Without a measurable way to directly connect the action to the outcome, there is simply no way to tell if they are in any way related. The process will be about as scientific as a dowsing rod or rain dance.
Therefore, any proposed action needs to be extremely well-defined, actionable, and (most importantly) attributable to the outcomes. This requires the proposer to think about the key metrics that they are expecting to be tracked.
Now let’s discuss why our earlier example includes a proposed action that isn’t good enough:
Proposed Action: “Increase our marketing spend by 200% and our cold calling by 100%”.
It might be surprising, but connecting your software/platforms/systems to BI platforms is actually the easy part most of the time as long as you have connectors, which are add-ons that work like adapters, making one thing compatible with another. In fact, one of Bear Cloud’s specialties is making connectors for our platform of choice, DOMO. So whatever you’re working with, you can have confidence that there’s a connector for it, and if there isn’t, we’ll build one.
I’ll end this section with the following:
Part 2: Supporting an estimated outcome with data
This part always feels as if it should be the easiest data element to provide.
That said, I’ve been part of consultations at organizations where someone proposes a new course of action that the data conclusively does not support. We see this all the time with organizations which have a ‘too big to fail’ mentality.
These organizations entertain every proposed action, chase every suggested possible outcome… and it often feels a lot like complacency.
Now, if someone in an organization is suggesting that the business should support a specific action, the first question a sensible proposer should expect to be asked is “why should we do that, and what is it for?” Therefore, a good hypothesis should always be front-loaded to include what the expected outcome should be if we follow the proposed action.
In our earlier example we focus on a lofty estimated outcome:
“We will grow our sales by 3M USD to a total of 5M USD.”
In this example we would need to understand exactly how much of an increase in sales would be the result of the increases in marketing spend and cold call. We would need to understand the selling trends in different regions, the number of customers we believe we can recover, and more in order. to understand where we should be directing our efforts.
For example, if we keep things as simple as having the business perform the exact same marketing and cold-calling activities we are currently doing, with the only change being an increase in spending, we might actually see diminishing returns and marketing oversaturation that stalls or even reverses our growth.
The expected outcome needs to be supported by strategy and vice-versa.
And how do we identify the right strategy? Data.
As leaders in/of companies, we are expected to think of new methods to make a company better, more efficient, more cost-effective, more profitable, etc. Whether we’re shooting to make the company an extra few hundred million in revenue or reduce support calls to lessen our breakfix budget for next year, we need data to come up with the strategy to support those hopeful outcomes.
Part 3: A measurable change from a fixed and known starting point
When a proposal hits my figurative desk for ‘the next big way we can increase X’, the first question I ask is “where do we stand on that now?”
If a proposal is worth a discussion, the proposer will have already done the legwork including:
“Where do we stand on that now” has always been a question that served me well.
The answer to that question is either an incredible conversation starter or a telltale sign that the proposal isn’t ready for a discussion.
But WHY do I ask this question, specifically?
It’s about vetting and setting.
While the other anatomic parts of a hypothesis help me to understand the reason behind the proposal, this last part is entirely about vetting and setting the priority.
Vetting: If the proposer does not know where we stand now on this, how could they possibly understand the value of their expected outcome.
“I want us to reach $500,000 in sales!” is a statement much less exciting if we’re already at $499,999.
Setting: If I have multiple proposals in front of me and limited resources to commit, where we are now will help to dictate the priority of these proposals.
As a business leader, if a proposer is telling me that we are at $0 in sales and they can raise it to $1,000,000, then that would be my number one priority, even if other things may be lacking. But, if a proposal reached me to increase our annual sales from $4M to $5M but our costs are outpacing our revenue… I would gladly take a cost-saving proposal as a priority.
That concludes this section on ensuring that there is data to support your hypothesis.
Quick Recap on the three parts we need:
Once you have the big three parts of your hypothesis, you’re ready for step 3.
Not if. Your gut instinct isn’t always right.
If we can all get behind that as a frustratingly simple starting point, this article has (mostly) done its job. In fact, when we all admit to each other, and ourselves, that questioning gut feelings isn’t an attack on our competence, we begin the journey to creating a data-driven culture.
But throughout this article I’ve maintained that gut instincts are still valuable. It’s often true that your gut is responsible for “getting the ball rolling” and making things happen. When you think back to the good leaders you may have encountered in your life, you’ll instantly remember the times when those individuals made quick decisions that dug people (potentially you) out of a tight situation.
That super power isn’t something I’m against. I want to help you use that power responsibly, like Spider-Man has taught us all. With that in mind, the goal of this article is to show you how you can use it more wisely and much more effectively by using data as a means of ‘training’ your gut instinct. We can’t escape the fact that there will always exist some situations where time is not a luxury we have, but even in those situations someone used to basing their decisions on data every day will naturally be more consistent than one who isn’t used to it.
At the very worst, when we draw the wrong conclusions from data and make a bad call, we gain more insight on the type of decisions we’re great at and whether or not we have areas that need improvement. That’s why it’s crucial to consistently review your data and decisions and learn from the failures as well as the successes.
Think about your gut instinct then as being an important tool for both a) pointing you in a promising direction and b) helping you act decisively in time-sensitive and critical matters. And of course, it can have other soft benefits, too. I think most of us would agree that having a good gut instinct shows confidence.
Just never let it be the only tool in your toolbox! Never stop checking in with your business to see “where it’s at.”
Remember the process:
If you’ve made it this far, my gut instinct tells me that you’re genuinely focused on creating value for your business and your people. If you’re unfamiliar with our services, you should know that Bear Cloud Studios shares the same focus.
Our data, analytics, and business intelligence solutions focus on creating data-driven cultures in organizations. From live speaking engagements, to data pipelining, to business intelligence and AI/ML solutions, we do it all.
If you’re looking at turning your leaders into data-driven, decision-making machines, we’d love to talk with you about your goals. If you’re curious about how connectors can open up BI as an option for your business, we’d love to answer your questions.
Either way, keep checking out our site, since this article is just the beginning of a weekly column I’ll use to share my knowledge, experience, and perspective on BI and related topics. I have a gut feeling you’ll be back! But as always, I’ll be keeping an eye on the data.
