Data Friction Matrix

The Data Friction Matrix: The AI Readiness Reality Check

The Data Friction Matrix: The AI Readiness Reality Check

[AI Won't Fix Bad Data]

If you throw GenAI at a broken data model, you don’t scale intelligence. You just accelerate bad decisions with absolute confidence.

That line landed harder than expected last week. Several people asked the obvious follow-up: how do I know if my organization is actually at risk?

The short answer: you don’t know — and that’s the problem. Data friction doesn’t live in one system. It lives in the handoffs, the manual reconciliations, and the reports nobody trusts, which is why it survives budget reviews and then resurfaces as a failed AI initiative. The fastest way to get an honest answer is to score it. The 12-point Data Friction Matrix measures bottlenecks across four pillars, weights each by damage, and returns a single number out of 72. At 19 or above, you are already paying the Data Friction Tax. Three minutes, no signup.

Here's where we're headed—so you can jump straight to what matters most to you
    Add a header to begin generating the table of contents
    Scroll to Top

    The AI Silver Bullet Is a Data Problem

    Most leaders can’t answer the question above — not because they’re careless, but because data friction is distributed. It doesn’t live in one system. It lives in the handoffs between systems, the reconciliation someone does every Monday morning, the report nobody trusts, the dashboard that gets rebuilt every quarter because the last one “didn’t feel right.”

    Friction spread across forty small inefficiencies is invisible in aggregate. That is exactly why it survives budget cuts, and exactly why AI initiatives built on top of it fail — not loudly, but expensively.

    Before we talk about audits, we need to talk about how you find out whether you need one.

    What the Matrix Measures

    Opinions about data maturity are cheap and plentiful. Numbers are what get you a budget line.

    The Data Friction Matrix is a 12-point diagnostic that scores your operational bottlenecks across four pillars:

    📊 Data & Architecture — silos, manual reporting, zero single source of truth

    🕸 Strategic Process — IT miscommunication, misaligned priorities, tech debt

    ⚙ Technical Execution — inability to scale, legacy blockers, lost engineering hours

    🧠 Cognitive & Team — complex user adoption, decision overload, process fatigue

    Each point is weighted by how much damage it does, and the tool calculates a total out of 72.

    The threshold matters more than the score. At 19 or above, you are actively paying a Data Friction Tax — and that tax is quietly eating your AI readiness and your project margins at the same time.

    The diagnostic is free, takes three minutes, and requires no signup. It’s in the Featured section of my LinkedIn profile. Run it. You’ll have a number by the time your coffee goes cold.

    Why a Score Isn't an Audit

    Here’s where most people stop, and where the expensive mistakes begin.

    A score tells you that you have friction. It doesn’t tell you:

    • Where it’s concentrated. A 32 driven by legacy architecture and a 32 driven by decision fatigue require completely different interventions — and one of them costs ten times more to fix.

    • What it’s actually costing. “High friction” is a diagnosis. “€180K a year in lost engineering hours plus a stalled AI initiative” is a business case.

    • What order to fix it in. Almost every organization has more friction than budget. The question isn’t what’s broken — it’s what breaks first if we do nothing.

    • Whether your team agrees. The most useful output of an audit isn’t the report. It’s the moment the COO and the CTO look at the same number and realize they’ve been measuring different things.

    A self-diagnostic gives you a symptom. It cannot give you a treatment plan.

    The 72-Hour Data Friction Audit

    This is where I have a bias, so take it as such.

    The standard move is to bring in a large firm. They arrive with a framework, interview three stakeholders, disappear for six weeks, and return with a 60-page document that confirms what two people in the room already knew — now with a deadline attached and nobody who owns the fix.

    The thicker the report, the easier it is to file it away and move on to the next crisis.

    What actually works is narrower and faster. A structured audit should answer four questions and stop:

    1. Where is the friction concentrated, by pillar and by process?
    2. What is it costing, in euros and in hours, per year?
    3. What order do we fix it in, given the budget we actually have?
    4. Who owns each fix, by name?

    That’s it. No maturity model. No transformation roadmap. No 60 pages.

    Once you know your score, the next step is applying a Strategic Alignment Framework to actually close the gap.

    This is exactly what the 72-Hour Data Friction Audit is built to deliver. Three days, one business process, four answers, one owner per friction point. Bounded scope, bounded cost, and a result your team can act on the following Monday — because the people who will implement the fix are in the room while it’s being designed.

    If you want to bypass the guessing game entirely, you can book a comprehensive data friction audit to map out your infrastructure in 72 hours.

    The Point

    AI doesn’t create data problems. It exposes them, at scale, with confidence.

    If your organization is planning any AI initiative in 2027, the sequence matters: architecture first, automation second. The organizations that get this right will not be the ones with the biggest models. They’ll be the ones whose data was clean enough for the model to be useful.

    You have roughly eleven weeks before 2027.

    That’s enough time to run a diagnostic, understand what it’s costing you, and decide what to fix first.

    It is not enough time to rebuild from scratch.

    Start with the number.

    Frequently Asked Questions

    What is a data friction score?

    It’s a single number out of 72 that quantifies how much hidden cost your data architecture is generating. The Data Friction Matrix calculates it from 12 weighted indicators across four pillars: Data & Architecture, Strategic Process, Technical Execution, and Cognitive & Team. Higher scores mean more friction, not more data.

    What does a score above 19 mean?

    At 19 or above, your organization is actively paying the Data Friction Tax — the cumulative cost of manual reconciliation, unreliable reporting, and stalled initiatives. It does not mean your data is broken. It means the friction is structural enough to erode margins and undermine any AI project built on top of it.

    Do I need an audit if I already have a data team?

    Yes, and for a specific reason: your data team is usually too close to the problem to see it. They know where the workarounds are because they built them. An external audit maps friction across processes and teams, costs it in euros and hours, and assigns an owner per fix — something an internal team rarely has the mandate to do.


    #DataFriction #DataStrategy #DataGovernance #EnterpriseArchitecture#TechDebt

    Data Friction signs
    data vanity metrics

    Data Friction Myth Busters 4 — The “Vanity Metrics” Trap

    Data Friction Myth Busters 4 — The Vanity Metrics Trap

    Why Your Green Dashboard Hides a Red P&L

    If your data platform launch was a “massive success” because you built 50 new data assets and signed up 200 users, you might be running out of cash while celebrating fake growth. Tracking data volume instead of active adoption—or total sign-ups instead of retention—is the classic trap of data vanity metrics. They create an invisible layer of friction that masks real risks and makes you feel productive while the business stalls.

    This article is part of the Data Friction Myth Busters series. Read the full series →

    Here's where we're headed—so you can jump straight to what matters most to you
      Add a header to begin generating the table of contents
      Scroll to Top

      The Illusion of Output Metrics

      In modern data teams, there is immense pressure to prove value. So, teams start measuring what is easy to count: pipelines deployed, tables created, dashboards published, or total registered users on a new platform.

      These are output metrics. They measure activity, not outcomes.

      A dashboard that nobody looks at is not an asset; it is technical debt. A user who signs up but never returns is not a customer; it is a leaky bucket. When you focus on data vanity metrics, you are essentially driving a car by looking at how fast the wheels are spinning, rather than where the car is actually going.

      The Three Systemic Failures

      Measuring the wrong thing doesn’t just give you a false sense of security; it triggers three specific failures across the organization.

      1. The Financial Waste (The SME Threat)
      Startups and SMEs are the most vulnerable to the data vanity metrics trap. Celebrating top-of-funnel numbers (like total registrations) while ignoring active retention is a fast track to bankruptcy. If your charts show a beautiful cumulative growth curve, but your product has a leaky bucket problem, you will run out of runway. Data should highlight your risks, not mask them.

      2. The Corporate Silo War (The Enterprise Threat)
      In large organizations, vanity metrics become weapons. Different departments use them to defend their turf. Marketing celebrates “leads generated,” while Sales complains about “poor lead quality.” Data Engineering celebrates “pipelines deployed,” while the business complains about “slow reports.” Everyone has a green KPI dashboard, but the company’s P&L is in the red.

      3. The Governance Breakdown (The Tech Debt Threat)
      True data governance is not about creating a massive, 6-month bureaucratic council to catalogue every junk table. It’s about locking down the data quality of the top 5 KPIs that actually move the business. Over-reporting on 200 unvalidated metrics creates massive cognitive friction for stakeholders and forces data engineers to maintain useless assets.

      How Data Friction Hides in Plain Sight

      This is exactly the kind of invisible friction the Data Friction Matrix is designed to measure.

      If you score high on the Matrix for “Cognitive & Team” (decision overload, process fatigue) or “Strategic Process” (misaligned priorities), it is almost certain that your organization is suffering from data vanity metrics. You are spending engineering hours producing data, and spending leadership hours arguing about whose data is right, instead of driving the business forward.

      The Point

      High-performing organizations govern for value, not volume.

      If a metric doesn’t trigger a specific operational action or impact a financial line item, it shouldn’t be on your dashboard. It shouldn’t be in your weekly meeting. And it certainly shouldn’t be a KPI for your data team.

      Before you build another dashboard, or deploy another pipeline, ask a simple question: “What decision will this change, and how will it impact the P&L?” If there is no clear answer, you are just adding noise.

      Stop measuring how much data you can stack up. Start measuring how much friction you can strip out.

      Frequently Asked Questions

      What are data vanity metrics?
      Data vanity metrics are measurements that look impressive on a dashboard but do not correlate with business value or strategic outcomes. Examples include total registered users (instead of active retention), number of dashboards created, or total data volume stored, rather than the business actions those metrics drive.

      Why are data vanity metrics dangerous for enterprises?
      In large enterprises, vanity metrics create silos and internal friction. Departments use them to prove their own value while ignoring the overall company P&L. This leads to wasted budgets on redundant tools and creates a culture where teams argue over whose dashboard is correct instead of aligning on a single source of truth.

      How do I know if my dashboard has vanity metrics?
      The simplest test is the “Action Test.” Look at a metric on your dashboard and ask: “What specific action will we take this week if this number goes up or down?” If the answer is “nothing” or “we will just monitor it,” it is likely a vanity metric.

      📌   This article is part of a series on Data Friction and the Growth Gap — a series breaking down the biggest tech myths enterprise ecosystems fall for every year.

      Read the previous article: Data Friction Myth Busters 3 — The “More is Better” Trap

      #DataFrictionMythBusters #DataFriction #DataStrategy #BusinessOperations #TechDebt

      Data Friction signs
      dashboard sprawl

      Data Friction Myth Busters 3 — The “More is Better” Trap

      Data Friction Myth Busters 3 — The "More is Better" Trap

      Why Dashboard Sprawl Won't Make You Data-Driven

      There is a familiar moment in many growing companies: a leader looks at the business, feels that something is being missed, and concludes that the answer is more data. More metrics. More dashboards. More integrations. More automation. It feels decisive, modern, and ambitious.

      But this is often where teams enter the “more is better” trap. The problem is not that they lack data. The problem is that they are adding data without adding clarity.

      In this third edition of the Data Friction Myth Busters series, the myth is simple: if we track everything, we will understand everything. The reality is less comfortable: tracking everything usually makes it harder to see what matters.

      This article is part of the Data Friction Myth Busters series. Read the full series →

      Here's where we're headed—so you can jump straight to what matters most to you
        Add a header to begin generating the table of contents
        Scroll to Top

        What Is the “More Is Better” Trap?

        The more is better trap is the assumption that data value grows in direct proportion to data volume. In practice, it does not. A company can double its number of dashboards and still make slower, worse decisions. It can connect every tool in its stack and still lack a reliable answer to basic operational questions.

        The trap appears when leaders confuse:

        • Collection with understanding — having more data is not the same as knowing what it means.

        • Activity with progress — more charts and automations create the appearance of maturity without the substance.

        • Speed with control — fast integrations can move data quickly in the wrong direction.

        When this happens, data stops being an asset and becomes another source of friction.

        How the Trap Shows Up in Enterprises

        In larger organizations, the more is better trap often takes the form of dashboard sprawl. Different teams build their own reports, define metrics differently, and present competing versions of the same story. Soon, meetings require pre-meetings just to agree on which number is correct.

        The dashboard becomes less of a decision tool and more of a territorial one. Each department protects its own view of the business. Instead of creating a shared language, the data stack creates more dialects.

        The cost is not only time. It is trust. When leaders cannot agree on the definition of revenue, churn, pipeline, or utilization, data loses its authority. People fall back on opinion, hierarchy, or the loudest voice in the room.

        How the Trap Shows Up in Startups

        In startups and SMEs, the same more is better trap looks different but produces similar damage. Speed is critical, so teams stitch together tools using no-code automation platforms. A form triggers an update. An update triggers a Slack message. A Slack message triggers a CRM change. A CRM change triggers an invoice.

        At first, it feels like operational leverage. But without ownership, monitoring, and documentation, these automations become hidden production systems. No one knows exactly how they work until a schema changes, an API limit is hit, or a field is renamed. Then the pipeline breaks silently, and the team keeps making decisions on stale or incomplete data.

        The startup is not suffering from a lack of tools. It is suffering from a lack of architecture.

        The Hidden Cost of Dashboard Sprawl

        The more is better trap is expensive in ways that are easy to underestimate. The obvious cost is tooling and maintenance. The deeper cost is decision latency.

        Every extra dashboard creates another place to check, another metric to reconcile, and another opportunity for disagreement. Every unmonitored integration creates another failure point. Every unnecessary data layer increases the distance between a question and a confident answer.

        There is also a psychological cost. More data can create false confidence. Leaders feel informed because they have access to many reports, even if those reports contradict each other or are never used. A dashboard nobody trusts is worse than no dashboard at all, because it creates the illusion of control.

        Why Smart Leaders Fall for It

        This trap is not a sign of incompetence. It is usually a sign of pressure. Leaders worry about missing a signal, falling behind competitors, or making a decision without enough evidence. Adding a new dashboard feels safer than removing an old one. Adding an automation feels faster than fixing a process.

        Vendors reinforce this instinct. Every tool promises more visibility, more connection, more insight. But visibility is not the same as clarity. Connection is not the same as integrity.

        The trap feels productive because adding is visible. Subtracting feels risky. But in data, subtraction is often the more mature move.

        How to Escape the “More Is Better” Trap

        Escaping the trap starts with a shift in identity. A mature data function is not a collection department. It is an editorial function. Its job is to decide what matters, what does not, and what should be removed.

        A useful test is to ask whether each dashboard or report helps answer three questions:

        1. What happened?
        2. Why does it matter to the P&L?
        3. What action should we take before Friday?

        If a report cannot support those questions, it is probably decoration, not decision infrastructure.

        From there, teams can take practical steps:

        • Audit before adding. Map existing dashboards, reports, and automations to real decisions. Remove what no one uses.

        • Define a single source of truth. Agree on core metric definitions and ownership.

        • Treat automations as production systems. Assign owners, monitor failures, document logic, and use schema contracts where possible.

        • Reduce before you optimize. Sometimes the fastest way to improve reporting is to delete half of it.

        • Measure friction, not volume. Track how long it takes to get a trusted answer, not how many charts exist.

        The Point

        True data maturity is not measured by the number of tools connected or charts displayed. It is measured by how quickly the organization can move from question to trusted action.

        The “more is better” trap ends when leaders stop asking, “What else can we track?” and start asking, “What can we stop tracking without losing the decisions that matter?”

        If your team struggles with manual reporting, conflicting metrics, or no single source of truth, the problem may not be a missing dashboard. It may be dashboard sprawl.

        Debunking the myth is easy. Auditing your own infrastructure is harder. That is why the Data Friction Matrix is available: a short audit designed to help you identify where your data architecture is creating friction and what to fix first.

        Stop adding layers. Start auditing your bottlenecks. The path to being data-driven is not more data. It is less noise.

        📌   This article is part of a series on Data Friction Myth Busters— a new series breaking down the biggest tech myths enterprise ecosystems fall for every year.

        Previous: Data Friction Myth Busters 2 — The Real-Time Obsession

        Coming next: Data Friction Myth Busters 4 — The Vanity Metrics Trap

        #DataFrictionMythBusters #DataFriction #DataGovernance #DataStrategy #TechDebt #DataQuality

        more is better trap
        Real-time data architecture pipeline vs batch processing illustration

        Data Friction Myth Busters 2 — The Real-Time Data Obsession

        Data Friction Myth Busters 2 — The Real-Time Obsession

        Why a Real-Time Data Architecture is Burning Your Budget

        There is a moment in many leadership meetings when someone says, “We need real-time data.” It sounds bold, modern, and decisive. The implication is clear: if we can see what is happening right now, we can act faster, smarter, and better than the competition.

        But for most organizations, this is not a strategy. It is the real-time data obsession—a costly distraction that confuses speed of data with speed of decision.

        In this second edition of the Data Friction Myth Busters series, we are tackling the myth that real-time architecture is the ultimate sign of data maturity. For the vast majority of companies, it is not. It is technical debt in disguise.

        This article is part of the Data Friction Myth Busters series. Read the full series →

        Here's where we're headed—so you can jump straight to what matters most to you
          Add a header to begin generating the table of contents
          Scroll to Top

          What Is the Real-Time Data Obsession?

          Is the belief that business value increases as data latency approaches zero. It assumes that if a dashboard updates every second, the organization becomes more agile.

          In reality, agility is not determined by how fast data moves through a pipeline. It is determined by how fast an organization can legally, operationally, and culturally act on that data.

          Most strategic business decisions—pricing changes, hiring plans, budget reallocations, product pivots—are made on a weekly or monthly cadence. Building a complex streaming pipeline so a management team can glance at a static KPI chart over morning coffee creates significant infrastructure cost and near-zero business value.

          The real-time data obsession is not about meeting a real need. It is about appearing advanced.

          Why Real-Time Sounds So Attractive

          Real-time data is often sold as the pinnacle of data maturity. Vendors showcase live dashboards, event streams, and sub-second updates. The message is subtle but persistent: if you are not real-time, you are behind.

          This narrative is reinforced by stories from companies where real-time genuinely matters—high-frequency trading, fraud detection, live logistics routing, dynamic pricing engines. These are valid use cases. But they are the exception, not the rule.

          For most businesses, the decision-making loop is not measured in milliseconds. It is measured in days, weeks, or quarters. The real-time data obsession becomes dangerous when leaders apply a solution designed for microseconds to problems that operate on a completely different clock.

          The Startup Trap: Engineering Vanity on a Runway Budget

          For fast-growing SMEs and startups, the real-time data obsession is particularly expensive. Lean teams rarely have the cash or headcount to design, build, and babysit a streaming architecture.

          Streaming systems require specialized skills, continuous monitoring, schema management, and robust error handling. They are production systems, not side projects. Forcing a small engineering team to maintain real-time infrastructure means diverting scarce resources away from the features and customer feedback loops that actually drive market fit.

          The result is predictable: higher cloud bills, slower product velocity, and a data platform that looks impressive in a demo but creates no competitive advantage. The real-time data obsession burns runway on engineering vanity.

          The Enterprise Trap: Governance Blind Spots and Panic Dashboards

          In large enterprises, the problem is different but equally costly. Internal data quality and reconciliation cycles take time for a reason. They exist to ensure that numbers are accurate, definitions are aligned, and reporting is auditable.

          Forcing data to move in seconds can bypass these governance controls. The result is a stream of operational noise presented to stakeholders who are not trained to interpret it. A temporary daily anomaly can trigger panic, micro-management, and knee-jerk reactions that cost far more than the delay ever would.

          The real-time data obsession in an enterprise often produces a paradox: more data, faster, with less trust. Executives see fluctuations in real time but lack the context to know whether they matter. Decision quality goes down, not up.

          The Hidden Costs of Real-Time Architecture

          The real-time data obsession carries costs that are easy to underestimate:

          • Infrastructure spend. Streaming platforms, compute, storage, and monitoring tools add up quickly.

          • Engineering overhead. Real-time systems require ongoing maintenance, on-call support, and specialized expertise.

          • Governance risk. Faster data can outpace data quality checks and access controls.

          • Organizational confusion. Real-time dashboards create the expectation of real-time action, which most operating models cannot support.

          • Opportunity cost. Every dollar and hour spent on unnecessary streaming is a dollar and hour not spent on reliability, governance, or decision support.

          None of these costs appear on the initial project plan. They accumulate quietly, long after the launch demo.

          The Question Every Leader Should Ask

          Before approving a real-time project, ask a simple question:

          “If this chart updates five minutes from now instead of tomorrow morning, what specific action will we take at 10:05 AM?”

          If the honest answer is “we will review it in our weekly alignment meeting,” then you do not need a real-time stream. You need a reliable, well-governed batch process.

          This question cuts through the real-time data obsession because it connects data latency to decision latency. If the decision cannot be made faster, the data does not need to arrive faster.

          Matching Data Refresh Rate to Decision Speed

          The mature approach is not to reject real-time data entirely. It is to match your data refresh rate to the actual decision-making speed of your business.

          Some processes genuinely require real-time: fraud detection, dynamic routing, safety monitoring, live customer interactions. Others require near-real-time: inventory alerts, support escalations, operational anomalies. Many require only daily or weekly refresh: financial reporting, strategic planning, executive dashboards.

          The real-time data obsession fails because it applies one speed to every problem. A well-architected data platform uses different latencies for different needs. It is intentional, not uniform.

          How to Escape the Real-Time Data Obsession

          Escaping the real-time data obsession starts with reframing the conversation. Instead of asking “How fast can we move data?”, ask “How fast can we act on it—and what is the cost of waiting?”

          From there, teams can take practical steps:

          • Map decisions to latency requirements. Identify which decisions actually change based on fresher data.

          • Challenge real-time requests. Ask what action will be taken differently and when.

          • Invest in reliability before speed. A trusted daily report beats an unreliable live stream.

          • Segment your architecture. Use batch for strategic reporting and streaming only where operational value is proven.

          • Measure decision latency, not data latency. The metric that matters is time from question to trusted action.

          The Point

          True data maturity is not measured by how fast your pipelines run. It is measured by how well your data architecture supports the decisions your organization actually makes.

          The real-time data obsession ends when leaders stop equating speed with sophistication and start equating fit with maturity. A batch process that delivers trusted numbers on Monday morning can be more valuable than a streaming pipeline that delivers noise every second.

          Dispelling myths is only the first step. The second is auditing your own infrastructure. The Data Friction Matrix is now available — a 12-point weighted self-diagnostic for scaling companies. You’ll get your friction score in under three minutes.

          Match your data refresh rate to your decision-making speed. Anything else is just an expensive tech flex.

          📌   This article is part of a series on Data Friction Myth Busters.

          Read the first article: Data Friction Myth Busters 1 — The AI Silver Bullet

          Coming next: Data Friction Myth Busters 3 — The “More is Better” Trap

          #DataFrictionMythBusters #DataFriction #DataStrategy #BusinessOperations #DataArchitecture

          real-time data obsession
          Isometric vector illustration of a robot arm pouring fragmented data into a briefcase, symbolizing the AI data readiness problem.

          Data Friction Myth Busters 1 – AI Data Readiness: Why GenAI Fails Without It (The Silver Bullet Myth)

          AI Data Readiness: Why GenAI Fails Without It (The Silver Bullet Myth)

          If your underlying data model is full of friction, silos, and conflicting definitions, AI will just hallucinate incorrect business metrics with absolute confidence. Garbage in, automated garbage out.

          The short answer: Generative AI cannot clean a broken data model; it can only accelerate whatever is already there. If your teams argue over basic definitions like “Churn” or “Revenue,” a natural language interface won’t create alignment. Before you deploy a GenAI agent, you need to ensure your AI data readiness is solid by establishing a single, coherent data language.

          This article is part of the Data Friction Myth Busters series. Read the full series →

          Here's where we're headed—so you can jump straight to what matters most to you
            Add a header to begin generating the table of contents
            Scroll to Top

            The Promise (and the Trap) of GenAI

            Right now, IT consulting firms and enterprise leaders are under immense pressure to deploy AI everywhere.

            The promise is tempting: skip the painful, boring, expensive work of data alignment, and let Large Language Models (LLMs) do the heavy lifting. Just connect a GenAI agent to the database, hand the commercial team a chat box, and watch productivity soar.

            It makes for a great boardroom presentation. It makes for a terrible operational reality.

            In enterprise operations, AI doesn’t eliminate data friction. It amplifies it. Here are the three ways deploying AI on a fractured data ecosystem will break your business.

            1. The Context Collapse

            Your finance team and your product team define “Churn” differently. Your sales team calculates “Pipeline” using a completely different set of rules than the CRM.

            When a human looks at a dashboard, they might notice the discrepancy. They might ask a question. They might apply judgment.

            An AI agent doesn’t. It picks one definition at random and presents it with absolute confidence. You haven’t empowered business users; you’ve decentralized confusion. You’ve built a machine that generates very articulate, highly formatted wrong answers.

            2. The Hidden Cloud Bill

            AI is computationally expensive. Querying a clean, well-modeled data warehouse is expensive enough. Querying messy, unindexed, fragmented data to feed a context window is exponentially worse.

            You’re paying a premium in compute costs to retrieve, stitch together, and hallucinate over bad data. That’s not an AI strategy. That’s a very expensive way to get wrong answers faster. For example, Gartner estimates that poor data quality already costs organizations an average of $12.9 million annually—adding AI to that mix only inflates the bill.

            3. The Governance Nightmare

            An SVP queries the AI for a forecast. The AI pulls data from an unvalidated staging table, blends it with a legacy table, and generates a beautifully formatted insight. The SVP makes a million-euro decision based on it.

            Who is responsible when it’s wrong? The AI vendor? The data engineer who built the pipeline? The SVP?

            When you put a natural language interface on top of an ungoverned data layer, you remove the friction that used to warn people they were looking at bad data.

            The Truth: AI Data Readiness is the Only Prerequisite

            AI is an incredible velocity accelerator. But it is not a data cleaner.

            Before you give your business stakeholders a natural language interface, you need to ensure the data layer speaks a single, coherent language. If your engineering team and your business leads aren’t aligned on the metrics, no amount of prompt engineering will save the project margin.

            True AI data readiness isn’t about the model you buy; it’s about the friction you remove before you turn it on.

            What AI Data Readiness Actually Means

            “AI data readiness” is not a model-selection exercise. It is the work of making your data layer coherent enough that a model can be useful on top of it. In practice, that means five things:

            • Agreed definitions. Core metrics have one owner and one definition, documented and enforced.

            • A single source of truth. Business users and AI agents query governed layers, not staging tables.

            • Clear lineage. You can trace any number—or any AI answer—back to its origin.

            • Defined ownership. Someone is accountable for each dataset and each metric.

            • Access and governance controls. Permissions and validation are built in, not bolted on.

            Without these five, “AI data readiness” is a slide in a deck. With them, it is the difference between an AI assistant that answers questions and one that generates risk.

            How to Audit Your AI Readiness

            The Data Friction Matrix is designed to measure exactly this. If you score high on Data & Architecture (specifically #1 Silos and #3 No single source of truth), your AI project is at severe risk.

            But before you even open the Matrix, you can start the audit yourself with four practical checks:

            1. Audit before you automate. List the metrics where departments already disagree. Those are your AI’s future hallucinations.
            2. Fix the highest-impact metrics first. You don’t need perfect data everywhere—just a governed core that covers the decisions that matter.
            3. Establish a semantic layer. Give both humans and models one place to resolve “what does this metric actually mean?”
            4. Constrain the AI’s surface area. Start with a narrow, validated domain rather than open access to the whole warehouse.

            You don’t need to buy a bigger model. You need to fix the foundation.

            Start by asking your team a simple question:

            “If we asked an AI to tell us our most profitable customer segment right now, would every department agree with the answer?”

            If the answer is no, turn off the AI pilot. Turn on a data audit.

            The Point

            Don’t buy a silver bullet. Buy a foundation.

            The organizations that win with AI in 2027 will not be the ones that deployed the most agents in 2026. They will be the ones whose data was clean enough for the agents to actually work.

            Stop automating chaos. Fix the model first.

            Frequently Asked Questions

            Can GenAI clean our messy data?
            No. Generative AI models are designed to predict and generate based on patterns, not to perform rigorous data cleaning or master data management. If you feed it conflicting data, it will either hallucinate a blended answer or confidently pick one source without knowing it’s wrong. Data cleaning must happen before the AI layer.

            What is “Context Collapse” in AI?
            Context Collapse happens when an AI model is given conflicting definitions of a metric (e.g., Finance and Sales both define “Revenue” differently) and arbitrarily selects one to answer a prompt. Instead of resolving the business disagreement, the AI hides it behind a confident-sounding response.

            How do I know if my data is ready for AI?
            Your data is ready for AI when it has a single source of truth, clear metric definitions agreed upon by all departments, and governed access. A good starting point is running a diagnostic like the Data Friction Matrix. If you score high on data silos and lack of a single source of truth, you need to fix the architecture before deploying AI.


            📌 This article is the first of a series on Data Friction Myth Busters — a new series breaking down the biggest tech myths enterprise ecosystems fall for every year.

            Coming next: Data Friction Myth Busters 2 — The Real-Time Obsession

            #DataFrictionMythBusters #DataFriction #DataGovernance #GenerativeAI

            AI Data Readiness
            Shitty Data

            The Root Cause of Data Friction (and the Tax You’re Paying): Shitty Data

            The Root Cause of Data Friction

            And the Tax You're Paying: Shitty Data

            Let’s call it what it is: shitty data.

            Not “data quality issues.” Not “data inconsistencies.” Not “legacy system challenges.”

            Shitty data.

            The kind that makes your CFO question the numbers. The kind that makes your engineers waste entire sprints fixing pipelines instead of building features. The kind that makes your CEO decide based on gut feeling because the dashboards can’t be trusted.

            We’ve been dancing around this problem for years. We’ve given it polite names. We’ve blamed the tools, the team, the budget.

            But the root cause is simpler — and more uncomfortable — than most companies want to admit.

            Here's where we're headed — so you can jump straight to what matters most to you
              Add a header to begin generating the table of contents
              Scroll to Top

              What "Shitty Data" Actually Means

              Shitty data isn’t just “wrong” data. It’s data that:

              • Lives in silos — so no one has the full picture.

              • Has no single definition — so “Active User” means three different things to three different teams.

              • Is manually verified — because nobody trusts the automated numbers.

              • Arrives too late — so decisions are made on last month’s reality, not today’s.

              • Breaks under scale — so what worked at 20 people collapses at 200.

              It’s not a technical problem. It’s an alignment problem. And it’s the raw material of Data Friction.

              How Shitty Data Becomes Data Friction

              When your data is shitty, every layer above it suffers.

              Strategy gets built on assumptions that don’t match reality. Execution gets built on requirements that were never properly translated. Decision-making slows down because no one agrees on the facts. And team morale drops because everyone is busy fixing symptoms, not solving causes.

              Shitty data doesn’t stay in the database. It leaks into every meeting, every sprint, every strategy session.

              And that leak has a cost — one we’ve explored before under a different name.

              The Friction Tax, Revisited

              If you’ve read the previous article in this series, you know about the Friction Tax: the hidden cost of misalignment between business strategy and technical execution.

              Shitty data is what makes that tax so expensive — and so persistent.

              The visible cost is easy to estimate. Add up the hours your leadership team spends reconciling numbers, debating definitions, or manually verifying dashboards. Multiply by their loaded salary. The number is rarely small.

              The invisible cost is harder to measure — but far more damaging. Features that ship late. Campaigns optimized against the wrong metrics. Customers who leave before anyone notices the warning signs.

              If you want the full breakdown of how to calculate it, I covered it in detail here.

              Why Companies Tolerate It

              If shitty data is so expensive, why does it persist?

              Because it hides.

              It looks like “normal operational noise.” It looks like “that’s just how it is.” It looks like “we’ll fix it when we scale.”

              But every month you wait, the tax compounds. The pipelines get more tangled. The definitions drift further apart. The trust erodes.

              And the worst part? The longer you tolerate it, the more expensive it becomes to fix.

              What to Look For (Without a Full Audit)

              You don’t need a three-week engagement to know if shitty data is a problem in your company. Start with these three signals:

              1. The reconciliation ritual.
              If your team spends the first days of every month manually verifying numbers, that’s not diligence. That’s a symptom.

              2. The definition drift.
              Ask three different teams what “active user” means. If you get three different answers, your data isn’t aligned. Your company isn’t either.

              3. The late signal.
              If your customer success team finds out about churn from a cancellation email rather than from usage data, your data is arriving too late to be useful.

              These aren’t technical failures. They’re structural ones. And they’re fixable — but only once they’re named.

              The Bottom Line

              Shitty data isn’t a technical problem. It’s a business problem.

              It’s the root cause of Data Friction. It’s what makes the Friction Tax so expensive. And it’s the reason so many scale-ups stall just when they should be accelerating.

              The first step is naming it. The second is measuring it. The third is deciding what to do about it.

              In the next article in this series, I’ll share the framework I use to map where the data breaks — and why.

              #DataFriction #GrowthGap #ScaleUp #BusinessStrategy #TechExecution #MMTrufinStratEdge

              Data Friction signs
              Glowing calculator beside a percent symbol, cracked pipe, and falling coins

              The Friction Tax: : What Data Misalignment Is Really Costing You

              The Friction Tax: What Data Misalignment Is Really Costing You

              Data Friction isn’t just an engineering inconvenience. It’s a cash-flow drain.

              Last week, we looked at how to spot the operational red flags of a data translation problem — like engineering sprints being derailed by pipeline maintenance, or managers spending the first three days of every month double-checking numbers in Excel.

              But to get business leaders to prioritize data architecture, you have to move past technical symptoms.

              You have to talk about the Friction Tax.

              This article is part of The Data Friction Playbook. Read the full series →

              Here's where we're headed — so you can jump straight to what matters most to you
                Add a header to begin generating the table of contents
                Scroll to Top

                What Is the Friction Tax?

                The Friction Tax is the hidden cost of misalignment between your business strategy and your technical execution. It’s not a line item on your budget. It doesn’t appear in your quarterly reports. But it’s there — quietly draining your time, your margin, and your team’s capacity to execute.

                It shows up as:

                • Meetings that turn into debates about whose numbers are “correct.”

                • Engineers spending sprints fixing pipelines instead of building features.

                • Marketing teams optimizing campaigns against metrics that don’t reflect reality.

                It’s a tax you pay every day. And unlike other taxes, you can eliminate this one.

                A Simple Calculation: The Visible Cost

                Let’s do the math for an established business with around 100 employees.

                Assume you have 6 department heads or senior managers earning an average of €90,000/year.

                Because data is trapped in department silos or unaligned systems, each manager wastes just 3 hours a week hunting down numbers, arguing over mismatched KPIs, or manually rebuilding reports for alignment.

                That’s 18 hours of leadership capacity completely wasted every single week.

                Across a year, that adds up to nearly 900 hours of lost decision-making time.

                In cold, hard cash? That’s over €40,000 paid annually just to tolerate bad data friction.

                And that’s just the visible cost.

                The Invisible Costs Are Far Worse

                The Friction Tax doesn’t stop at wasted hours. It compounds across your entire operation.

                Delayed product features.
                Your developers are playing data mechanics — fixing pipelines, reconciling definitions, untangling mismatched metrics — instead of building the features your customers are asking for. Every sprint spent on friction is a sprint not spent on growth.

                Misallocated marketing spend.
                Your campaigns are optimized against lagging attribution metrics. You think you’re investing in what works. In reality, you’re investing in what you can measure — which may not be the same thing. The result? Budget spent on channels that look good in reports but don’t move the needle.

                Customer churn you see too late.
                Your customer success team only sees usage drops after the client has already decided to leave. By the time the data reaches them, the relationship is already broken. That’s not a customer success problem. It’s a data latency problem.

                The cost of this misalignment is well-documented. According to a 2024 IDC survey, IT-business misalignment is the third biggest organizational challenge for CIOs. A 2025 Gartner survey also found that 48% of digital initiatives fail to meet their intended business outcome targets. You can read more about the data behind this problem in this article on the hidden cost of IT-business misalignment.

                How to Calculate Your Own Friction Tax

                You don’t need a full audit to estimate what data friction is costing you. Start with these three questions:

                1. How many hours per week does your leadership team spend reconciling, debating, or manually verifying numbers?
                2. What percentage of your engineering sprints go to fixing data pipelines rather than building new features?
                3. How much marketing budget is allocated based on metrics you don’t fully trust?

                Multiply those numbers by your average salary or budget allocation. That’s your Friction Tax. And if it’s more than €0, it’s already a priority.

                What to Do About It

                Data friction is rarely a technical limitation. It’s a financial leak — and it’s already on your P&L.

                Fixing it doesn’t start with new tools or new hires. It starts with a clear diagnosis.

                That’s exactly what the 72-Hour Data Friction Audit was designed to do. In three days, I step into your operations, look under the hood of your business goals and your technical architecture, and hand you a brutal, 2-page assessment:

                1. The Short-Circuit: The exact spot where your business strategy and your code architecture are misaligned.
                2. The Fix: A clear, actionable roadmap to realign both teams, protect your margins, and speed up deployment.

                The Bottom Line

                If a data issue doesn’t impact the bottom line, it isn’t a priority.

                But if it does — and it almost always does — then the Friction Tax is already eating into your margins, your speed, and your team’s capacity to execute.

                The first step is naming it. The second is measuring it. The third is fixing it.

                📌   This article is part of The Data Friction Playbook — a 5-part series on the hidden costs of misalignment between business strategy and technical execution.

                Previous: How to Spot Data Friction → · Next: How to Close the Gap →

                🧭 You've seen what friction costs. Now measure your own.

                Take the free 12-Point Data Friction Matrix assessment. Score your friction in five minutes, get your personal Friction Tax estimate, and see where you stand.

                Get Your Friction Score →
                The friction tax
                AI without strategy

                AI Without Strategy: The Hidden Cost of Automating Chaos

                The Real Cost of AI Without Strategy

                The Hidden Cost of Automating Chaos

                Last week, someone sent me a comment that stopped me in my tracks:

                “Do NOT touch AI for a single thing unless you have thought everything out and laid your vision of what you want or think. I will not warn again — if you don’t listen, your IQ will be less than a monkey in 3-4 months.”

                It’s a brutal statement. But it contains a truth that most companies scaling with AI are ignoring.

                The problem isn’t AI. The problem is AI without strategy.

                Why AI Without Strategy Fails?

                Here’s why:

                Here's where we're headed — so you can jump straight to what matters most to you
                  Add a header to begin generating the table of contents
                  Scroll to Top

                  1. AI doesn't give you vision. It gives you speed.

                  If you don’t know where you’re going, AI will help you get there faster. But “there” might be nowhere.

                  I’ve seen teams deploy AI-powered tools to automate processes that were already broken. They didn’t fix the process. They just made the mess faster. And more expensive.

                  AI amplifies what you already have. If you have clarity, it multiplies it. If you have chaos, it multiplies that too. AI without strategy just makes your existing problems bigger and faster.

                  2. The translation gap also applies to AI

                  If you can’t explain what you want in plain language, AI can’t execute it. And if your business team and your technical team don’t share the same definitions, AI will just add another layer of confusion.

                  I’ve spent 20 years bridging the gap between business strategy and technical execution. The same rule applies to AI: if the intent isn’t clear, the output won’t be either.

                  3. The real cognitive risk isn't stupidity. It's dependency.

                  AI won’t make you less intelligent. But it will make you more dependent if you don’t understand the “why” behind what you’re doing.

                  If you outsource your thinking to AI, you lose the ability to challenge it. And if you can’t challenge it, you can’t lead it. You just follow.

                  That’s the real danger. Not that AI makes you dumber. That it makes you slower to notice when it’s wrong.


                  The Transparency Layer

                  This isn’t just an internal operations problem; it’s a systemic one. Look at the broader conversation happening right now about AI alignment — the idea that AI systems should behave in ways that are safe, predictable, and aligned with human values. The Open Alignment Initiative, led by Hugging Face and Thomas Wolf, is pushing for more transparency in how alignment is evaluated. Their argument is simple: alignment shouldn’t be solved behind the closed doors of a handful of frontier labs.

                  I think the same principle applies to any company adopting AI. If you don’t understand how your AI tools make decisions, you can’t align them with your strategy. You’re not in control — you’re just renting someone else’s logic.

                  Transparency isn’t just a technical requirement. It’s a strategic one.

                  Why this matters more for scale-ups

                  In a scale-up, every decision is amplified. A small misalignment in strategy becomes a big misalignment in execution. AI doesn’t fix that — it accelerates it.

                  I’ve seen scale-ups deploy AI tools to “move faster” without fixing the underlying process. Six months later, they’re moving faster in the wrong direction, with higher costs and more technical debt.

                  The companies that win with AI aren’t the ones that adopt it first. They’re the ones that adopt it last — after they’ve clarified the problem, the process, and the metric.

                  How to Fix It: The Strategy-First AI Checklist

                  If you’re about to introduce AI into a process, ask yourself three questions first:

                  1. Do we know exactly what problem we’re solving? If the answer is vague, stop here.
                  2. Can we explain it in plain language to someone outside the business? If not, AI won’t understand it either.
                  3. How will we measure success? If you can’t define it, AI can’t optimize it.

                  If you can answer all three, you’re ready. If not, fix the strategy first. Then bring in AI.

                  The bottom line:

                  AI is an accelerator, not a GPS. You set the direction. It provides the speed.

                  If you don’t have a clear strategy — if you haven’t thought through what you want, why you want it, and how you’ll measure success — AI will just make your existing problems bigger and faster.

                  But if you do have clarity, AI becomes a multiplier. It executes what you’ve already designed. It amplifies what you’ve already built.

                  The question isn’t “Should we use AI?”

                  The question is: “Do we know what we’re trying to achieve?”

                  If the answer is no, fix that first. Then bring in AI.

                  the real cost of ai without strategy
                  digital transformation business strategy

                  Digital Transformation: Unleashing Business Strategy’s Inner Geek

                  Digital Transformation: Unleashing Business Strategy's Inner Geek

                  In a world where business meets the digital galaxy, it’s no longer enough to simply dip your toe into the tech waters. No, my friend, you need to dive headfirst into the deep end of digital transformation to unlock a world of supercharged business strategies. Brace yourself for a cosmic adventure as we explore the geeky wonders that technology brings to the table.

                  The Power of Digital Transformation: Embrace Your Inner Geek

                  Digital transformation is like the Batmobile of business strategies. It swoops in, integrates digital technologies into every nook and cranny of your organization, and leaves your competitors scratching their heads in bewilderment. But what’s so amazing about it? Buckle up, my tech-loving friend, and prepare to have your mind blown.

                  1. Enhancing Customer Experience: It’s Superhero-level Awesome!

                  Picture this: your customers are treated to a personalized experience that feels like it was created just for them. Digital transformation empowers you to harness the incredible power of data analytics and automation. You’ll have the ability to anticipate their every need, deliver tailored solutions, and win their undying loyalty faster than a speeding bullet.

                  2. Driving Innovation and Agility: Unleash Your Inner Tony Stark!

                  Digital transformation is your Iron Man suit of business strategies. It unleashes your innovation superpowers and propels you to new heights. Embrace cutting-edge technologies like AI, blockchain, and the Internet of Things (IoT). They’ll catapult you ahead of the competition, leaving them with the firepower of a water gun at a laser tag tournament.

                  3. Optimizing Operations: Zap Those Pesky Villains of Inefficiency!

                  Holy operational efficiency, Batman! Digital technologies streamline your processes and vanquish the villains of inefficiency. With process automation, data-driven decision-making, and real-time analytics, you’ll be able to eliminate manual errors faster than The Flash on caffeine. Say goodbye to high costs and hello to streamlined operations that run smoother than a well-oiled R2-D2.

                  4. Unlocking Growth Opportunities: You’re the Business Jedi!

                  With digital transformation, you become the Yoda of growth opportunities. Expand your market reach, venture into new territories, and create innovative products and services. Harness the force of digital channels to target new customer segments and ride the wave of emerging trends. Your competitors won’t stand a chance as you soar to the top of the business galaxy.

                  Challenges and Strategies: Facing the Dark Side of Transformation

                  Even superheroes face challenges, but fear not! We’ve got the strategies to help you conquer the dark side of digital transformation and emerge victorious.

                  1. Leadership Commitment: Your Batman Signal of Success!

                  Gather your league of extraordinary leaders and make a commitment to digital transformation. Embrace your inner Batman and lead the charge with a clear vision and unwavering dedication. Inspire your team to join forces and create a culture of innovation and continuous learning. With great leadership comes great power!

                  2. Talent Development: Training Like a Superhero Academy!

                  Equip your team with the skills they need to conquer the digital world. Provide them with training programs, unleash their inner superheroes, and watch as they soar to new heights. Foster a culture of digital literacy and empower your workforce to embrace new technologies like Hulk-sized champions.

                  3. Agile Implementation: Super Speed Ahead!

                  Leap tall buildings in a single bound with an agile approach to digital transformation. Break down complex projects into bite-sized, manageable tasks. This allows you to test, learn, and adapt quickly, leaving your competitors in the dust like a turbo-charged DeLorean. Zoom past the competition and embrace change like a true superhero.

                  4. Collaboration and Partnerships: Avengers Assemble!

                  Don’t face the challenges alone—team up with external partners and unleash your inner Avengers. Join forces with tech vendors, startups, and industry experts. Their knowledge and resources will be your secret weapons. Together, you’ll conquer the digital universe and emerge as the ultimate business superhero.

                  Conclusion:

                  In the thrilling saga of digital transformation and business strategy, the geeks shall inherit the business world. Embrace your inner superhero, dive headfirst into the digital abyss, and watch your organization skyrocket to new heights of innovation, growth, and success. It’s time to unleash your geeky powers and conquer the galaxy of business!

                  digital quotient in marketing management

                  From IQ to DQ: How the Digital Quotient Is Redefining 21st Century Skills

                  From IQ to DQ: How the Digital Quotient Is Redefining 21st Century Skills

                  We’ve long measured intelligence through IQ and emotional intelligence through EQ. But in the 21st century, a new metric has emerged: the Digital Quotient (DQ). Coined by McKinsey and developed further by the DQ Institute, it aims to capture how individuals and organizations navigate the digital world.

                  During the XXth century we became accustomed with concepts like the Intelligence Quotient (IQ) and Emotional Quotient (EQ). The XXIst century, brings us a new concept: the Digital Quotient (DQ).

                  In the next paragraphs I’m going to discuss the whereabouts of these pathways of science.

                  A short history of brain testing

                  Firstly, the notion of intelligence testing appears around the 1880s for the first time, and continues to evolve ever since. The IQ test was originally designed by the French psychologist Alfred Binet in the early 1900s. He was a psychologist who wanted a more scientific way of evaluating students. He wanted to remove teacher bias and to identify at-risk students who would need help and extra attention to succeed in school. So Binet developed the first intelligence test. Through his research, he identified aspects of thinking that contribute to a student’s ability to do well at the things taught in school.

                  But how do you define and measure intelligence? The consensus seems to be that you can’t, but the IQ test is the closest we’ve ever come. The downside is that this general intelligence test which had innocent origins in French schools, was later used to support discrimination, eugenics, and genocide. 

                  Secondly, the history of Emotional Intelligence begins with Edward L. Thorndike in 1920 when he states that the personality reflects social intelligence. In his classic formulation: “By social intelligence is meant the ability to understand and manage men and women, boys and girls — to act wisely in human relations”  .

                  In short, while the IQ measures the cognitive intelligence — the ability to learn or skilled use of reason –, the EQ measures the emotional intelligence — the ability to use emotions and cognitive skills. Therefore, while the former employs the neocortex, while the latter relays on the limbic part of the brain.

                  And these two seemed to cover quite completely the brain activity in the final days of humanism.

                  Then a new day has dawn, and with it the dark clouds of posthumanism and AI takeover scenarios. Therefore we’re getting to the point.

                  What is the DQ?

                  The original concept is patented by McKinsey & Company in February of 2015 as a service mark for the Insurance & Financial Service industry.

                  The definition McKinsey give to DQ is:
                  “Digital Quotient®  measures an organisation’s performance across four key dimensions of digital maturity: Strategy, Culture, Organisation and Capabilities.“

                  The context: the World Economic Forum’s Transformation Maps — a constantly refreshed repository of knowledge about global issues, from climate change to the future of work.

                  The Digital Quotient Framework by McKinsey
                  The Digital Quotient Framework by McKinsey

                  Rapidly, the concept grows to encompass all digital education, endeavour put forward by the World Economic Forum via the newly founded (2020) DQ Institute.

                  The DQ Institute aims to set “the world’s first global standards for digital literacy, digital skills and digital readiness”, helping thus the global education system to prepare skilled workforce for the XXIst century.

                  DQ Institute Vision & Mission

                  It is common knowledge that keeping up with the latest technology is central to the success of every organization. Companies worldwide are recognizing that adapting only their business and technology strategies is not enough — they need to build digital talent as part of building a digital organization.

                  Here’s the problem in a nutshell: the job opportunities that are available today are 21st-century jobs. But the way most people perform these jobs is still stuck in the previous century. As is the way our society is training and educating people.

                  A short history of work

                  In the 19th century, with the Industrial Revolution, there was a massive movement of the population from rural to urban areas. The primary and secondary education system was created to train the workforce for the “new world” of manual and clerical work in cities.

                  In the 20th century, work was dominated by factory jobs. The education system that was built in the previous century was, with some modifications, still suited to training good factory workers and their managers. The early 60s bring about a new industrial revolution — the second. Which made Aldous Huxley, author of Brave New World, to argue that by the year 2000, technology and automation would make workers richer and more productive.

                  Then in the 60s, the rise of the machines. The 70s bring about the rise of the conglomerate — a multi-industry company. Conglomerates are always large and multinational. In the 80s the World Wide Web came online. The 90s see the rise of leading technologies such as cell phones, e-mails, PDAs.

                  The management focused on a series of tools to optimize this kind of work: operational efficiency, something called Taylorism, and eventually some management philosophies called Six Sigma. Management was mostly done face to face, while health insurance, a social safety net, and other benefits were bundled into inflexible labour contracts.

                  Then the millenium changed. And the decade of 2000 brings us skype, google, gmail, facebook, twitter, youtube, the iPhone. By 2010 Artificial Intelligence connects everything we do in business, governing and everyday life. The cloud enables workers to store and access huge amounts of data online.

                  So we’re seeing the rise of new work models such as freelancing and remote work. In the  companies, teams are learning to be more agile, to work with distributed and remote teams, and to scale up and down to adapt to ever-changing conditions.

                  Therefore, we get to the second part of this article:

                  How the DQ is measured

                  According to Business Insider — the Spanish edition — here are the 10 must-have digital skills to be able to work in the 21st century:

                  In conclusion

                  We have another Revolution going on — the Fourth Industrial Revolution, characterized by digital transformation in all industries and across societies.

                  Where do you think it’s leading us?