Does Expertise Really Matter Any More?

On mastery, AI, and what it actually means to build for "pro"

In 2008, Malcolm Gladwell reframed the American dream in his book Outliers through the 10,000 hour rule: If you work hard and long enough you can achieve mastery in any skill or field. The implication was clear: success is within reach for each of us and is ours to claim. All that's necessary is to put in the time and work, and have access to the right tools.

Furthermore, his premise carried with it several unchallenged, widely accepted assertions:

1. 10,000 hours is a really long time, and is therefore hard for one person to accomplish without help
2. Since 10,000 hours effectively equals a decade, there simply aren't enough hours or years for a person to master too many things
3. There are few shortcuts that can be taken to achieve true mastery other than practice, repetition, and a desire to improve ones craft in the chosen domain. Settling for just 'good enough' was for amateurs, not experts.

"Practice isn't the thing you do once you're good, it's the thing that makes you good" — Malcolm Gladwell

What followed Gladwell's popular axiom in Outliers was a decade and a half of access to technology that put the tools to learn, practice, and hone virtually any craft into the hands of millions.

It forced us to re-evaluate fundamental assumptions around the value of artistic work, ownership, intellectual property and compensation for that work. I was there, building Mavaru, an online music store in the post-Napster pre-Spotify era for independent artists in music to capture revenue on a pay-what-you-want model. I saw the issues of fair compensation for artistic work play out with consumers and fans of indie artists before it spread across the cloud. I watched as social media giants established walled gardens and centralized value chains, treating users as callously as the record industry had treated artists, launching products and profiting off of user data without consent, challenging the norms of privacy and ownership across many an industry and product category.

But with giving up privacy came universal access, and the advent of social media platforms like Myspace (RIP), Facebook, Instagram, Youtube and a panoply of mobile & web applications meant not only could anyone learn and improve at whatever area they wished to hone, but we could all participate in the journey and feel we had a stake in each others success.

Suddenly, we saw ourselves through the success of others with unprecedented voyeurism and scale, at times to the detriment of our collective mental health. But it filled us with dreams of our own. What felt undeniable was that the collective promise of the creator economy (and its hustle culture) were built on the unassailable belief that mastery was relevant, worthy, and possible.

But now it feels like that's all changing.

I am of course, talking about AI.

What happens to the value of 10,000 hours now? You've probably asked yourself, as we all have, will AI replace me?

But the real question is, if everyone has access to this technology — who will use it best, and why?

In a world where knowledge and skills can be compacted, scaled and shared, is mastery still a relevant, worthy and possible pursuit?

Absolutely. Because without it, you can't judge or value output. And you can't bend AI-driven tools to do what they do best: make you better.

As I've spoken to artists, leaders, and builders on the frontier of AI, in particular in the music and art world through our work at NAMM, the preeminent conference for pro audio gear — I am convinced that we already know the answer to the fundamental questions that drive our hopes & fears about AI and its impact. For example, when we held a private roundtable of artists and spoke to operators at NAMM, we asked them about the future of music creation in a world where AI tools could create content with the press of a button. What emerged of course, were strong opinions — those for and against such tools. But what both camps agreed on was a fundamental insight: none of them planned to stop making the work they loved. Real artists don't get into the game just to 'press the button' and they wouldn't be stopping any time soon. They do it for the love: the craft was the point, and the opportunity to create was value enough regardless of consumption. There had always been a mass market for what we now knew as 'slop' and as long as there were artists who loved their craft, they would create great work that was competitive, built by humans. And they'd always judge crap work (AI or not) with a ruthless bullshit detector. Real pros don't want slop.

And so, as I have explored and investigated these questions from music to B2B to the entire AI landscape, I have continued to encounter one word over and over again.

That word is "Pro".

Here's why I think that matters, and you should too.

Experts define what deserves to be "pro"

The word "pro" is doing enormous work right now. In pro audio it's the difference between a Neumann and a USB microphone. In B2B SaaS it's the tier that justifies the price jump. In AI tools, virtually every product reaching for credibility and enterprise adoption uses it. And yet nobody has defined what it actually means to earn it: commercially, creatively, or culturally.

That ambiguity is a problem. And it's also an opportunity.

Because what we're watching play out in music, in enterprise software, and in every category facing AI disruption is a market sorting itself into tiers in real time. The brands and professionals who understand which tier they're in, and why they belong there, will hold their position. The ones who don't will get flattened by the ones who do and wonder why customers do not believe their prices are worth it anymore.

The proof lies in a few tenets that I would argue today, as was the case 15 years ago, remain unchallenged yet currently misunderstood:

1. We do not inherently value tool-driven work, because its value proposition is often rooted in the expectation that it will reduce cost and effort required, not 'merely' increase quality. Ironically, tool use is seen as a defining human characteristic a la Homo faber 'Man the Maker.'
2. Quality of work matters. We don't want to pay (as much) for work that feels generic or which we do not perceive to be distinctive. But, just as there has always been, there is a two-ish tiered market for products or services that fit into a 'mass' lower tier, and a market for that which is 'best practice' 'enterprise-grade' or in shorthand — 'pro.' We are still willing to pay for better, we just have to believe its worth it. Positioning still matters.
3. Discernment. Knowing what good looks like before you make it is the only thing that gets you through points 1 and 2 correctly. And discernment cannot universally be downloaded in a markdown file, without understanding and crafting it to begin with. Probabilistic tools revert to the mean.

Consider what this looks like in practice:

A brand puts out AI-generated campaign assets at scale. The mass consumer doesn't notice, doesn't care, moves on. The pro consumer, the photographer, creative director, or engineer who's spent a decade developing their craft, notices immediately. Not the tool itself. This is a daily struggle for anyone who uses these tools and knows what 'pro' work looks like, and doesn't always get it on the first press of the 'button'.

In a B2B context, Peter Block outlines this truth through what he calls the 'Pair-of-Hands' concept in Chapter 2 of his book Flawless Consulting. He discusses the three reasons why companies look to outside consultants, what role the outside consultant should play, and how it affects the fees they can command. This is tremendously relevant in the agentic AI era, where the expectation we hear from executives every day is "can't we just do this with AI?"

Know your role. Then frame your value.

Block describes three modes a consultant operates in: the pair of hands, the collaborator, and the expert. Each commands a different level of leverage, and a different fee.

The pair of hands consultant takes orders. The client defines the problem, prescribes the solution, and the consultant executes. Low resistance, low leverage, low price; highly executional, and looked to when internal resources at the enterprise are low. The collaborator works alongside the client with shared ownership of the problem and the outcome. It's a blend of strategic and operational partner, but the consultant is primarily there to iterate, feedback, and offer fresh perspective down to the granular level. Finally, the expert consultant brings irreplaceable judgment to the table, honed from decades of experience, edge cases, and specialized work. The client comes with a problem, not a prescription. The expert diagnoses, recommends, and is trusted to be right.

AI neutralizes the pair of hands almost entirely. It executes faster, cheaper, and without complaint. The collaborator role gets compressed and potentially enhanced: AI fuses into the feedback loop, handling the assembly work that used to justify the hours, freeing up time for higher order collaboration and focus. But the expert role? That gets more valuable. Because the expert isn't being paid for execution. They're being paid for the judgment that tells you whether the output is any good, and AI cannot evaluate its own output against a standard it doesn't have. In fact, it gets riskier and worse unless you limit its scope and focus to defined tasks.

Therefore, the pro who can bend AI to their will, directing it, shaping it, and knowing when it's wrong delivers the highest quality output at scale. The one who can't is just a more expensive pair of hands. And the market is figuring that out fast.

What it actually means to be Pro

This is why I believe we are entering an era where mastery gets redefined, not replaced. That definition is happening around what it really means to be "Pro."

The romanticized version of the solitary craftsperson logging ten thousand hours in isolation gives way to something more practical and more demanding. Mastery now means knowing how to direct powerful tools toward outcomes that only someone with deep domain knowledge could even recognize as good.

The brands and professionals who will win in this environment share four moats we keep seeing across every category:

Taste. The ability to know what good looks like and to hold the line on it. Taste is one of last moats that can't be replicated at scale.

Data-Driven Insight. Data as the thing that tells you who you're actually building for, what they actually value, and where the gap between what you think is true and what is true actually lives. Taste without data is just an instinct. Data without insight is just noise. But together, they are the ultimate combination. The hard part is knowing when to follow the data to its uncomfortable conclusion, or to buck it. Which leads to…

Context. Every situation is different. And specialization is what allows a human to make the right call in a specific situation when a generalist, human or machine, would revert to the mean. This is Gladwell's premise holding under pressure: you can't master everything, which means the depth you've built in one domain is genuinely irreplaceable. An AI trained on everything knows a lot about the average but it knows very little about your specific moment. But a real pro — knows how to shape an AI to interrogate context and continually evaluate it. The expert who's spent a decade in one category carries context to meta-evaluate the AI's performance and improve it. Any AI that claims otherwise is hallucinating with confidence, which is precisely why what comes next matters so much.

Trust. This is where the myth of infinite possibility collapses. Yes, AI can help you do more. But can you really use it to do everything better, across every domain, at professional quality? The context switching alone is brutal. And often, AI tools never tell you when they're wrong. The amateur faking expertise won't catch it but the pro will. And that's why they are trustworthy. Trust is what you earn when the people around you know that what you produce is genuinely yours, shaped by real judgment, not just generated and forwarded. And that starts with a willingness to be clear what you won't do. It's also the first thing that goes when you ship slop.

Robert C. Martin, whose book Clean Code became a bible for developers who take their craft seriously, put it plainly:

"Writing clean code is what you must do in order to call yourself a professional. There is no reasonable excuse for doing anything less than your best." — Robert C. Martin, Clean Code

The new pro isn't the person with the most tools. It's the person who knows exactly what the tools are for, how to use them, and what they're not for.

What This Means

I don't believe the 10,000 hour rule has stopped being true. It's gotten more urgent. What Gladwell was really describing, underneath all the talk about practice and repetition, was the irreplaceable value of depth and focus. The judgment you've built that can't be shortcut, scaled, or prompted away. It still matters, it just shows up now in a different technological and historical context with new tools.

I still believe mastery is a relevant, worthy, and possible human pursuit. And I think we need to have a conversation about that because AI is forcing a reckoning that is now impossible to avoid: if you don't know who you are building for and what you offer that no one else does, you are at risk of being disintermediated. There is less room than ever for undifferentiated, generic work and your reputation and ability to command price all depend on it. Don't position yourself as a pair of hands.

Brands and leaders who can answer the hard questions with specificity, data-driven insight, and genuine depth of expertise have never been more valuable.

The Pro Perspective is where I, and others across the MOST Wanted Co family, will think out loud about these questions. And we will share research, strategy, and thinking as this plays out in real time.

If you're building for pros, developing a "Pro" offering, or trying to figure out what that actually means in this moment for you and your area of expertise: welcome. Let's figure this out.