Making decisions under uncertainty - how mathematics can solve real world problems in B2B SaaS
A guest piece written by Gökçe Ceylan, Principal at Oxx.
This piece is a guest piece written by Gökçe Ceylan, Principal at Oxx.
Gökçe joined Oxx in 2021 and is a Principal based in London. With her background in seed and growth stage investing and corporate venture, as well as on the other side of the table in an early-stage startup, she brings to the team a wealth of investment expertise, first-hand operator experience, and a deep understanding of B2B software. Gökçe is a SaaS enthusiast with a particular interest in the Future of Work.
For most people, mathematical theorems can feel very abstract and far from being applicable to real life. However, like the laws of physics, they are at play even when they go unrecognised. In my work as a venture capital investor, I see time after time how math theorems can help explain and address some strategic as well as operational topics in the world of B2B SaaS. Over the past year, I’ve delved into a number of these to provide tools for entrepreneurs as they encounter challenges on their company building journeys. Let’s look at a couple of these:
1.“Data versus gut feel” – how Bayesian statistics can make gut feel a measurable quantity
There is a misconception that gut feel-led and data-driven decision making are polar opposites. Maths, however, provides us with a theorem which combines the two and proves that you indeed can bring your gut feel into a data-driven decision-making process in a legitimate way.
In the world of VC and entrepreneurship, you are encouraged to look for Founder-Market-Fit, and years of experience in a respective field is considered a strong indicator of it. Through experience, these founders have developed a particular ‘gut instinct’ for how their field operates. But at the point of critical decision-making, you don’t want feel, you want evidence. So when does a founder’s years of accumulated experience come into play? Well, Bayesian statistics has the answer.
Bayesian statistics provides a mathematical framework for updating and revising beliefs based on new evidence. The theorem labels your current belief (market knowledge, experience, etc.) as ‘prior’, and new information (on sales, marketing campaign performance, etc.) is used to update your beliefs to achieve greater accuracy, which is then called ‘posterior’. Essentially, it allows you to continually improve your predictions or decisions based on new information.
This offers the best of both worlds: you start with intuition, which at every step becomes more accurate, based on new evidence. It’s a particularly strong framework as it allows you to not only account for your existing beliefs but also how confident you are about those beliefs; enables data-driven decisions based on probabilities rather than just statistical significance (so you don’t have to be a data scientist to leverage it); and works even with small datasets.
By showing the direction of travel rather than disparate data points, it enables decision-makers to avoid reacting to outliers or the highs and dips of rogue information.
Product and tech teams have been leveraging Bayesian statistics for various use cases; from A/B testing product feature releases, to anomaly detection in data, to building predictive models.
However, this methodology doesn’t need to be confined to the use cases of product and tech teams. As a business leader, you can also leverage Bayesian statistics to predict likelihood of your sales teams hitting their quota or test whether a launch in a new market is performing to expectation, and ensure your commercial decisions remain adaptive and data-driven.
2.Bringing clarity to a decision making process and overcoming decision fatigue by using decision trees
Research shows that 40-60% of B2B purchase processes fall through because buyers are unable to make a confident decision between the options presented to them. If they have too many choices, features, or setups thrown at them, it’s difficult to compare and decide exactly the best solution for them. However, if viewed from the right perspective, no decision should be overly complicated. We can borrow decision trees from mathematics to make sure that we are at the right “zoom level”, bringing the right perspective and therefore clarity to strategic and operational decisions. Maths to the rescue again!
Decision trees help in making decisions by breaking down a complex problem into simpler, step-by-step choices. They are the representation of a decision process, visually mapping out different possible outcomes based on various decisions, like a flowchart. Their purpose is to help identify the best possible decision by considering all potential consequences. Every decision you make takes you further down a particular branch of the tree, and you can logically track your decision-making – generating a high level of confidence in your final decision.
Let's put this into a B2B SaaS perspective, using sales as an example. B2B Account Executives aim to get buyers to purchase their product, such as an ESP (Email Service Provider). First, they must understand customers’ problems to be solved, (i.e. do they even need an ESP in the first place) then build trust through understanding their key purchasing criteria (KPCs), or even guiding them on what their KPCs should be, if unclear to the prospect. However, many sales executives skip to the ‘buy this product’ stage, bypassing essential checks the buyer needs, and start throwing ‘shiny features’ at them: here is how to create a GDPR compliant list, look at this long list of integrations, here’s a new AI feature that takes written commands from you, etc. With a feature-first sales approach, prospects get overwhelmed by choice without being guided through what matters for them in an ESP. Skipping steps in the decision tree can thus prove fatal for the Account Executive’s sales goals.
By applying the mathematical framework of a decision tree, prospects can streamline their decision-making and enable comparison based on a clear hierarchy of needs. Shifting focus from the buyer to the seller, such a methodical approach can also inform a company’s marketing or product strategy - i.e. if we lose them at a certain level or ‘node’ in the decision tree, it means that this criteria is important but it hasn’t been appropriately clarified. A client may not yet know it’s a key criteria to them, but their engagement in the process can inform hierarchy, and the marketing manager can use this information to complement their strategy by implementing for instance additional nurture programs or content pieces to build buyer confidence.
3.Win/Lose or Win/Win – Positive sum games and pricing negotiations
In game theory, there is a subset called the ‘zero sum game’. This is where, for example, a game might have a $5 prize to share between participants. If there are two people, for every extra dollar one person takes, the other person loses a dollar: the sum of gains and losses always equals zero. As SaaS invariably involves a subscription model, discussions around pricing and in particular, price rises, are a regular part of business, but should always be handled with care. Many businesses struggle to set their pricing, and when they choose to change their model or raise their prices, the communication around it becomes of utmost importance. This is about creating a positive sum game. Framing a price increase as a positive sum game during pricing negotiations can actually be a way to bolster customer retention.
Why is this? Well, when B2B providers position price increases as solely a raise in cost for the customer, it’s a zero sum game. Why should a customer lose for the provider to make more money? This becomes a straightforward game of give and take, with no value creation involved in the conversation. And the buyer in this case feels like they are being fleeced. However, in positive sum games, both participants can gain from taking a specific action. Companies should reframe the conversation around price increase as an opportunity to release new features, or discuss the gains already being made from the adoption of their software. By demonstrating to a customer, ‘We’ve raised the cost by $X to deliver these additional features/benefits, but at the same time you’ve already benefited by >$X from the product,’ you treat price increases as positive, and an opportunity for both parties to win.
As you can see, maths is ubiquitous. Recognising this, figuring out which theorem is at play, and following that framework will help you in your role – either as an entrepreneur, operator or investor.
Gökçe is an investor at Oxx. You can read more from her at oxx.vc and follow her series on Math meets B2B SaaS, where she regularly deep-dives into a new mathematical theorem and how it is applied to the B2B SaaS world.
