Almost every skill you have is on a depreciation schedule right now. Not because you are getting worse at it, but because AI is getting better at it, and the gap between what you can do and what a model can do is closing on nearly everything. Synthesis. Drafting. Framing. Analysis. Each model release quietly marks another capability down.

But there is one skill that runs the other way. It does not depreciate when the model improves. It appreciates. And almost nobody is organizing their career around it, which is exactly why I want to name it.

Part 1: The Skill That Runs Backward

The skill is judgment under ownership. Specifically: the ability to decide, with accountability attached, which of the things AI can now produce is actually the right thing to do, and then to own the consequence of that decision.

Watch why this one inverts the usual math. When a model can generate ten viable strategies, five plausible roadmaps, and three defensible business cases in the time it takes to read this paragraph, the scarce thing is no longer producing options. It is choosing among them and standing behind the choice. The more fluent AI becomes at generating possibilities, the more valuable the person who can say “that one, not the other nine, and here is why, and I own what happens next” becomes. Abundance of options makes judgment scarcer, not cheaper.

This is the part that trips people up. They assume that when AI gets better at the analysis, the human who did the analysis becomes less necessary. The opposite is true for anyone whose real contribution was the decision, not the deck. AI flooding the zone with credible options raises the price of good judgment, because now someone has to be accountable for picking correctly from an overwhelming set, and the model cannot be accountable for anything.

Part 2: Why Judgment Can’t Be Generated

Here is the structural reason judgment appreciates while everything around it depreciates. A model can produce a recommendation. It cannot own an outcome. Those are different in kind, not degree.

When you make a call and attach your name to it, you are supplying something the model structurally cannot: consequence. You will be in the room when it works or does not. You carry the relationship with the stakeholders affected by it. You absorb the cost of being wrong and earn the standing of being right. That accountability is not a feature that a better model eventually acquires. It is the one thing that cannot be generated, because generation and ownership are opposites. The moment a decision has a real owner, a human is doing something no model does.

And this is why the skill compounds. Every time you make an owned call and it holds up, you build a track record that the next decision draws on. Judgment is one of the few things that gets more valuable the more you exercise it, precisely because it accrues to a person and a reputation rather than to a reproducible output. The model starts fresh every time. You do not.

So while your artifact-production skills are being marked down with each release, your judgment, if you are actually exercising it under ownership, is being marked up. The catch is that most BRMs are not positioned to exercise it. They are still positioned as producers, generating output the model now generates too, instead of as owners making the calls the model cannot.

Part 3: Reposition Around the Thing That Appreciates

If judgment under ownership is the asset that appreciates, the obvious move is to reorganize your role around it: to shift from producing artifacts toward owning outcomes, decisions, and realized value. But that shift is not automatic, and the maturity model most BRMs were trained on actively points the wrong way. It rewards advisory polish, the exact thing AI just made cheap.

That is the gap The Operator Shift was built to close. It is a field supplement to Earn Strategic Trust, and it lays out which capabilities appreciate as AI improves and which depreciate, a five-level Operator model built around ownership rather than advisory maturity, and a self-assessment to find where you actually sit today. If you want to spend the next few years building the skill that gets more valuable every time the model does, this is the map.

Get it here: The Operator Shift ($17.99, datasciencecio.com).

Think different for different results.

Most conversations about AI and the BRM role are set in the future tense. What will happen to the role. How it might change. What you should prepare for. That framing is comfortable because it keeps the whole thing at a safe distance, somewhere on the horizon.

I want to take the comfort away. The sort is not coming. It is already running, in live budget cycles and staffing decisions and quiet reorganizations happening right now. And it is separating BRMs into two piles. The hard part is that most people cannot tell which pile they are standing in.

Part 1: The Two Piles

Here is the line the sort runs along, and it is not seniority, tenure, or title.

In the first pile are the BRMs whose value came primarily from producing artifacts. The stakeholder map. The capability model. The maturity assessment. The strategy deck. The roadmap. The executive briefing. For years this was legitimate, valuable work, and a lot of the role’s credibility was built on being the person who could synthesize all of that and present it well.

In the second pile are the BRMs whose value comes from owning outcomes. Not the deck about the initiative, the initiative. Not the roadmap, the tradeoff decisions the roadmap represents. Not the assessment of value, the realized value itself, with their name on the number.

For a long time both piles looked the same from the outside, because both produced impressive output and both sat in the same meetings. AI is what pulled them apart. When a senior leader can generate a credible stakeholder map or capability model or strategy deck in minutes, the first pile’s core value does not shrink gradually. It drops to near zero almost at once. The second pile is untouched, because you cannot generate an owned outcome from a prompt.

Part 2: Why You Can’t Feel the Sort Happening

The unsettling part is that the sort is nearly invisible from the inside, and there is a specific reason for that.

If your value came from artifacts, AI did not announce its arrival by taking your job. It arrived by quietly making your best work cheap. The deck you would have spent two days on still gets made, so nothing feels different in your daily experience. What changed is offstage, in how the people who fund and staff the role now perceive that work. The effort that used to read as evidence of your value increasingly reads as something the org can get without you. You are producing the same output and getting a different amount of credit for it, and that gap is almost impossible to feel in the moment. You only notice it later, when the strategic work starts routing around you.

So the reassuring signals are exactly the misleading ones. Still busy. Still producing. Still in the meetings. None of that tells you which pile you are in. The only question that actually locates you is uncomfortable and simple: if the artifacts you produce could be generated for free tomorrow, what would you still be paid for? If you have a clear, concrete answer, you are in the second pile. If the honest answer is a pause, the sort has already begun placing you.

Part 3: You Can Still Choose Your Pile

Here is the part worth staying for. This sort is not a verdict. It is a window, and it is open right now. The BRMs moving from the first pile to the second are not doing it by working harder at artifacts. They are repositioning, deliberately, toward ownership, judgment, and realized outcomes, the parts of the role that get more valuable as AI gets better rather than less.

But repositioning requires knowing precisely where you stand today and what the target actually looks like, and the maturity model most BRMs were trained on does not describe this world at all. That is exactly the gap I built The Operator Shift to close. It is a field supplement to Earn Strategic Trust, and it maps the AI compression of the role, which responsibilities are most exposed, which capabilities appreciate, and a five-level Operator model with a self-assessment so you can locate yourself honestly and move before the role moves without you.

If you have read this far, you already suspect which pile you are in. This tells you what to do about it.

Get it here: The Operator Shift ($17.99, datasciencecio.com).

Think different for different results.

There is a moment near the end of every initiative that feels like the finish line. The build is done. The system is live. The demo went well. Someone updates the status to green, the project gets marked complete, and the team exhales. Delivered.

And then, six months later, nobody can tell you what it was actually worth.

I have watched that gap open up for twenty years, as an engineer, a CIO, and now leading data and analytics portfolios. It is one of the most expensive gaps in all of technology work, and it hides in plain sight because we use one word when we mean two. “Delivered” and “realized” sound like synonyms. They are not. And your stakeholders, the ones who fund the next thing, only ever cared about the second one.

Part 1: The Word Swap Nobody Notices

Here is the swap. “Delivered” means the thing exists. The pipeline runs, the dashboard loads, the integration passes its tests. It is a statement about the output. “Realized” means the value showed up. The cycle time actually dropped, the cost actually came down, the decision actually got faster, and someone can point to it in a number. It is a statement about the outcome.

We celebrate delivery because delivery is visible, dated, and ours. It happens on a timeline we control, and it feels like completion. Realization is none of those things. It happens later, off our timeline, in someone else’s part of the business, and it requires the change to actually take hold in how people work. So we mark the project green at delivery and quietly assume realization will follow on its own.

It does not follow on its own. That is the entire problem.

Part 2: The Handoff Where Value Goes to Die

Watch what happens at the exact moment of delivery. The build team’s involvement ends. The system gets handed to the business, the project closes, the team rolls to the next initiative. And in that handoff, ownership of the value silently disappears.

Think about who is responsible for realization the day after go-live. The build team considers the job done; they delivered. The business considers it IT’s system; they did not build it. So the follow-through that turns a live system into a measured outcome, the adoption push, the process change, the measurement, the “did this actually move the number” question, falls to nobody. It is not that someone drops it. It is that it was never anyone’s to hold.

This is why the demo can impress and the value can still never materialize. The two events are separated by a handoff with no owner, and value does not survive an ownerless handoff. It evaporates in the gap between “it works” and “it worked,” and by the time anyone notices, the team has moved on and the trail has gone cold.

And here is the compounding cost. When you cannot show realized value from the last initiative, you walk into the funding conversation for the next one with nothing but a promise. The whole cycle gets harder, because you spent your credibility delivering things instead of realizing them.

Part 3: Practice, Not Theory

Closing that gap is not a mindset problem and it is not a measurement trick. It is a practice: a set of concrete things you do before, during, and after delivery so that realization has an owner, a mechanism, and a number attached to it. Most technical leaders were never taught that practice, because our entire training points at delivery. We are extremely good at making the thing exist and largely untrained at making the value show up.

That practice is what I wrote Deliver Real Value to teach. It is the third book in the BRM Accelerator Series, and it is deliberately about execution over theory: how to carry an initiative past the finish line of “delivered” all the way to the finish line that actually matters, “realized.” If your work keeps shipping green and disappearing from the value conversation, this is the book that closes the gap.

Get it here: Deliver Real Value on Amazon ($24.95, paperback).

Think different for different results.

Ask most people how they earned a stakeholder’s trust and they will tell you about a moment. The crisis they handled. The impossible deadline they hit. The executive presentation that landed. There is always a story, and the story always has a hero beat in it.

I used to believe that too. Twenty years in, working across engineering, the CIO seat, and now leading data and analytics portfolios, I have come to think it is almost entirely wrong. The big moments do not build trust. They spend it. And what you get to spend in the crisis was quietly deposited long before, in a hundred moments so small you barely remember them.

Part 1: The Deposit Nobody Watched

Think about the last person you genuinely trusted at work. Not liked. Trusted, in the specific sense that you would hand them something important and stop worrying about it. Now try to name the single moment that trust came from. You cannot, and that is the point. It did not come from a moment. It accumulated.

It came from the email they said they would send by Thursday, and it arrived Wednesday. The number they promised to double-check, and they actually did, and they came back and told you it was fine before you had to ask. The meeting they could have let drift, and instead they closed the loop in two lines so you knew exactly where things stood.

None of those were heroic. None of them would make it into a performance review. And that is precisely why they built trust: because nobody was watching, which means the person did them for the work and not for the credit. Trust is not earned in front of an audience. It is earned in the follow-through that has no audience at all.

Part 2: Why the Follow-Up Is the One You Skip

Here is the mechanism that quietly erodes relationships. The follow-up is always the lowest-status task on your list. It is the two-line email confirming what you already discussed. It is the small loop that feels closed enough. It is the update nobody explicitly asked for. So under pressure, it is the first thing to go. You tell yourself they know it is handled. You tell yourself you will circle back. You almost skip it, and often you do.

But from the other side of the relationship, that skipped follow-up is not invisible. It is the exact spot where a small doubt takes root. Not a dramatic one. Just a quiet “I am not sure where that landed” that makes the stakeholder check in, or hedge, or keep a task on their own list that should have left it. Every skipped loop transfers a little anxiety back to them. And a stakeholder who has to carry anxiety about your reliability will never hand you the strategic work. They cannot afford to.

This is the part people miss when they chase the big moment. You do not get the crisis to be a hero in unless someone trusted you enough to put you near the crisis in the first place. And that trust was built in the follow-ups, long before the stakes were high.

Part 3: Mindset Over Mechanics

None of this is a technique you bolt on. You cannot fake consistent follow-through, because the whole signal is that you did it when nobody made you. It comes from a different place: a mindset about what the relationship is for and who you intend to be inside it. Get the mindset right and the small reliable behaviors follow on their own. Try to reverse-engineer the behaviors without the mindset and stakeholders feel the performance immediately.

That is the argument at the center of Earn Strategic Trust, the second book in the BRM Accelerator Series. It is not a book of relationship tactics. It is a book about the mindset that produces trust as a byproduct, so that when the big moment finally comes, you already have the standing to meet it. If your influence has ever plateaued despite doing good work, the gap is almost always here.

Get it here: Earn Strategic Trust on Amazon ($24.95, paperback).

Think different for different results.

Listen to how technical and relationship leaders describe their own work in funding conversations, and you will hear a specific phrase over and over. “The hard savings are around forty thousand, and then there are some soft benefits on top of that.” It sounds humble. It sounds honest. It sounds like someone being careful not to overclaim.

It is also the exact moment the value evaporates.

I have spent twenty years watching capable people undercut their own initiatives, and the “soft benefits” phrase is the most common self-inflicted wound I see. The person saying it thinks they are being credible. What they are actually doing is standing in front of the finance lead and handing them permission to round half the value down to nothing.

Part 1: What “Soft” Actually Signals

Here is what you think “soft benefit” means: a real benefit that is harder to measure precisely.

Here is what the person holding the budget hears: a benefit the presenter could not be bothered to quantify, and therefore one I am free to discount to zero.

Those are not the same message, but you only control the first one. The second one is what determines whether your number survives. When you label something “soft,” you are not adding a caveat. You are making a ranking decision on your own initiative, and you are ranking the benefit at zero. Finance simply accepts the ranking you volunteered.

The tragedy is that most “soft” benefits are not soft at all. Reduced rework, faster decision cycles, lower analyst burnout, fewer escalations, better data trust across a business unit. Every one of those has a defensible dollar figure sitting underneath it. It is not there because someone stopped digging one layer too early and reached for the word “soft” as a shortcut.

Part 2: The Anatomy of a Benefit That Got Away

Take “this saves the team a lot of manual effort.” That is the sentence that dies as a soft benefit. Now watch it get resurrected.

How many people? Say four analysts. How many hours per week does the manual work consume? Say six each. That is twenty-four hours a week, roughly twelve hundred hours a year. What is the loaded hourly cost of an analyst at your organization? Call it eighty dollars. You are now looking at just under one hundred thousand dollars in annual capacity, and not one number in that chain was invented. Every figure is one your organization already knows or can produce in an afternoon.

That is the difference between a soft benefit and a hard one. It is not the nature of the benefit. It is whether you did the arithmetic. “Soft” is almost never a property of the value. It is a description of how far you stopped short of quantifying it.

And here is the part that should sting a little: the analyst-hours example is the easy one. The genuinely valuable benefits, the strategic ones, are the ones people abandon fastest because the chain looks longer. So the most important value in your portfolio is precisely the value most likely to get waved away as “soft.”

Part 3: Stop Discounting Your Own Work

The fix is not to overclaim. Overclaiming gets you caught, and getting caught once costs you credibility for years. The fix is to build the traceable chain from the technical work to the dollar figure, so that when you say a number, you can defend every link of it, and you never have to reach for “soft” as a hedge.

That discipline, turning vague benefit language into quantified, defensible figures without inflating anything, is a learnable method. It is the core of what I wrote Quantify Your Impact to teach. The book exists because I watched too many strong BRMs and product owners talk themselves down to zero in rooms where their work deserved a real number. Every “soft benefit” you have ever mentioned had a figure underneath it. This book is how you find it.

Get it here: Quantify Your Impact on Amazon ($24.95, paperback).

Think different for different results.

There is a particular kind of quiet that happens in a budget review. Not the good quiet. The kind where you finish presenting the initiative you are proudest of, the one your team spent two quarters building, and the room just moves on. No hostility. No debate. Someone nods, says “let’s revisit next cycle,” and the next line item comes up. Your work was not rejected. It was skipped.

I have watched that happen to genuinely excellent work more times than I can count. And after twenty years of sitting on both sides of that table, as an engineer, a CIO, and now leading data and analytics portfolios, I have stopped believing the problem is the work.

Part 1: The Work Was Never on Trial

Here is the uncomfortable truth. In the room where funding gets decided, nobody is evaluating your architecture. Nobody is admiring your integration design or your data model or the elegance of how the pipeline handles edge cases. Those things are real, and they matter, but they are not what is being weighed.

What is being weighed is a list of numbers. Every initiative that survives that meeting arrives as a number: a cost avoided, a cycle time reduced, a revenue stream protected, a risk priced. The ones that arrive as descriptions, however impressive, get filed under “sounds valuable, can’t rank it.” And you cannot fund what you cannot rank.

This is the trap that catches strong technical leaders specifically. The better your work is, the more you assume it speaks for itself. It does not. In a budget review, work does not speak. Numbers speak. Your work sat there, mute, next to six other line items that had learned the language.

Part 2: What Actually Happened in That Room

Play the scene back and watch the mechanics. The finance lead is not being difficult. They are being forced to sort. They have a fixed pool of money and a stack of requests that exceeds it, and their entire job in that hour is to produce a ranked order. Anything that cannot be placed in that order gets deferred, not because it lacks value, but because it lacks a coordinate.

Your competitor for that funding, the initiative that got approved while yours got skipped, was very likely not better work. It was better positioned. Someone attached a defensible number to it early, socialized that number before the meeting, and walked in with a figure the finance lead could drop straight into the ranking. That is the whole game. The number does not have to be big. It has to exist, and it has to hold up when someone pushes on it.

This is why I keep saying the leader who keeps their funding is rarely the one with the strongest portfolio. It is the one who puts a number on the table before anyone asks for it.

Part 3: The Number You Should Have Had

So what was the number that would have saved your initiative? Not a made-up one. Not a hopeful projection you cannot defend. A quantified, traceable figure that connects the technical work you did to a business outcome someone in that room already cares about.

Most technical leaders do not have that number for a simple reason: nobody ever taught them how to build one. Quantifying the value of data, analytics, and integration work is a specific skill, and it is not intuitive. It requires a repeatable method for turning “we improved the pipeline” into “we reduced X by Y, worth Z annually.” That translation is learnable, and once you have the method, you never walk into a budget review naked again.

That method is exactly what I wrote Quantify Your Impact to teach. It is the first book in the BRM Accelerator Series, and it exists because I got tired of watching good people lose funding for good work simply because they never learned to price it. If your best work has ever been quietly skipped in a room full of numbers, this is the book that fixes the cause, not the symptom.

Get it here: Quantify Your Impact on Amazon ($24.95, paperback).

Think different for different results.