This post was originally published on Medium
The proverb, “See the forest for the trees,” celebrates individuals that can see the big picture while at the same time disparages those that focus on the little details — the trees.
In my recent Medium article about how it is possible to see both forests and trees simultaneously, I posit that high performers should seek to develop this skill.
The next logical step is to wonder about what it means to think this way. Proverbs can be cute, and sometimes even inspiring, but generally fail to provide practical application for real-life situations.
What does it mean to see the forest for the trees? The following is my attempt to bridge this gap for you.
When we discuss thought processes — seeing forests vs. seeing trees, for example — we are really discussing perspective, decision making, reasoning, and logic.
Once we break down these topics, we can finally understand how they lead to creative problem-solving.
There are two classical schools of reasoning: deduction and induction. A third, abduction, is a sophisticated hybrid of the two.
Deduction, the method used in deductive reasoning, is generally defined as “the deriving of a conclusion by reasoning.”
Deductive reasoning leaves no room for uncertainty; all decisions are based on factual evidence and universally agreed upon conclusions.
If we agree that:
…then we can deduce that all finches have feathers.
If we agree that the definition of a sandwich is “two pieces of bread or a split roll with a filling in between,” then we must deduce that a hot dog is a sandwich.
Deductive reasoning is a top-down approach. It starts with a broad, but true statement then moves to a more specific but true statement (called the major premise). Then we can use that knowledge to make a more specific and also true statement (the minor premise).
In our first example, all birds have feathers is our true statement and all finches are birds is our major premise. Our conclusion, or inference, is that all finches have feathers.
If deduction is the top-down approach, induction is the bottom-up approach.
Inductive reasoning is the act of using individual observations to make generalized inferences or theories.
For example, Charles Darwin famously observed various species of birds in the Galapagos to draw his theory of evolution.
Suppose you were waiting in line at Shake Shack — a fast-casual burger restaurant — and overheard multiple customers in front of you each ordering the chicken sandwich. In that case, you might infer that the chicken sandwich is a better option than the hamburger.
Key Takeaway: Deductive Reasoning uses generalized ideas to reach specific conclusions; Inductive Reasoning uses specific ideas to reach broad conclusions.
Deductive Reasoning is a principle often taught in business school, mainly to finance or management majors, as a practical method for improving organizational operations. It is effective at identifying problems within business processes and often helps point to the best solution.
In an article titled, “Learn Why Employers Value Deductive Reasoning, and How You Can Show It,” Anastasia Belyh describes how a business might use deductive reasoning to determine budgets for business expenses, such as the gas budget needed to power a delivery truck.
The process starts with asking many questions, such as:
How many deliveries does the truck make each month?
What’s the total number of kilometers traveled by the delivery truck during each trip?
How much gas does the truck require per kilometer?
Has the number of deliveries been consistent for the last six months?
You can then use this information, along with the current price of gasoline, to reach your desired budget.
Belyh describes other operational problems that deductive reasoning can help solve, including one that hits home for me personally: resolving client issues.
If a client were to express that they were unhappy, we could ask them a series of questions about why they felt this way. Belyh notes:
The client might note that he doesn’t know the status of the project or the milestones that have been reached. He is not even sure whether your team is still working on the project. He probably sent an email to a member of the project team and got no feedback.
Using the information given by the client, you can form two premises. The major premise is that the client is not happy. The minor premise is that the client’s unhappiness stems from the lack of communication by the project team.
From this, you can deduce that regular communication between the client and the communication team will keep the client happy.
Deductive Reasoning is a tactic often used by outside consultants brought in to improve organizational operations. A consultant can ask a lot of questions and then provide recommendations based on their observations.
While this appears to be an efficient and fast way to uncover opportunities for improvement, there are tradeoffs. The consultant chooses this route because they do not have the time to fully immerse themselves in a business to the point where a more well-rounded, creative solution could have been formed.
There are two gas stations in my town; both are right next to each other. Drivers can see both locations and the bright signs displaying their current gas prices from either direction.
One of the stations is always two or three cents cheaper than its neighbor. This is where I choose to fill up, and it’s typically busier than the more expensive option next door.
But some people still willingly choose the other station! This absolutely blows my mind, as rational drivers have enough time to read both locations’ prices before deciding where to pull in.
No amount of Deductive Reasoning could ever explain why some people choose the more expensive option. Trust me, I have been attempting to solve this mystery each morning for the last two years.
Deductive Reasoning leans on universal truths; there is no room for uncertainty. It is not enough to say that most drivers make their decisions based on price, while some drivers care more about other factors (or just don’t care at all).
While Deductive Reasoning can be useful when solving operational or financial problems, it cannot be used to solve complex marketing problems. Why? Because marketing problems are at the mercy of people, and people do not behave in ways that can be defined as logical or rational.
As discussed, induction is the bottom-up process of using specific ideas to reach broad conclusions.
As it relates to my previous article, inductive reasoning is a way to describe the act of seeing the trees.
Practical inductive reasoning requires a handful of skills:
Jack Maple is a legendary New York City police officer known for his obsession with the small details. Maple, a transit officer, would study the intricate patterns of people and crimes committed near the 42nd street subway station and eventually predict when and where certain crimes would take place with extreme precision.
Inductive Reasoning takes small sample sizes and draws larger conclusions. This, by itself, lends itself to bias and other risks that could lead to faulty conclusions.
In a marketing/advertising environment, I have often found that induction leads to Survivorship Bias* — drawing false conclusions from incomplete datasets.
*I wrote about Survivorship Bias in PPC campaigns here.
These biases can cause organizations to lose focus the entire reason that the phrase see the forest for the trees exists in the first place.
I was recently analyzing a client’s website and was shocked by the number of available conversion options. Users experience a half-dozen forms and calls-to-action that crowd every page: They can purchase products directly through an ecommerce cart, request a free sample, or receive a custom quote.
They can also fill out a more vague Contact Us form or chat with a sales representative through a live-chat window. You can also call them directly; there are three different phone numbers listed throughout the site.
If none of these options appeal to you, you can email them directly at a sales@ email address listed at the bottom of the page.
The amount of choice presented here is both confusing and paralyzing. When I asked the client about it, he said they received conversions from each of these options. “In fact,” he continued, “some customers have specifically said that they liked that we had a chat function, so I am afraid that if we removed any options, it would have turned them away from the site.”
While he makes a good point, this conversation bears a striking resemblance to a scene from the 2016 Michael Keaton film, The Founder.
Keaton plays Ray Kroc, the founder of the McDonald’s franchise. In this particular scene, Kroc is traveling around to the various McDonald’s locations and finding them dirty, unkempt, or worse — offering food options not included under the original model.
Kroc confronts the franchise owner and lambasts him for offering additional items, including fried chicken and corn on the cob. The franchisee didn’t see the big deal. People loved the fried chicken, and families have appreciated the additional options. The franchisee boasted about these options, stating proudly, “We want to have something for everybody!”
As history would conclude, Kroc was correct. He saw the forest for the trees.
Abductive logic would have helped prevent a situation where our client offered a dozen conversion options, as well as the McDonalds’ franchisee from being scolded by Kroc.
The results from a small sample size (a few customers, for example), should not always guide your global business decisions.
Abductive Reason seeks to understand why. It is pragmatic, and therefore OK with best guesses; it is not positivist, like deductive reasoning, which exclusively focuses on proven truths.
It starts with an observation or set of observations and then seeks to find the simplest and most likely conclusion from the observations. This process, unlike deductive reasoning, yields a plausible conclusion but does not positively verify it.
If this sounds familiar, it should: Abductive Reasoning is the basis of AI and Deep Learning.
In a presentation titled “How Abductive Reasoning Promotes Robust and Reliable Research Practices,” Peter A. Bamberger describes how Hugh Laurie’s character in House uses Abductive Reasoning to pioneer his approach to differential diagnosis.
According to Bamberger, neither Deductive Reasoning and Inductive Reasoning can lead to new insights. Therefore we need a new approach to driving empirical exploration: one that uses data to give us first suggestions. But the suggestions do not have to prove anything to be useful.
First suggestions simply provide us with a few ideas from which we can apply further research.
Abductive Reasoning is a process of switching between Inductive Reasoning and Deductive Reasoning. It is the way we can see both forests and trees.
The process starts with Inductive Reasoning to develop mini-hypotheses. It then uses Deductive Reasoning to validate the mini-hypotheses and create new knowledge statements: all while seeking to understand the why behind a particular phenomenon.
Let’s use a simple example of a conversion rate experiment. There are many methods that marketers can employ to improve conversion rate; some tactics include Deductive Reasoning, others include Inductive Reasoning.
For example, we could conduct several customer surveys and develop a buyer persona, which is then used to inform the landing page changes. This is Inductive Reasoning.
We might find the following commonalities between a sample of customers:
The buyer persona is one of the oldest and most popular marketing tactics. We could add elements to the page that would cater to this persona’s specific interests, hoping to improve our conversion rate.
While this might help our conversion rate, we should recognize that we are attempting to squeeze our entire audience into one generalized bucket. Doing so will likely alienate many individual customers that do not fit this mold.
Instead, we might introduce an entirely new landing page and run an A/B split test experiment as a second strategy. This is a top-down, Deductive Reasoning approach. Eventually, we might learn that Variant B converts at a higher rate than Variant A, so we declare that the winner.
This approach’s flaw is that we are not introducing any new knowledge to our thought process (other than declaring a winner). We do not understand why our customers prefer one page to the next, and we continue to alienate possible customers that may have preferred the original Variant.
The solution is to use a blend of both schools of thought, methodically switching back and forth to develop new hypotheses and validate new knowledge statements.
Ian Chalmers and Iffat Jokhio of Pivot Design Group use the following graphic to describe how Abductive Reasoning sits at the intersection of the two classical approaches. The following illustration depicts how it applies to the UX process.
In our previous examples, we could start with an inductive approach via buyer persona analysis and then use our new hypotheses to test a new landing page variant, which can be validated or disproven through eye-tracking software. And so on and so on…
In the fall of 2019, we began working with a retailer that sold custom themed glassware. The products are unique and novel, the sort of thing that few people would buy for themselves, but are very popular as gifts.
We conducted a handful of interviews with previous customers — an inductive approach to learning more about their audience — and came away with a few mini-hypotheses about what could help increase overall site revenue.
Our primary hypothesis was that we should introduce new products that included more humorous slogans and phrases. This would be a new twist on the established brand that would excite some customers and influence repeat purchases. However, this also could alienate some of our existing customers that wouldn’t appreciate this new direction.
We were afraid that our small sample size would lead us to a false deductive knowledge statement. This is one of the most common mistakes in marketing strategy.
If the new product line would cause friction with a significant percentage of our existing customer base, we would conclude that we should not introduce them under the current brand umbrella.
This is a deductive conclusion, a new knowledge statement; these new products would damage our brand image.
We had many unanswered questions and several ways in which it could have gone wrong. For example:
In both scenarios, we’re not adding any additional information that could help develop new knowledge.
Instead, we leaned on Abductive Reasoning to help find a solution. A series of hypothetical questions followed…
…What if we created an entirely new brand?
…What are the costs, risks, and time commitments associated with creating a new brand?
…What would a new brand look like and feel like? What would we call it? Can we decide on a name that has an available domain name that we can lock down immediately?
…How would we bring these new products to market? Is there a way for us to figure out which existing customers would be interested in these products and exclusively target them with ads?
…What additional research do we need to conduct to determine if this is the right course of action?
…Is the client going to hate this idea? How should we present it to them?
You’d be amazed at how energizing this exercise could be.
Abductive reasoning helps answer these questions.
After a long process, we determined that a new brand was the best course of action. The client loved the idea, and within a matter of weeks, the new site was live.
This was one of the riskiest campaigns we’d ever launch. We dedicated an initial budget of 15K to advertise the new product line, and our team watched the real-time analytics with bated breath.
The new brand was a huge hit. We generated 120K in incremental revenue in the first month, and the site passed the 1MM mark within six months.
If not for Abductive Reasoning, the idea for the new product line would have been a disaster, or it would have never seen the light of day.
The act of switching between differing logical thought processes allows us to discover the most creative solutions to complex problems.
We can see both forest and trees simultaneously; it is a skill, and it can be developed. Start by asking hypothetical questions, then exhaustively attempt to prove or disprove your hypothesis, and then start all over again. You’ll get there soon enough.
We'll get back to you within a day to schedule a quick strategy call. We can also communicate over email if that's easier for you.