Summary

Amy Edmondson’s Right Kind of Wrong gives CMOs a practical way to separate preventable mistakes from intelligent failures. For marketing leaders under pressure to innovate, especially around AI, the lesson is not to “fail fast” for the sake of it. It is to design smaller, smarter bets that help teams learn without getting reckless.

Why Failure Needs Better Definitions

Marketing leaders are under pressure to innovate without wasting money, move faster without creating chaos, and adopt AI without turning the team into a science experiment. That tension makes failure a loaded word.

In a CMO Huddles Expert Huddle, Harvard Business School professor Amy Edmondson reframed the issue for B2B marketing leaders as laid out in her book Right Kind of Wrong. The goal is not to celebrate failure for its own sake. The goal is to understand which failures are avoidable, which are complex, and which are intelligent enough to produce useful learning.

That distinction matters because marketing teams often treat every miss the same way. A broken email merge field, a campaign that underperforms because the buyer insight was wrong, and a disciplined AI pilot that disproves a hypothesis are not the same kind of failure. If the CMO treats them the same, the team learns the wrong lesson.

Basic, Complex, and Intelligent Failures

One of the most useful parts of Edmondson’s framework is the distinction between basic, complex, and intelligent failure.

A basic failure happens in familiar territory and is usually preventable. In marketing, this might look like a bad list upload, a campaign going live with the wrong personalization field, a webinar follow-up that never fires, or a sales enablement asset built from outdated messaging. These are not failures to romanticize. They point to process gaps, quality-control issues, unclear ownership, or insufficient training.

A complex failure happens when multiple factors combine. For CMOs, this can show up in live events, product launches, rebrands, sales handoffs, new market entries, and large integrated campaigns. No single decision may explain the outcome. The real work is to understand the interaction: Timing, audience, product readiness, channel mix, internal alignment, customer expectations, and execution.

An intelligent failure is different. It happens in new territory, in pursuit of a clear goal, with a reasonable hypothesis, and with a bet sized small enough to learn without causing disproportionate harm.

“An intelligent failure is an undesired result in new territory,” Edmondson explained.

For CMOs, that is the useful kind of failure: Not a badge of honor or a mess to excuse, but evidence from a thoughtful test.

Why Playing It Safe Can Become the Bigger Risk

One of the strongest CMO takeaways from the Expert Huddle is that avoiding failure today can create a larger failure later.

Marketing changes too quickly for teams to rely only on familiar playbooks. Buyer behavior shifts. AI changes workflows. Search becomes less predictable. Channels mature. Sales motions evolve. Competitive narratives move. A team that never ventures into new territory may look safe in the short term, but it becomes more vulnerable over time.

This is especially relevant for CMOs operating under intense CEO, CFO, or board pressure. The instinct is often to lower risk, protect targets, and avoid anything that could be perceived as a miss. But if the market is changing, playing not to lose can quietly become a strategy for irrelevance.

The answer is not reckless experimentation. It’s disciplined experimentation.

How to Make Experiments Smart, Small, and Useful

In the Expert Huddle, Edmondson emphasized that an intelligent failure must be tied to a goal and kept no bigger than necessary. That gives CMOs a practical operating principle.

Before approving a new campaign test, AI workflow, positioning experiment, event format, channel bet, or creative concept, the team should be able to answer:

• What are we trying to learn?
• What do we believe will happen?
• Why do we believe this is worth testing?
• What would prove us wrong?
• How small can we make the test and still learn something useful?
• What will we do with the learning afterward?

This turns experimentation into a management discipline. It also protects the team from vague innovation theater, where people say they are experimenting but have not defined the hypothesis, the risk, or the decision the learning will inform.

For example, a CMO might pilot a new AI-assisted content workflow with one campaign team before scaling it across the organization. If the workflow saves time but weakens insight quality, that is useful information. The test did not “fail” in a simplistic sense. It revealed where human judgment still needs to be built into the system.

Psychological Safety Is Risk Management

The Expert Huddle also clarified a frequent misunderstanding: Psychological safety is not about comfort, niceness, or avoiding standards. It is about making candor possible.

Marketing failures often become more expensive when people see warning signs but do not speak up. A campaign concept feels off. Sales is quietly ignoring an asset. A customer proof point is weaker than claimed. An AI-generated output sounds plausible but is wrong. A product launch timeline is too compressed. If the team is afraid to raise those issues early, the business risk increases.

“High quality bets depend on high quality conversations,” Edmondson said.

CMOs can build that environment through the questions they ask. 

Instead of asking, “Are we on track?” they can ask:

• What are we seeing that doesn’t fit our assumption?
• Where might this break?
• What would make this bet smarter or smaller?
• What are we afraid to say out loud?
• What did we learn that we should reuse?

What CMOs Should Do Differently

The practical shift is to build a marketing operating system that distinguishes learning from carelessness.

• Add “what we learned” to campaign readouts.
• Separate preventable execution issues from intelligent market-learning moments.
• Require a hypothesis for meaningful experiments.
• Keep experimental bets small enough to protect trust and budget.
• Ask teams to identify early warning signs before launch.
• Make it normal to discuss weak signals before they become public failures.
• Teach teams that curiosity is often more useful than being right.

The best CMO teams are not reckless. They are learning systems. They prevent what should be prevented, diagnose what is complex, and use intelligent failure to keep the company from standing still.

Q&A

What is intelligent failure in marketing?

Intelligent failure is an undesired outcome from a thoughtful experiment in new territory. For marketing teams, it should have a clear goal, a hypothesis, and limited downside so the CMO can turn the result into useful learning.

How is intelligent failure different from a mistake?

A mistake usually happens when a team deviates from known process or good practice. Intelligent failure happens when the team is exploring something new and learns from a well-designed test.

Why should CMOs care about psychological safety?

Psychological safety helps teams surface risks, weak signals, dissent, and early learning. For CMOs, that improves decision quality and reduces the cost of preventable failures.

How can CMOs make AI experimentation safer?

CMOs can keep AI tests small, define the learning goal upfront, clarify where human judgment is required, and review outcomes before scaling workflows across the marketing organization.

What should marketing teams do after an experiment fails?

They should identify what kind of failure occurred, capture what was learned, decide whether to stop, revise, or scale the idea, and reuse the insight in future campaigns or workflows.

‍

Want to hear more? Listen to the full conversation on Renegade Marketers Unite.

CMO Huddles helps B2B marketing leaders win by bringing together peers, fresh perspectives, and opportunities to build stronger personal brands. Want to join the huddle? Learn more about CMO Huddles and apply to join the community.