Stop Playing Chess With Your Pipeline: Sales Is a Game of Poker

I recently finished Annie Duke’s Thinking in Bets: Making Smarter Decisions When You Don’t Have All the Facts, and I absolutely loved it as a fractional VP of sales.

Not because I have any particular interest in becoming a poker player.

It was because Duke, a former professional poker player and World Series of Poker bracelet winner who has since focused her work on decision-making, gave me language for something I have experienced throughout my career in sales leadership:

“We spend far too much time pretending we know things we cannot possibly know.”

Sales forecasts are presented as commitments. Customers describe buying decisions as though the answer can be objectively calculated. RFPs and tenders attempt to reduce complicated business decisions to scores. Salespeople tell managers that deals are “90%.” Buyers tell salespeople that they have “done their homework.”

Then something happens that nobody saw coming. And everyone wonders how they missed it.

Maybe we didn't miss it. Maybe we were simply playing the wrong game.

B2B Sales Is Not Chess

One of the ideas behind Thinking in Bets is the distinction between decisions made with complete information and decisions made under uncertainty.

Chess is largely a game of complete information. Both players can see the board. You know where every piece is. You know the rules. You know what moves are available. You don't know what your opponent will do next, but the current state of the game is visible to both players.

Poker is fundamentally different.

You can see some of the cards. You cannot see others. You know something about the other players, but not everything. You can calculate probabilities, study behaviour, and make an intelligent decision, but information is always missing.

And then there is luck.

You can make a statistically sound decision and lose the hand. You can also make a terrible decision and win. That distinction sits at the heart of Duke's work on decision-making under uncertainty. A good outcome does not necessarily prove that we made a good decision, just as a bad outcome does not necessarily prove the decision was bad.

Tell me that doesn't sound like sales!

You Can Run a Great Sales Process and Still Lose (Sales Strategy)

Anyone who has sold for long enough has experienced it.

You've done the discovery. You've identified the problem. You've met the decision-makers. The customer likes the solution. The business case works. Your champion tells you you're the preferred vendor.

Everything appears to be going your way. And then you lose.

Maybe the CEO froze spending. Maybe your champion didn't have as much influence as you thought. Maybe procurement added another supplier at the last minute. Maybe a competitor dropped its price. Maybe the customer was acquired. Maybe priorities changed. Maybe someone inside the organization simply wanted another supplier to win.

There were cards you couldn't see. That's poker.

The opposite happens too. A salesperson skips discovery, sends a mediocre proposal, discounts unnecessarily, and somehow receives the purchase order. They won. But did they make good decisions? Not necessarily.

One of the dangers Duke identifies is judging decision quality entirely by its eventual outcome (a heuristic she calls "resulting"). If winning automatically means we made good decisions, we can accidentally reinforce terrible behaviours.

This matters enormously in sales management. If one rep follows a rigorous process and loses because circumstances changed, while another ignores the process and gets lucky, which behaviour should we reinforce?

The result matters. But so does the quality of the decision that produced it.

"This Deal Is Definitely Closing"

Now consider the sales forecast. A salesperson walks into a pipeline meeting and says:

“This deal is definitely closing by Friday.”

There are few phrases in sales management that should make us more nervous. Definitely? Based on what?

This is where Duke's deceptively simple idea becomes useful: Wanna bet?

There is something psychologically interesting about attaching a wager to a belief. When nothing is at stake, certainty is cheap.

“This is absolutely happening.”

“I'm positive.”

“No question.”

But imagine saying: “Would you bet your commission on it?”

Suddenly:

“Well... legal hasn't actually completed the contract.”

“Procurement still needs to approve it.”

“We haven't spoken directly to the CFO.”

“They are still talking to another vendor.”

Interesting. Nothing about the opportunity changed in those five seconds. What changed was the willingness to interrogate the belief.

Duke discusses this idea of treating beliefs as bets because putting something at stake encourages us to scrutinize what we know, how we know it, and how certain we should really be. And I think there is an incredibly useful sales lesson hiding inside that idea.

Without Skin in the Game, Assumptions Are Cheap

Sales organizations are filled with assumptions.

“This customer loves us.”

“The budget is approved.”

“The existing supplier isn't performing.”

“The CFO will sign off.”

And customers make assumptions just as casually.

“Our existing supplier will continue taking care of us.”

“We can solve this internally.”

“We can wait another year.”

“Our tender process will identify the best supplier.”

Maybe every one of those statements is correct. But how hard have we actually tested them?

That is where the concept of skin in the game becomes useful. If there is no consequence to being wrong, there is very little pressure to examine what we believe. Put something behind the belief, and our thinking tends to change.

Money. Time. Growth. Reputation. Resources. A future commitment. Those things turn an opinion into something much closer to a bet. I often tell salespeople you will never know your full abilities until you are willing to remove all security like an entrepreneur.

The Customer Is Playing Poker Too

This is the part salespeople frequently miss. We understand that we have incomplete information. What we sometimes fail to challenge is the assumption that the customer's buying process has somehow eliminated that uncertainty.

It hasn't. Customers are playing poker too.

Think about a sophisticated tender or RFP. A company identifies requirements, develops scoring criteria, asks suppliers to answer questions, compares pricing, and checks references. The process is designed to create fairness, consistency, and better information.

Those are good things. But it can also create a dangerous illusion: We evaluated everything; therefore, we know which decision is best.

No. You evaluated everything your process allowed you to see. That's not the same thing.

A tender cannot tell you everything about the next five years. The supplier that looks financially strong today could experience problems two years from now. The brilliant technical person who impressed your committee could resign next month. The market can change. Technology can change. Your own requirements can change.

Here's the line I keep coming back to: A tender does not eliminate uncertainty. It organizes the uncertainty you know about.

There are still hidden cards.

This Is Where Great Salespeople Challenge the Customer

This connects directly to The Challenger Sale.

The Challenger approach doesn't say great salespeople simply answer questions politely and hope the customer reaches the correct conclusion. It emphasizes teaching customers something new, challenging assumptions, and creating constructive tension.

That doesn't mean being obnoxious. It means being willing to say: “I think there is a risk in the way you're looking at this.”

Suppose a customer tells you: “Our tender process is comprehensive. We're confident it will identify the best supplier.”

A weak salesperson says: “Great. We'll make sure we complete every section.”

A salesperson who is prepared to challenge the customer's thinking might say: “Your tender will give you a very structured way to compare what you know today. Can I challenge one assumption, though? Some of the things that will determine whether this decision succeeds three years from now can't possibly appear on today's scorecard. What are you betting on being true about whoever you select?”

Now the buyer might recognize that they are betting the supplier remains financially healthy, key talent stays, and service levels remain strong. None of those observations prove that the tender process is bad. They simply reveal that the process cannot create certainty where certainty does not exist.

Challenge the Assumption. Then Find the Stake.

If you really want someone to examine a belief, make the consequences of that belief visible.

Consider: “We can wait another year.”

You could ask, “Why?” (That's fine.) You could argue, “I think you should act now.” (That's probably not very effective.) Or you could challenge the assumption:

“What are we assuming remains true during that year?”

Then, find the stake: “If we're wrong and this problem gets worse, what does that year cost the business?”

Maybe it's $50,000. Maybe it's lost production or customer churn. Never manufacture the stakes. Discover them. You are helping the buyer identify the consequence embedded in a decision they are already considering.

Once you discover that consequence, you can ask a much more powerful question: “Are we comfortable making that bet?”

Now there is skin in the game.

Doing Nothing Is Also a Bet

Customers frequently treat change as risky and the status quo as safe.

Buy a new system? Risk. Change suppliers? Risk. Do nothing? Somehow, that gets treated as the absence of risk.

But doing nothing is a decision. And therefore, it is a bet.

Keeping the existing supplier means betting their performance will remain acceptable. Waiting a year means betting the problem will not become materially worse. Delaying an equipment replacement means betting the equipment will continue operating reliably.

Those may all be reasonable bets. The point isn't that change is always better; the point is that the status quo deserves the exact same interrogation as the alternative.

Get the Customer to Put Something Behind the Statement

There is another way sellers can create skin in the game without becoming confrontational: Ask for a corresponding action.

Customers will sometimes say: “If you could demonstrate X, we'd seriously consider moving forward.”

Okay. Now put something behind that statement.

“If we can demonstrate that, what would you be comfortable agreeing to as the next step?”

Maybe it's bringing in the CFO. Maybe it's a technical review or conducting a pilot. The point isn't to trap the customer. The point is to determine whether the belief has substance. If the seller has to keep proving things while the buyer is never willing to put anything behind their statements, you've just uncovered a hidden card.

Stop Asking, "Are We Going to Win?"

Sales managers should apply exactly the same thinking internally. Most pipeline reviews revolve around some version of: “Are we going to get the deal?”

That invites a prediction. I'd rather ask: “What are we betting on being true?”

We have the budget. (Do we know that, or believe it?)

The CFO supports the project. (Have we heard that directly?)

We're the preferred supplier. (Preferred by whom?)

The existing supplier is out. (Are they actually out?)

Those questions force us to separate facts from assumptions. Then ask: “What haven't we learned yet?” That is a much better sales conversation than asking a rep to defend an arbitrary probability in the CRM.

Probabilities Are Useful, But Don't Confuse Them With Facts

There is still value in probabilistic forecasting. If you have a $100,000 opportunity and reasonably estimate a 60% chance of winning it, its mathematical expected value is $60,000.

But that doesn't mean $60,000 is coming. You're either likely to receive something close to the full deal value, or nothing at all. Expected value is useful across a portfolio of decisions. It is not a certainty about an individual one.

The useful managerial question isn't, “Is the CRM percentage right?” It's: “What new information would materially change our probability?”

Then go get that information.

Learn When to Fold

Poker gives us another lesson salespeople desperately need: You don't play every hand.

Yet salespeople have an extraordinary ability to keep dead opportunities alive, falling victim to the Sunk Cost Fallacy:

“We've already invested three months.”

“I've traveled there twice.”

“We're too far into this to walk away.”

But the time you already spent is gone. The relevant question is: Knowing everything we know today, would we make the same investment again?

Would you spend another ten hours on it? If you wouldn't place the next bet, why are you still playing the hand? Being good at sales is not only about knowing when to push. Sometimes it's about knowing when to fold.

Run a Pre-Mortem Before the Decision

Psychologist Gary Klein introduced the business pre-mortem as a way to surface risks before a project begins. Instead of asking, “What might go wrong?”, you assume failure has already happened.

Imagine it's twelve months from now. The project failed. The deal never closed. Now ask: “What happened?”

Think about using that with a customer: “Let's assume it's a year from now and this decision turned out badly. What happened?”

Suddenly you may hear:

“The implementation team wasn't capable.”

“We underestimated integration.”

“Our people didn't adopt it.”

Those are the hidden cards. You still cannot see all of them, but now you're actively looking for them.

The Objective Isn't to Remove Uncertainty

You can't. That's what many sales processes get wrong.

We create frameworks, methodologies, tenders, and forecasting processes because we want certainty. These tools can absolutely improve decisions, but they cannot make the future knowable.

The objective should not be certainty. It should be better decisions under uncertainty.

What do we think we know?

What are we assuming?

What information are we missing?

What happens if we're wrong?

And perhaps most importantly: What are we already betting on being true?

Sales Is Poker

Your forecast is a bet. Your qualification decision is a bet. Choosing which opportunity deserves another ten hours is a bet. Your customer's tender is a bet. Acting today is a bet. Waiting another year is a bet.

The sophisticated buyer and the sophisticated salesperson aren't the people who somehow know exactly what is going to happen. They're the ones who understand that they don't.

They search harder for the hidden cards. They challenge their own assumptions, and each other's. They understand the stakes. And then they make the best decision they can with the information available.

So perhaps we shouldn't actually look across the table at a prospect and say: “Wanna bet?”

But we should absolutely be thinking it. Because somewhere underneath almost every meaningful business decision is a wager. And the question worth asking is:

What are we betting on being true?

References

Duke, Annie. Thinking in Bets: Making Smarter Decisions When You Don't Have All the Facts. Portfolio/Penguin, 2018. Duke's work examines decision-making under uncertainty, probabilistic thinking, and the difficulty of separating decision quality from eventual outcomes.

Dixon, Matthew and Brent Adamson. The Challenger Sale: Taking Control of the Customer Conversation. Portfolio, 2011. The Challenger methodology emphasizes teaching customers something new, tailoring the conversation, taking control, and using constructive tension to challenge established assumptions.

Klein, Gary. “Performing a Project Premortem.” Harvard Business Review, September 2007. Klein describes the pre-mortem technique for surfacing concerns and potential causes of failure before a project begins.

He, Zichang and Wen Jiang. “Evidential Supplier Selection Based on Interval Data Fusion.” 2017. The research characterizes supplier selection as a multi-criteria decision problem containing unavoidable uncertain information.

Jiang, Dizuo et al. “A Possibility Distribution Based Multi-Criteria Decision Algorithm for Resilient Supplier Selection Problems.” 2018. The study examines supplier selection where decision makers must work with conflicting criteria and imprecise information.

Rezazadeh, Alireza. “A Generalized Flow for B2B Sales Predictive Modeling: An Azure Machine Learning Approach.” 2020. The work treats B2B opportunity outcomes probabilistically and examines using historical opportunity data to improve predictions.