How Should I Use AI for Every Part of My Business?

Let AI do the legwork of finding customers—scanning markets, cutting segments, flagging opportunities, running outreach experiments and reading the results. What stays yours is the judgment: which market to chase, how to position, which relationships to build, and which tradeoffs protect your reputation. The Dermer Rule here isn't more prospecting tools bolted on—it's letting AI widen the field of possibilities while you decide which promise is worth making.

Lead generation is mostly volume and pattern work, so hand AI the sourcing, list-building, scoring, personalized outreach at scale and the constant testing of what lands. Your 20% is human: reading which leads are actually a fit, owning the positioning behind the message, and standing behind the brand every touch represents. Don't just automate the old funnel—rebuild it so AI multiplies qualified conversations and you own who's genuinely worth pursuing.

Before the conversation and after it, AI earns its keep: researching the account, prioritizing who to call, tailoring proposals, logging the CRM, chasing follow-ups and analyzing what happened on the call. But a deal closes on trust, real discovery, persuasion and the judgment to know when to push or walk. Redesign selling so reps spend their hours in the room with the buyer—because the reason someone says yes stays stubbornly human.

AI can carry the production line of content—researching topics, drafting, tailoring versions, publishing across channels, testing headlines and reading the analytics. What it can't supply is taste, a point of view, cultural read, or the promise your brand is staking. Treat this as a redesign, not a faster copy machine: let AI flood the pipeline with drafts and data, and keep yourself accountable for why any of it deserves attention.

Marketing runs on repeatable motion—campaign builds, audience research, personalization, distribution, endless testing and reporting—and all of that is AI's to run. Your job is the part machines fake badly: the positioning, the emotional truth of the message, the cultural timing and the reputation on the line. The Dermer Rule says don't wire AI into last year's plan; redesign the whole engine so humans own meaning and AI owns motion.

Sourcing candidates, sifting résumés, scheduling, drafting job docs and surfacing signals from the pipeline—hand all of it to AI, which handles volume without fatigue. The human 20% is the weighty part: judging potential, protecting fairness, and reading whether someone will actually fit the culture. A hire changes a person's life, so the Dermer Rule is firm here—let AI narrow the field, but never let it make the call only accountable humans should own.

Day-to-day people management has a mechanical layer AI can absorb: documenting one-on-ones, scheduling, tracking performance signals, surfacing who needs support and suggesting coaching prompts. What can't be delegated is the coaching itself—the empathy, the hard conversations, the fairness calls and the culture you set by example. Rebuild the manager's week so AI clears the admin, and the leader spends real time on the human work that makes people stay and grow.

Cash flow is a data problem before it's a judgment problem, so let AI run the forecasts, monitor balances daily, flag anomalies, model scenarios and keep collections intelligent. The decisions—how much risk to carry, where the money goes, which tradeoff you can live with—stay with you and your accountability to stakeholders. The point isn't a prettier dashboard; it's redesigning the finance rhythm so the economics are always visible and a human owns the calls.

When you're raising, AI is a tireless analyst: building the model, stress-testing scenarios, assembling the data room, monitoring metrics and drafting the numbers behind your story. Investors, though, back conviction—so the risk appetite you signal, the tradeoffs you defend and the relationships you build are irreducibly yours. Don't bolt AI onto a stale pitch; redesign the raise so the analytics are bulletproof and you carry the vision and the accountability into the room.

Strategy work is drowning in inputs, and that's exactly what AI should absorb—synthesizing research, generating scenarios, monitoring the landscape and drafting the decision memos. But vision, courage, values and the priority calls that define a company can't be outsourced; they're the human 20%. Use AI to raise the quality of what you're deciding on, then own the decision itself—the Dermer Rule treats better inputs and braver choices as two different jobs.

AI is good at the raw material of prioritization: pulling the data, laying out tradeoffs, modeling what each path costs and returns, and keeping the list current. Choosing what actually matters—and what you'll deliberately ignore—takes conviction and values no model holds. The reframe: let AI make the options and consequences crystal clear, and reserve for yourself the courage to say no to good things so the great ones get your attention.

Most decisions get worse from missing or messy information, so put AI on the synthesis: gather the evidence, frame the options, run the scenarios and stress-test your assumptions. The choice itself—weighing values, accepting the risk, owning the outcome—belongs to a human who can be held accountable. Rather than a decision tool bolted onto old habits, build a practice where AI sharpens the inputs and you supply the judgment.

Give AI the work around the work—scheduling, drafting, prepping for meetings, chasing follow-ups, automating the routine tasks that eat a calendar. What no assistant can decide is your intent: where your focus should go, which boundaries to hold, and what's actually worth your finite attention. The Dermer Rule on time isn't cramming more in; it's redesigning your days so the machine handles the busywork and you protect the hours only you can spend.

Scaling multiplies the same motions—finding new segments, testing offers, standardizing outreach, analyzing what works and doubling down. AI thrives on that repetition, so let it run the experiments and the optimization at a pace no team could match by hand. Your 20% is deciding which markets and relationships are worth the stretch and guarding the reputation growth can strain. Scale is a redesign of how work happens, not just doing the old thing faster.

To cut costs well, let AI do the forensic work: model spend, monitor every line, detect anomalies, benchmark vendors and run scenarios on where the fat actually is. The judgment—what to protect, what risk a cut introduces, which stakeholders it touches—is yours to own. Don't just point AI at the ledger to trim blindly; redesign the cost base so the economics are transparent and a human decides what the business can't afford to lose.

Pricing has a heavy analytical base AI should own—modeling elasticity, analyzing competitor moves, segmenting willingness to pay, testing offers and forecasting the revenue impact of each option. But price is a signal of value and trust, and the final number takes nerve and a read on the customer relationship that stays human. The Dermer Rule: let AI expose what the market will bear, and keep the courage to set—and defend—a price yourself.

A launch generates mountains of prep—research synthesis, messaging drafts, test plans, documentation, analytics dashboards—and AI can produce and iterate on all of it fast. What decides success is more human: whether you're solving the right problem for the right customer, and the taste behind what ships. Don't automate a checklist and call it transformation; redesign the process so AI accelerates the building while you own the judgment about what's worth putting into the world.

Building leans on AI for the grind—synthesizing user research, spinning up prototypes, writing docs, running tests and turning analytics into the next iteration. The irreducible 20% is choosing which problem is worth solving, holding the customer's context, and the taste that separates a feature that ships from one that shouldn't. Treat AI as a redesign of how you build, not a faster keyboard, and keep humans accountable for what actually deserves to exist.

The ordinary ticket—password resets, order status, how-to questions, routing and logging—should resolve without a human ever touching it, and AI does that around the clock. The moments that decide whether a customer stays are different: the angry escalation, the recovery, the exception no script covers. Automate the routine ruthlessly so your people are free for exactly those moments—because when trust is on the line, a real human owning the relationship is the whole point.

Everything before you sit down is AI's to prepare: researching the other side, modeling scenarios, mapping your alternatives, pricing concessions and drafting proposal options. At the table, though, it's all human—reading the room, timing a move, building trust and finding the courage to hold or fold. The Dermer Rule reframes negotiation as preparation plus presence: let AI make you the most prepared person in the room, and keep the reading of the person to yourself.

AI is a superb early-warning system—monitoring continuously, spotting anomalies, modeling how risks cascade and flagging what needs a look before it becomes a crisis. Deciding which risks to accept, which cross an ethical line, and who owns the consequences is squarely human. The shift isn't more alerts; it's redesigning risk so AI helps you see further and sooner, while accountable people decide which risks the business will actually carry.

In a crisis the fog is the enemy, so let AI cut through it—synthesizing what's known, modeling scenarios, drafting communications and tracking the situation as it moves hour to hour. The calls that matter—what to stand for, what to sacrifice, how to lead scared people—demand courage and values no model carries. Use AI to keep the picture clear and the drafts ready, and keep the hard, human decisions and their accountability with you.

Productivity gains hide in the work around the work, and AI can take most of it—drafting, scheduling, prepping, automating handoffs and suggesting what to tackle next. What stays yours is intent: deciding what actually deserves effort and protecting the focus that produces real output. Don't sprinkle AI over a broken workflow; redesign how the work flows so the machine absorbs the friction and humans spend their energy where judgment pays off.

Point AI at your processes and it will map the steps, design the automation, run the flow, monitor it and catch the exceptions. The human work is drawing the lines: where automation must stop, what triggers an escalation, and who owns it when something goes wrong. Automate aggressively—but the Dermer Rule insists you define exactly where the machine hands control back to an accountable person, because a workflow with no human boundary is a liability, not an upgrade.

Deciding what to automate starts with AI doing the discovery—mapping tasks, timing them, scoring which are repeatable and modeling the payoff of each. But the choice of what should be automated, versus what must stay human for judgment or trust reasons, is yours to make and own. The reframe: let AI show you where the mechanical 80% really lives, then apply human judgment to protect the 20% that gives your business its edge.

Rolling out AI is itself a project AI can help run—discovering use cases, mapping capabilities, benchmarking options, analyzing workflows and drafting the implementation plan. The strategy behind it, though, is human: which priorities matter, how much risk you'll tolerate, and where you draw the line that must stay human. The core Dermer Rule move is to treat AI as a redesign of how work gets done, not a pile of tools you install and hope helps.

AI can personalize the training itself—assessing each person's level, generating practice, answering questions on demand and tracking who's actually adopting it. What it can't do is lead the change: building trust, redesigning roles honestly, and giving people a reason to embrace the shift instead of fearing it. Teach the whole team to use AI fluently, but the Dermer Rule is protective—guard the human capabilities that make your company worth working for as you do it.

AI can help you audit the work—mapping every task, benchmarking what it can now do and modeling what a redesigned role looks like. But the line itself—what must stay human because it depends on trust, ethics, taste or accountability—is the most important judgment a leader makes, and it can't be delegated to the thing being evaluated. Draw that boundary deliberately: the Dermer Rule is as much about protecting the 20% as automating the 80%.

An AI strategy needs a clear-eyed inventory, and AI can produce it: discovering use cases, mapping your capabilities against them, benchmarking competitors and modeling where the return is real. Setting the priorities, the risk tolerance and the human boundaries is leadership work you own outright. Strategy here means designing how the whole organization will work differently—not shopping for tools—so keep the accountability for that redesign firmly in human hands.

Measuring AI's return is a data exercise AI runs well—tracking usage, comparing to baselines, quantifying time and cost saved and reporting the trend. What a model can't settle is what counts as value, how much of a gain to credit to AI, and whether to keep investing. Measure outcomes, not activity: the Dermer Rule warns against mistaking busy AI for valuable AI, and keeps a human accountable for deciding whether the return is real.

A business plan is built on research, and AI can assemble all of it—market intelligence, competitor teardowns, financial models and a stack of scenarios to pressure-test. But a plan is ultimately a bet, and the vision, the tradeoffs and the conviction to commit are yours alone. Let AI help you understand every possibility on the table; the Dermer Rule leaves the courage to choose a single future—and stand behind it—to the human.

To validate an idea, put AI to work sizing the market, scanning competitors, modeling unit economics and generating the scenarios where it wins or dies. The verdict, though, rests on human conviction and a willingness to kill a favorite idea when the evidence says so. Don't let AI's tidy analysis substitute for the founder's judgment—use it to see the possibilities clearly, then own the choice to pursue, pivot or walk away.

Market research is information work at its purest, so hand AI the heavy lifting: gathering intelligence, mapping segments, analyzing competitors, synthesizing trends and modeling how the market might move. Interpreting what it means for your specific bet—and having the courage to act on it—is the human 20%. The reframe isn't a faster report; it's letting AI understand more of the market than any analyst could, while you decide which future is worth choosing.

AI can watch your competitors relentlessly—pulling their moves, comparing positioning, modeling scenarios and synthesizing where the gaps and threats are. Turning that intelligence into a differentiated bet takes human judgment, taste and the nerve to zig where the market zags. Rather than a one-off competitor deck, build an always-on picture AI maintains, and keep for yourself the strategic choice about where you'll actually win.

Defining your ideal customer starts with AI mining the data—clustering your best accounts, analyzing behavior, personalizing profiles and testing which segments respond. But choosing who you're truly for, and the promise you'll make them, is a positioning decision only a human should own. Don't let the algorithm pick your customer by default; use AI to reveal the patterns, then apply the taste and judgment that decide who you build the business around.

AI can supply the raw material of positioning—researching how the market talks, analyzing competitors' claims, drafting variations and testing which messages resonate. The positioning itself, the promise and the cultural read behind it, is the human 20% that defines the brand. This is a redesign of how you find your words, not a slogan generator: let AI expand and test the options, and keep the meaning and the reputation on the line yours.

A brand needs constant production—content, personalized touchpoints, campaign execution, testing and distribution—and AI can run that machine tirelessly. What it will never own is taste, point of view, cultural timing or the trust a brand accrues over years. The Dermer Rule here protects the 20%: automate the output so it's everywhere and consistent, and guard the human judgment about what the brand stands for and why anyone should believe it.

Sales emails are a volume game AI wins—researching each prospect, personalizing at scale, drafting sequences, timing follow-ups and analyzing what gets replies. The trust that turns a reply into a relationship, and the read on what a specific buyer actually needs, stays human. Let AI write more and better first drafts than any rep could by hand; keep yourself responsible for the genuine connection no template, however personalized, can manufacture.

Prospecting is largely research and repetition, so let AI build the lists, score the fit, personalize the opener and keep the CRM honest. Judging whether a prospect is truly worth pursuing, and building the early trust, is the human 20%. The reframe: stop measuring reps on activity they shouldn't be doing—redesign prospecting so AI handles the grind and people spend their time only on conversations that could actually become deals.

Qualification has an obvious mechanical layer AI should own—scoring leads against your criteria, enriching records, prioritizing the queue and flagging who's gone cold. The real discovery, the read on whether a fit is genuine, needs a human who can ask the question behind the question. Use AI to rank and route so nothing slips; keep the judgment about who's truly ready to buy—and worth your reps' hours—firmly human.

Call prep is tailor-made for AI: researching the account and the person, summarizing history, drafting talking points, anticipating objections and prepping the proposal. What happens live—the discovery, the trust, the persuasion—is irreducibly human. Don't send reps in cold or over-scripted; redesign prep so AI makes them the most informed person on the call, and let them spend that advantage on genuinely connecting with the buyer.

Sales forecasting is a numbers problem AI handles well—analyzing pipeline, weighting deals, spotting patterns in what closes and modeling the range of outcomes. Calling the number you'll commit to, and reading the deals AI can't see inside, takes human judgment and accountability. Let AI produce a sharper, faster forecast than any spreadsheet ritual; keep the human on the hook for the commitment and the story behind it.

Pipeline hygiene is exactly the repetitive work to give AI—updating stages, flagging stalled deals, prioritizing follow-up, surfacing risk and keeping the CRM current. Deciding which deals get your reps' scarce time and reading what a stuck deal really needs is the human 20%. Rebuild pipeline management so AI keeps the board clean and honest, and people apply judgment to the handful of deals where attention actually changes the outcome.

AI can do the observation work of coaching—analyzing call recordings, spotting patterns, benchmarking reps and surfacing exactly where each one loses deals. The coaching itself—the trust, the tough feedback, the belief that lifts a struggling rep—is human and can't be automated. Let AI find the teachable moments at scale; the Dermer Rule keeps the human relationship that turns a data point into someone who actually gets better.

Advertising rewards relentless testing, and that's AI's strength—producing variations, targeting audiences, executing campaigns, optimizing spend and reading performance in real time. The creative idea, the promise and the brand-safe judgment behind it stay human. Don't just let AI optimize a bad ad more efficiently; redesign the creative process so AI runs the media machine and people own the idea worth spending money to spread.

Social is a content treadmill AI can carry—drafting posts, scheduling, tailoring per platform, testing formats and reporting on what performs. The voice, the cultural read and the judgment about what to say when trust or reputation is at stake belong to a human. Automate the cadence and the analytics; keep the human hand on tone and timing, because one tone-deaf post AI would happily publish can cost more than a month of content earns.

SEO is heavy on repeatable production—keyword research, content drafts, technical audits, internal linking and tracking rankings—all of which AI does faster than any team. Deciding what your brand should actually be known for, and protecting quality over volume, is the human 20%. The reframe: let AI handle the mechanics of visibility, and keep yourself accountable for creating content people—and search engines—actually value, not just more pages chasing a keyword.

Ongoing market research—surveys, review mining, audience analysis, trend tracking and synthesis—is exactly the information work AI should run continuously rather than in occasional bursts. Interpreting what the findings mean for your positioning, and which signals to trust, is the human part. Don't treat research as a report you commission and shelve; redesign it as a living feed AI maintains, with people owning the judgment about what it changes.

Personalization is a scale problem AI was built for—segmenting audiences, tailoring content and offers, triggering the right message at the right moment and optimizing on response. The line between helpful and creepy, and the brand promise underneath it all, is a human call. Let AI personalize at a scale no team could touch; keep a person accountable for respecting the customer, because trust lost to a tone-deaf automation is expensive to win back.

Video has a long production tail AI can shorten dramatically—scripting drafts, editing, generating variations, captioning, clipping for social and testing thumbnails. The story worth telling, the taste in the cut and the emotional truth are human. Rather than churning out more forgettable clips, redesign production so AI handles the labor and people own the creative judgment that makes a video worth watching and sharing.

Producing a podcast is mostly post-production and packaging, so hand AI the editing, show notes, clip creation, titles, descriptions and cross-platform distribution. What earns an audience—your voice, your point of view, the guests you choose and the trust you build episode by episode—is irreducibly human. Automate the production line so you can publish consistently; keep for yourself the reason someone chooses to press play and stay.

Repurposing is pure mechanical leverage and belongs almost entirely to AI—turning one piece into threads, clips, newsletters, summaries and posts tuned to each channel. The editorial judgment about what's worth amplifying, and keeping the brand's voice intact across formats, stays human. The Dermer Rule reframes this as designing a content engine, not a copy-paste chore: let AI multiply reach while a person guards the taste that made the original worth repurposing.

AI can watch your reputation everywhere at once—monitoring mentions, reading sentiment, detecting an issue as it sparks and drafting a first response. But what the company actually says when trust is at stake demands human empathy, timing and the courage to take a stance. Let AI give you the early warning and the raw draft; keep a person accountable for the words, because a reputation is repaired or ruined by judgment no model can supply.

Screening is high-volume filtering, and AI does it faster and more consistently than a tired recruiter—parsing résumés, matching skills, ranking against the role and surfacing candidates you'd have missed. Judging genuine potential and guarding against bias in how that filter is built stay human. Use AI to widen and speed the top of the funnel; keep people accountable for fairness and for the judgment that a résumé never fully captures.

AI can shoulder the mechanics of interviewing—scheduling, generating role-specific questions, summarizing notes, transcribing and comparing candidates against a rubric. The interview itself, though, is a human read on potential, character and fit that shapes someone's livelihood. Let AI remove the logistics and the bias-prone shortcuts; the Dermer Rule keeps a person in the room and accountable for a decision that changes another person's working life.

New-hire onboarding has a repeatable backbone AI should run—provisioning docs, scheduling, answering routine questions, personalizing the learning path and tracking progress. What makes someone feel they belong—the mentorship, the culture, the early trust—is human work no workflow replaces. Automate the administrative first week so it's smooth and consistent, and free managers to do the human part that actually determines whether a good hire stays.

AI can assemble the evidence behind a review—compiling performance signals, summarizing feedback, drafting the write-up and flagging patterns over time. The judgment about someone's growth, the fairness of the call and the conversation itself are irreducibly human. Let AI end the scramble for data and the blank-page dread; keep the manager accountable for the assessment, because a review shapes a career and can't be handed to a model.

Coaching benefits from AI on the periphery—surfacing performance signals, suggesting development resources, drafting frameworks and tracking whether growth goals move. The coaching relationship—the trust, the honest feedback, the belief in someone's potential—is human to its core. Don't confuse a personalized learning plan with mentorship; use AI to inform the coaching, and keep the human connection that actually helps a person grow.

AI can prepare you for conflict—documenting what happened, summarizing perspectives, surfacing relevant policy and suggesting approaches to the conversation. The resolution itself takes empathy, presence and fairness no tool can fake, and the accountability sits with a human. Let AI handle the prep and the record; the Dermer Rule is unambiguous here—when emotions and relationships are on the line, a person owns the room, not a machine.

The paperwork around a termination—documentation, policy checks, timeline summaries, required notices—is where AI helps, handling the mechanics accurately. The decision and the conversation are among the most human acts in business, demanding fairness, dignity and full accountability. Never let AI make or deliver this call; use it only to ensure the process is clean and defensible, and keep the weight of a decision that upends someone's life squarely with a person.

AI can give you instruments for culture—analyzing sentiment, surfacing signals from surveys and communication, and flagging where morale is drifting. Culture itself is built by human example: the values you model, the fairness you enforce and the trust you earn daily. Use AI as the dashboard, never the driver—the Dermer Rule keeps leaders accountable for the culture, because a company's character is set by people, not measured into existence.

Budget-building is largely modeling and consolidation, so let AI forecast, roll up the numbers, run scenarios, flag variances and keep the whole thing current. Allocating capital, setting the risk appetite and defending the tradeoffs to stakeholders is the human 20%. Don't just speed up the annual spreadsheet ritual; redesign budgeting as a living model AI maintains, with a person owning the choices about where the money actually goes.

Forecasting is exactly what AI does best—ingesting history, modeling drivers, running scenarios and updating the outlook as reality shifts. Deciding how much to trust a forecast, what risk it implies and what to do about it stays human and accountable. The reframe isn't a fancier model; it's making the financial future continuously visible so leaders spend their judgment on decisions, not on rebuilding the spreadsheet every month.

Expense analysis is tedious and pattern-heavy—precisely AI's strength. Let it categorize spend, detect anomalies, benchmark against plan and surface where the money is quietly leaking. The call on what to cut, what to protect and which tradeoff to accept belongs to a human weighing context AI can't see. Automate the forensics so nothing hides in the ledger, and keep the accountable judgment about the business's spending with a person.

Collections runs on timing and repetition, so let AI monitor receivables, predict who'll pay late, draft reminders and prioritize the follow-up that recovers the most cash. Deciding how hard to push a valued customer, and protecting the relationship while getting paid, is a human judgment. Automate the chase so nothing ages unnoticed; keep a person accountable for the tone, because collections tests a relationship as much as a balance sheet.

AI can arm every capital decision—modeling returns, running scenarios, benchmarking options and monitoring how past bets are paying off. But allocating capital is the defining act of leadership: choosing which future to fund takes conviction, risk appetite and accountability no model holds. Let AI make the tradeoffs and consequences vivid; the Dermer Rule leaves the decision about where the company's money goes firmly, deliberately human.

Bookkeeping is about as mechanical as work gets, so hand AI the categorization, reconciliation, anomaly detection, reporting and the endless data entry. What stays human is the judgment on gray-area treatments, the context behind the numbers and accountability for their accuracy. This is a clear redesign, not an assist: let AI keep the books clean and current in the background, and reserve people for the calls that require actual judgment.

Producing the reports—compiling data, building statements, drafting variance explanations and formatting the pack—is work AI should own end to end. Deciding what the numbers mean, what to emphasize and standing behind them to stakeholders is the human 20%. Don't just generate prettier reports faster; redesign reporting so AI handles the assembly and a person owns the narrative and the accountability that a signature on the numbers implies.

AI can do the modeling behind tax planning—running scenarios, tracking obligations, flagging opportunities and organizing the documentation. The strategy, the risk tolerance on aggressive positions and the accountability for what you file demand human judgment and qualified advice. Use AI to make the tax picture visible and the options clear; keep a person—and the right professional—responsible for the decisions, because the consequences of getting tax wrong land on humans, not models.

AI can run the machinery of a project—building the schedule, tracking status, planning resources, flagging slippage and keeping the documentation current. What it can't do is navigate the exceptions, referee competing priorities or read the human dynamics on the team. Automate the normal path so nothing falls through the cracks, and keep the project manager accountable for the judgment calls and the coordination no Gantt chart captures.

Drafting SOPs is documentation work AI accelerates well—observing a process, writing the steps, formatting the doc and keeping versions in sync as things change. Deciding which processes deserve a standard, and owning the judgment about the exceptions the SOP can't foresee, is human. Let AI turn tribal knowledge into clear documents fast; keep a person accountable for whether the standard actually reflects how the work should be done.

Operational scheduling—shifts, resources, capacity, sequencing—is a constraint-solving problem AI handles better than any wall calendar, optimizing and re-optimizing as conditions change. The human 20% is the exceptions: the fairness of who works when, the last-minute judgment call and the coordination a solver can't feel. Automate the optimization so the baseline schedule builds itself, and keep people accountable for the exceptions where context beats the algorithm.

Procurement has a repeatable core AI should run—sourcing options, comparing quotes, tracking orders, monitoring spend and flagging exceptions. The relationships with key suppliers, the negotiation and the judgment on strategic risk stay human. Don't just digitize the purchase order; redesign procurement so AI handles the routine buying at speed, and people own the supplier relationships and the calls a price comparison alone can't make.

Inventory is forecasting plus monitoring, and AI excels at both—predicting demand, optimizing reorder points, tracking stock and alerting before you run short or overbuy. The judgment on a surprise, a supplier failure or a demand shock nobody modeled is human. Let AI keep the ordinary stock decisions humming automatically, and reserve people for the exceptions where field context and accountability matter more than the model's best guess.

Supply chain planning rewards AI's ability to model complexity—forecasting, optimizing routes and resources, monitoring the network and detecting disruptions early. Deciding how to respond to a real shock, which relationships to protect and what risk to carry is human. The reframe isn't a smarter planning tool bolted on; it's redesigning the chain so AI runs the normal flow and accountable people own the exceptions no plan survives.

AI can inspect at a scale and consistency people can't—monitoring output, detecting defects, spotting drift and documenting every check. Deciding what quality means for your brand, and owning the judgment when a borderline case or a customer's trust is at stake, is human. Automate the detection so nothing slips through, and keep a person accountable for the standard itself—because quality is a promise, and a promise is owned by people.

The administrative side of vendor management—tracking contracts, monitoring performance, flagging renewals and documenting issues—is AI's to run. The relationship, the negotiation and the judgment on whether a vendor is truly a fit stay human. Let AI keep every vendor honest and every deadline visible; the Dermer Rule keeps a person accountable for the relationships and the strategic calls a scorecard can inform but never make.

Getting a new customer up and running has a routine core AI should own—guided setup, answering common questions, surfacing the right resources and flagging when someone's stuck. The moment a customer feels genuinely cared for, especially when something goes wrong early, needs a human. Automate the standard journey so it's smooth and instant, and keep people ready for the exceptions that decide whether a new relationship takes root or quietly churns.

Retention leans on AI for the early signals—analyzing usage, predicting churn, triggering the right nudge and surfacing accounts that need attention. But saving an at-risk relationship takes human empathy, a real conversation and the trust a well-timed email can't manufacture. Let AI tell you who's slipping and why; keep people accountable for the human moments that actually change a customer's mind about staying.

AI can triage complaints instantly—classifying, routing, retrieving history and drafting a first response—so nothing sits ignored. The recovery itself, turning an angry customer into a loyal one, runs on empathy and trust only a human delivers. Automate the intake and the routine fixes; the Dermer Rule keeps a person on the moments that matter most, because how you handle a complaint often decides whether the relationship survives at all.

AI is excellent at spotting expansion signals—analyzing usage, flagging the right moment, recommending the next product and personalizing the pitch. Whether an offer feels genuinely valuable or merely extractive is a human judgment rooted in the relationship and its timing. Let AI find the opportunity and prep the account; keep a person accountable for making growth feel like service, not a squeeze—because trust, once spent on a bad upsell, is hard to earn back.

AI can inform a roadmap richly—synthesizing user feedback, analyzing usage, modeling scenarios and drafting the options. Choosing what to build, and having the taste and conviction to say no to the rest, is the human 20% that defines the product. Don't let the loudest data pick your roadmap; use AI to see every possibility clearly, and keep a person accountable for the prioritization that decides where the product goes.

User research generates a flood of raw material—interviews, surveys, session data—that AI can transcribe, synthesize and surface patterns from in a fraction of the usual time. Deciding which insights matter and reading the human context behind them stays with you. The reframe: let AI handle the synthesis so researchers spend their judgment on interpretation, not transcription, and keep people accountable for what the findings actually mean for the product.

Testing is repetitive and data-rich—ideal for AI to run: generating cases, executing tests, analyzing results and documenting what broke. Deciding what "good enough to ship" means, and owning the taste and judgment behind quality, is human. Automate the test grind so coverage is broad and fast; keep a person accountable for the call about whether the product is actually ready for the people who'll use it.

AI can generate, refactor, document and debug code faster than any developer typing alone—that's the mechanical 80%. Architecture, security, the system-wide tradeoffs and accountability for whether the thing should work this way are human. Let AI produce more code; the Dermer Rule keeps a person responsible for the design decisions and the judgment that separate code that merely runs from a system you'd stake the business on.

Software testing is exactly the repeatable, high-volume work AI should own—writing test cases, running suites, catching regressions and documenting failures. Judging what's worth testing, interpreting a subtle failure and owning the risk of shipping stays human. Automate coverage so quality scales with the codebase, and keep an accountable engineer deciding what "safe to release" means—because a green test suite is evidence, not a guarantee.

AI is built for security monitoring—watching traffic around the clock, detecting anomalies, correlating signals and flagging threats faster than any human eye. Deciding how to respond, what risk to accept and owning the consequences of a breach stays human. Let AI see everything and surface what matters; the Dermer Rule keeps accountable people making the calls, because security decisions carry consequences a model can't be held responsible for.

AI can prep the data, run the analysis, detect patterns, forecast and draft the visuals—the grind that used to eat analysts' weeks. Choosing the right question, judging correlation from causation and framing what to actually do about it is human. Use AI to find the patterns at speed; keep a person accountable for what the patterns mean, because a confident chart pointed at the wrong question is worse than no analysis at all.

Report-building is assembly work AI should own—pulling the data, running the analysis, drafting the visuals and writing the first-pass narrative. Deciding what the audience needs to know, and what the numbers really imply for the decision at hand, is human. Don't just generate more dashboards; redesign reporting so AI handles the production and a person owns the interpretation and the call to action that make a report worth reading.

AI can process the law at scale—searching case law, summarizing statutes, spotting issues and comparing documents far faster than manual review. The strategy, the ethical judgment and the advice a client relies on stay human and accountable. Let AI do the reading and the first-pass analysis; the Dermer Rule keeps a qualified person responsible for the counsel, because legal advice carries consequences a model cannot answer for.

Reviewing contracts is pattern work AI handles well—flagging non-standard clauses, comparing against your playbook, summarizing obligations and spotting missing protections. Judging which risks matter for this deal, and negotiating them, takes human legal judgment and context. Let AI catch what a tired reviewer misses late at night; keep a person accountable for the strategic calls and the advice that decide whether you sign.

Drafting benefits from AI's speed—generating first drafts from your templates, tailoring clauses, checking consistency and comparing versions. The judgment about what to include, what to concede and the ethics of the terms is human. The reframe isn't a faster template; it's letting AI produce the workmanlike draft so a lawyer spends their expertise on strategy and the accountability a client's signature ultimately rests on.

Compliance generates endless evidence and monitoring—exactly the continuous, repeatable work AI should run: mapping controls to policy, collecting proof, testing and reporting. What compliance actually requires in a gray area, the ethics and the accountability to a regulator, is human. Automate the evidence trail so audits stop being fire drills; keep a person responsible for the materiality judgments, because a checkbox is not the same as being genuinely compliant.

AI can prepare the board better than any late-night scramble—synthesizing the materials, benchmarking performance, analyzing risk and tracking follow-ups from last time. The governance itself—the hard questions, the independent judgment, the accountability—is irreducibly human. Use AI to raise the quality of what the board sees; the Dermer Rule keeps directors owning the challenge and the oversight, because governance is a duty, not a document.

Compiling an investor update is mechanical—AI can pull the KPIs, explain the variances, draft the narrative and format it cleanly. What builds investor trust is candor and the judgment about what actually matters, especially when the news is bad. Let AI assemble the update; keep a person accountable for telling investors the truth that matters, because relationships with backers are built on honesty a polished draft can't supply.

Selling a company is a process AI can optimize—researching buyers, supporting the valuation, organizing the data room, modeling scenarios and tracking every workstream. But what the company is worth to you, the timing, the legacy and the negotiation are deeply personal and human. Let AI run the mechanics flawlessly; keep the founder accountable for the decision, because the value of a sale is measured in more than the multiple on the term sheet.

AI can analyze an acquisition thoroughly—screening targets, building the model, testing synergy scenarios and reviewing documents at a pace no deal team matches. Whether the company is truly a fit, whether you trust its leaders and whether you can integrate it is human judgment. Use AI to understand the deal cold; the Dermer Rule leaves the conviction—whether you should actually own this business—to accountable people.

Diligence is document-heavy grind, and AI shines here—reviewing contracts, analyzing financials, screening for red flags and organizing findings far faster than a room of analysts. Interpreting what a finding means for the deal, and the judgment on integration risk, stays human. Let AI surface everything hiding in the data room; keep a person accountable for the read on whether the risks are ones the company can actually live with.

AI can inform succession—synthesizing candidate data, modeling scenarios, benchmarking readiness and tracking development plans. But choosing who carries the company's future is among the most human decisions a leader makes, rooted in values, trust and courage. Use AI to make the picture clear and the options honest; the Dermer Rule keeps the choice—and the accountability for the company's continuity—firmly with the people who'll answer for it.

Entering a new market is research-intensive, and AI can carry it—sizing the opportunity, segmenting, mapping competitors, running entry experiments and analyzing early results. The judgment about whether the market fits, how to position and which relationships and reputation risks to accept is human. Don't just run the old playbook faster in a new place; redesign the entry so AI tests more possibilities and a person owns the choice of market and promise.