Oxford EMBA lessons from our eight core modules
The core modules of my Oxford Executive MBA are done. My notebook is now a strange mix of discounted cash flow formulas, a conditional-probability example about pregnancy, and a running list of college dinners. We now have only The Oxford Leader, Global Complexity 🛫, and electives ahead of us.
I started with the J26 cohort in January 2026 and will hopefully graduate late 2027. Before the details fade, I wanted to write down the Oxford EMBA lessons I keep coming back to, both in class and in my fractional-CMO engagements.
I've grouped them by what they're useful for. Partly because that's how I use them, and partly because the time value of money came up in three different modules and I only want to explain it once.
Decisions: your gut votes first
Most decisions start as emotion and gut instinct, then get rationalized. The practical fix is to slow down how you frame the decision.
Your career is a bias too. I've spent my working life in fast-moving eCommerce and logistics companies, so I have a penchant to move fast and break things. My classmates in healthcare, finance, and infrastructure reminded me that in their worlds, the cost of breaking things is too high. Knowing your defaults gives you a beat to pause before you act on them.
Beware the prosecutor's fallacy. If you're a woman, your probability of being pregnant is maybe 3%. If you're pregnant, your probability of being a woman is 100%. Same two facts, opposite direction. Mixing up P(A|B) and P(B|A) is how a tiny chance of the evidence, given innocence, gets heard as a tiny chance of innocence.
A few small rules worth keeping:
- Keep a variable in your regression if its t-statistic is above roughly 1.96 (significant at the 95% level).
- When in doubt, build a decision tree.
- Ask "why" five times to get to the real reason. The first answer is usually a symptom.
Money: a dollar today beats a dollar tomorrow
Money in the future is worth less than money today. Three reasons: risk (you might not get it), inflation (it buys less), and opportunity cost (you could have invested it elsewhere). So future returns get discounted back into today's money before you compare anything.
Size a project with net present value. Discount the project's future cash flows at your required return and subtract what it costs. Positive NPV, do it. Negative, don't. Because the required return is already baked into the discount rate, a positive NPV means the project beats the return you needed from it.
The part that stuck with me as a marketer: only count marginal cash flows, the ones that happen because of the project. That's the same question as incrementality in marketing. Would this revenue have shown up anyway?
Two kinds of risk. Systematic risk affects everyone. The central bank raises interest rates, and every company feels it. Idiosyncratic risk is firm-specific, like not getting the patent your product depends on. You can diversify away idiosyncratic risk. You can't diversify away systematic risk, which is why investors only get paid extra for bearing that one.
Weighted average cost of capital (WACC) is the return a company is expected to pay everyone who funds it, its lenders and its shareholders, weighted by how much each one put in. Higher WACC means riskier, so a project needs higher returns to be worth it. The standard formula:
WACC = (E / V × cost of equity) + (D / V × cost of debt × (1 − corporate tax rate))
E is the market value of equity, D is the market value of debt, and V is the two added together. Debt gets the tax adjustment because interest is usually tax-deductible.
The Rule of 72. Divide 72 by an annual growth rate to get the years it takes to double. At 10% that's about seven years; at 3% it's 24.
Accounting basics worth knowing:
- Know your profitability, liquidity, and leverage ratios. Can you make money, can you pay your bills this year, and how much of the business is funded by debt?
- Revenue isn't recognized until it's earned. Getting paid upfront for a year of service doesn't mean you've earned a year of revenue.
- The development part of R&D can be capitalized into an asset once it meets the viability criteria: it's technically feasible, you intend to finish and use or sell it, you have the resources to do so, it'll likely generate economic benefits, and you can reliably measure what it cost. Research is always expensed. (That's under IFRS, which is what we were taught; US GAAP is stricter.)
Follow the best ideas. One finding from business finance: a fund manager's biggest, highest-conviction holdings tend to outperform the rest of their portfolio. The signal is in the bets they're most willing to concentrate on. (From class. I'm a CMO, so please don't treat this as investment advice.)
Operations: standardize what repeats
Of all my Oxford EMBA lessons, this is the one I reach for most in marketing ops.
Low variety and high volume = good candidate for automation. High variety and low volume = one-off project. Your weekly reporting and lead routing belong in the first bucket. A rebrand belongs in the second.
Six Sigma means cutting your process variation until the average output sits six standard deviations away from the line between acceptable and defective. That buffer is what keeps quality tight even when the process drifts. In practice it works out to about 3.4 defects per million (assuming some drift over time).
Strategy only works with a culture that can execute it. The module gave me frameworks for diagnosing when culture is the actual constraint.
Customers and pricing
When you price something with no benchmark, aim high. If you're too high, the market tells you fast. If you're too low, you never find out. This lesson has come up many times in my own practice.
Find the pricing sweet spot with both stated and revealed preferences. What customers say they'd pay in a survey and what they actually pay when the checkout page is in front of them are different numbers. Use both, and trust behavior when they disagree.
The service gaps model gives you four places a service experience can break:
- Listening: how well do you understand what your customers want?
- Service design: how well have you structured your product or service to deliver it?
- Performance: how do you actually deliver, and how does the customer receive it?
- Communication: how well are you setting expectations and keeping customers in the loop?
The fifth, the customer gap, is the distance between what customers expected and what they think they got. It's what the other four add up to.
The service recovery paradox. Customers who had a service failure that was then fixed well can end up more satisfied than customers who never had a problem. I love this one, with a caveat: the research is mixed, and the lift shows up more reliably in satisfaction than in long-term loyalty. Fix failures fast. Don't stage them.
Where I disagree: attribution. In class, multi-touch attribution got a fairer hearing than I'd give it. In B2B eCommerce and tech, I've already landed on marketing mix modelling as the better approach. Buying committees and long sales cycles break most touch-based models. Consider this my one formal objection to the curriculum.
People and networks
Two network shapes compound: closed and diverse, or open and specialized. A tight group with different perspectives (a good cohort) or a broad web of experts in one domain. According to the module, everything in between tends to deliver less. Networks also come in layers that grow roughly threefold: about 5 close relationships, then 15, 50, 150, and 500. Each layer has a natural ceiling, so invest at the right one.
Do more things. The research we covered on entrepreneurs points to varied experience, failures included, as a common thread among the successful ones. More industries and more roles give you a bigger pattern library to draw from.
Wrapping up
If I had to keep one of my Oxford EMBA lessons, it would be marginal cash flows. Count only what happens because of the decision. It's the finance version of the question I ask about every marketing channel, and now I have the math to back it up.
The core modules gave me the language a CFO and board use. I keep my roster to four retainer clients at a time, more so now that I'm in school, but I'm always happy to talk shop, so reach out.
Filed Under: Management, General