Founder & CEO, Valiotti Analytics | Better Data Analytics for Data-Driven Companies

Founders don’t lack data.
They lack confidence to make decisions with it.

I work as a Fractional Chief Data Officer for scaling tech companies, when analytics becomes business-critical — but hiring a full-time CDO is too early or too risky.

I’m the founder of Valiotti Analytics, a data consultancy trusted by 50+ teams globally. I work with companies across the US and Europe.

I have a PhD in data science, which helps when problems are complex — but my work is always grounded in business outcomes.

What I typically change:
• Data & analytics strategy aligned with business goals
• Clear metrics and decision frameworks
• Analytics org design (people, tools, processes)
• Systems that teams actually use, not just maintain

Before consulting, I was an in-house operator. I’ve lived through messy data, slow reports, and dashboards nobody trusts — and I know why “just build another report” never works.

Since 2019, I’ve helped marketing, sales, and product teams:
→ Scale revenue up to 25x
→ Eliminate manual reporting and data chaos
→ Turn analytics into a decision-making system, not a cost center

I usually work in a fractional format (3–12 months), directly with CEOs, CPOs, and CMOs — with optional execution support from my team.

Nikolay Valiotti

Data engineering is never “done.”

If you think it is, you’re already behind.

Data stacks don’t age gracefully.
Pipelines break. Schemas drift. APIs get “versioned” into oblivion. One quiet column rename, and your executive dashboard goes dark five minutes before a board meeting.

That’s not incompetence. That’s actually normal. The mistake is treating data engineering like a one-off build instead of a living system.
The job isn’t to design something flawless and walk away. It is to stay close enough to catch problems before the business does.

Strong data teams understand this:
→ Failure is inevitable — speed of detection is the real metric
→ Models will drift — adaptability matters more than elegance
→ Requirements will change — velocity beats perfection
This is why “set it and forget it” always turns into “why does no one trust the data?”

If you want data you can rely on, you need three things:
1 — Monitoring that screams the moment something breaks
2 — Engineers who can trace, debug, and patch quickly
3 — A habit of revisiting data contracts when sources evolve

Data is infrastructure. And no serious business builds infrastructure once and never inspects it again.
—
♻️ Repost if you’ve ever been paged because a silent schema change wiped out a revenue dashboard.
Follow me Nick for more unfiltered takes on data strategy, engineering, and keeping systems alive long after launch.

Nick Valiotti

Fractional CDO | Helping Scaling Tech founders turn data into faster decisions | Founder @ Valiotti Analytics

One thing I have to say out loud:
Excel is incredible. And this is exactly the problem with it.

The better Excel works, the longer companies cling to it. The longer they cling, the harder it becomes to outgrow.

Every team hits the same phase:
→ one spreadsheet
→ turns into twelve
→ turns into forty
→ turns into a folder full of excels

At that point, no one knows what’s real anymore.
This isn’t a tooling issue. It’s infrastructure debt.
And it compounds quietly — until decisions start disagreeing with each other.

In my book, Your Fractional CDO, I break down why spreadsheets become dangerous at scale — and what to fix before dashboards, KPIs, and leadership conversations drift out of sync.

Because the answer isn’t “ditch Excel.”
It’s: stop using it as your entire data strategy.

What actually works instead:

1 — Centralize your data.
Stop exporting CSVs like confetti. One shared source—Sheets, a small warehouse, anything—is better than ten “perfect” reports.

2 — Assign ownership.
If everyone owns the data, no one does. Quality only scales when accountability exists.

3 — Automate the pipes, not the panic.
Connect systems directly. Excel is fine as a front end—if the plumbing behind it isn’t chaos.

4 — Measure less. Decide more.
Fifty KPIs don’t make you data-driven. A handful of metrics that guide real decisions do.

Excel belongs in modern data stacks. Just not at the center of them.
Tools don’t create clarity. Frameworks do.

Fix the foundation before it fails you. And grab Your Fractional CDO — unless you genuinely enjoy living one broken formula away from disaster → https://lnkd.in/d--GjMpr

undefined undefined

ClickHouse Developer

Tableau Desktop Certified Associate

A Course for Executives

Digital Transformation Academy

Founder @ Valiotti Analytics | Marketing & Sales Dashboards for Growth-Ready Companies | PhD & MA in Data Analytics

🚀 Founded valiotti.com — the go-to data agency for fixing messy marketing analytics.
Top-rated on Upwork & Clutch. Trusted by growth-stage teams to build dashboards, clean up data chaos, and drive faster decisions. Partnered with AWS, Google Cloud, and Salesforce.

Ex-Head of Consumer Marketing at the largest eastern retail company.
I’ve been the one chasing targets with thousand-row tables, slow reports, and siloed data — I know the pain.

Since 2019, I’ve helped marketing and sales teams:
→ Scale revenue up to 25x
→ Save hundreds of hours on reporting
→ Get insights that actually drive action

Let’s talk: https://calendly.com/nick-valiotti

Let's discuss your project and grow you business with the power of data

Join Friday Data Snack for weekly data, leadership, and business insight

A data strategy that looks perfect on paper

usually dies in real life.

I’ve seen this pattern across companies of every size.
There’s always a polished deck, five-year vision, clean architecture, immaculate governance, dashboards that could win design awards.

And then you look at how decisions are actually made.
→ Sales still lives in spreadsheets
→ Marketing trusts the ads manager more than internal reports
→ Finance keeps a “shadow model” just in case
And that's not because people hate data, but simply because the strategy never met reality.

From a Fractional CDO standpoint, this is the core mistake:
teams confuse designing a system with changing behavior.

The thing is, data strategy is an operating model. So if adoption isn’t happening, the problem is rarely tooling. It’s usually one of these:
1. No clear decision owner
2. No agreed definition of “good enough” data
3. No incentive to switch from existing workflows
4. No tolerance for messiness during transition

A strategy that only works when everything is clean, aligned, and fully implemented is just a concept. Real data leadership starts smaller and rougher:

– Which decisions actually matter this quarter
– Who owns them
– What data is “good enough” to act
– Where the workflow already lives
If the answer doesn’t fit into how people already work, it won’t be used, no matter how elegant it looks.

Strategy isn’t supposed to be beautiful.
It’s supposed to survive contact with reality.

P.S. Had this convo with Marcel Dybalski a while ago, thought more people need to hear it!
-------

👋 Hi, I’m Nick — Fractional CDO. I help teams turn “data vision” into decisions people actually use. If your dashboards look great but nobody trusts them, that’s usually where I start.

AI & Data Platform Engineer | GCP & Fabric Certified

“Data-driven” — easy to say, brutal to explain.

I turned it into my elevator pitch.

Step 1 — Get all your data in one place
Your numbers live everywhere — website, ads, CRM, spreadsheets.
First, you just need to collect them.
→ Fivetran: connects the tools for you (free up to 500K rows).
→ Python: for people who like to build things themselves.
Goal: one single place where all your data lives.

Step 2 — Store it safely
Once you’ve collected it, you need to keep it somewhere reliable.
That’s BigQuery.
Think of it as a big digital storage room for all your numbers.
It’s cheap, fast, and handles anything you throw at it.

Step 3 — Make it visual
Turn all that data into something you can actually use.
That’s your dashboard.
It doesn’t really matter which tool you go for. Looker Studio, Tableau, whatever.
A good dashboard shows what matters — sales, traffic, growth —
without making your eyes hurt.

Bonus — real-time?
Want to see what users do on your site right now?
Use Segment.
Just don’t send every click straight into BigQuery — your CFO won’t like the bill.

That’s it.
Three steps. Zero $10K toys.
Just a simple stack that works.

♻️ Repost if this has been the most insightful elevator ride this quarter.
Follow me Nick to learn how to explain it to your grandma next.

First day at work in 2026. Fresh priorities. Same first request:
“I need a dashboard.”

Every analyst hears this. Regularly. Confidently. With zero additional context. Because 9 times out of 10, when a stakeholder says “I need a dashboard,” what they actually mean is:

“Something feels off and I don’t know why.”
The dashboard is just the physical form their confusion takes.

The moment you accept the request as is, you’ve already lost. You’re no longer solving a problem — you’re redecorating anxiety.

It usually goes like this: “Can we get a dashboard to track campaign performance?”

Sure. You build it. Then come the sequels:
→ “Can we slice it by region?”
→ “Actually, over time?”
→ “What if we try a different chart?”
→ “Can we add just one more KPI?” (famous last words)

Congrats. You’ve entered the Dashboard Cinematic Universe. This happens when analysts become BUILDERS instead of PARTNERS.

A good analyst doesn’t just execute. They investigate. They pause and ask the most dangerous question in analytics:

“What decision will this help you make?”

That one sentence ends most bad dashboards on the spot. Because sometimes the answer isn’t a dashboard at all. It’s:
– a short memo
– a mini-report
– or a five-minute conversation that saves 6 weeks of work

If you want to be seen as a partner, not a chart factory, this is the shift that matters.

And this is exactly why I built a simple dashboard discovery framework, the Dashboard Canvas.
No endless meetings. No vague requests. No dashboards that get ignored.
If you build dashboards for a living, it’ll save you hours → https://lnkd.in/djYMn3Fp

Repost it so others can use it too. Use it on your next project.

As a CEO and founder — this dashboard is just great. Simply because it respects how execs actually think.
They don’t open a dashboard to analyze. They open it to confirm or challenge a belief.

This one starts with the only question that matters:
Did we hit quota?
101.3%. Answered. No interpretation required.

Immediately after, it answers the second executive question:
Is this short-term noise or a real trend?
MTD, YTD, last month, last year — side by side, same visual language. 
The user doesn’t have to remember numbers. They just register direction.
Green means “we’re fine.” Red means “ask why.” That’s executive UX.

What’s also strong here is sequencing. Cumulative → period → comparison → distribution over time. That mirrors how leadership conversations actually happen in a room.

The total sales number isn’t there to impress anyone. It’s there to ground the conversation. Execs anchor on scale before they talk variance. This dashboard knows that.

The monthly chart at the bottom is doing quiet but serious work. It shows where execution deviated from expectation and how often that happens.
That’s a performance management view, not an analytics one.

Most importantly:
There is no moment where an exec would have to ask, “So what should I do with this?”
They can already tell:
– whether to push the team
– whether to stay hands-off
– whether to zoom into a specific month
– whether this is a sales issue or a planning issue

That’s why this is an executive dashboard.
Not because it has fewer charts. But because it collapses complexity into decisions, instead of pushing complexity onto the reader.

That’s the difference most teams miss. Bravo, Sebastine Amede!

----

👋 Hi, I’m Nick — fractional CDO. I break down what actually makes dashboards useful to leadership. Follow for reviews like this!

You hired a CDO.

Nothing changed. Why?
Because one person can’t fix an 8-person problem.

Most data disasters don’t come from bad talent.
They come from asking one human to be:
a strategist, an architect, an engineer, an analyst, a BI developer, an AI owner,a translator to execs, a firefighter on Slack. 

That’s not leadership. That’s organizational confusion.
So what happens? Your “data scientist” is building pipelines.
Your analyst suddenly owns AI because “they’re good with numbers.”
Your BI dev is trying to architect a warehouse.

Every team has dashboards. None of them agree.
Everyone is busy. Nothing is reliable.

This is what should be happening instead:
→ Data Engineers move data — clean, automated, invisible.
→ Data Architects design how everything fits and scales.
→ BI Developers turn tables into usable interfaces.
→ Analysts explain why numbers move and what to do next.
→ ML / AI roles only exist after the foundations are solid.

It’s a relay, not a solo marathon. Each role carries the baton at the right moment.

Now here’s the uncomfortable truth:
A full-time CDO is expensive. And most companies don’t actually need one sitting on payroll.

What they need is the CDO thinking:
→ role clarity
→ sequencing
→ ownership
→ alignment to revenue and growth

That’s why I wrote Your Fractional CDO. Not a leadership memoir. Not AI hype.
A practical playbook for building a data system that works — without hiring an army or burning budget. Available on Amazon → https://taap.it/Yt0nMRT

A common question I hear:
“What does a CDO actually do? And why can’t our senior data analyst just do that?”

It’s a fair question. The thing is:
A data analyst’s job is to answer questions.
A CDO’s job is to decide which questions matter — and which ones shouldn’t be asked at all.

Analysts work inside the system.
CDOs are responsible for the system itself.

That means:
→ deciding what data the company should trust
→ setting priorities when everything feels important
→ choosing what to build, buy, or ignore
→ clarifying ownership so decisions don’t stall in debate
None of this is about SQL, dashboards, or tools. It’s about judgment.

At Series A–C, this gap becomes visible. Data exists. Dashboards exist. Talent exists. But nobody is accountable for direction.

That’s usually when founders start feeling the drag: slow decisions, duplicated work, quiet confusion.

This is where a fractional CDO helps — not by replacing analysts, but by giving them direction.

I wrote a short guide explaining how this actually works in practice: what a CDO does, when the fractional model makes sense, and what typically changes in the first 90 days.

Feel free to grab it in my Featured or via the link → https://lnkd.in/d8TDg3DJ

----

👋 I’m Nick, Fractional CDO. If you’re looking for this kind of data leadership, DM me.

My feed and comment section keep panicking over “the death of the data analyst.” 

Which tells me most people still don’t understand what the job was in the first place.

A data analyst was never meant to be a SQL code generator. Not before. Not now. SQL was the easiest part of the job — so we turned it into the job.

For years, we hired analysts to write queries, ship exports, and keep the dashboard lights on. But that role always had an expiration date. AI just made it obvious.

Fast-forward to 2026.
→ LLMs write SQL in seconds.
→ Warehouses scale automatically.
→ “Can you pull this data?” is no longer a skill.

But here’s what didn’t change:
CMOs still don’t trust the numbers. CEOs still ask, “Wait — what’s included here?” Meetings still end with, “Let’s double-check that.” Because the business didn’t suddenly become analytical!!

No one is validating assumptions. No one is pressure-testing metric logic. No one is responsible when AI output sounds right but is wrong.
That responsibility moved to the analyst.

A strong analyst today:
— Lives inside the business, not the warehouse.
— Translates messy questions into precise metrics.
— Knows where models hallucinate and why.
— Owns conclusions, not queries.

SQL is cheap now. Dashboards are everywhere.
Judgment is still rare. That’s the job.

Nothing says “data-driven company” like KPIs updated by hand every Friday.
But sure, let’s talk about machine learning next quarter.

And yet, every quarter (if you’re lucky, if not — every month), someone asks for a “cool dashboard.”
As if design can fix dysfunction. Surpise-surprise, it won’t.

Here’s the quick sanity checklist every BI analyst to run before saying yes:

1 — Data foundation
If your inputs aren’t automated, every chart is a ticking time bomb. Dashboards built on manual data rot faster than milk in the sun.

2 — Metric alignment
If “Revenue” means one thing to Marketing, another to Finance, and a third to the CEO — you’re not visualizing insights, you’re hosting a family argument.

3 — Decision logic
Every chart should answer a question someone actually asks. If it doesn’t change behavior, it’s just decoration.

So next time someone says, “We just need a dashboard,” maybe ask: “Do we have data… or do we have Excel?”

Because there’s a big difference between building analytics and making PowerPoint interactive.

----------

I’m Nick, and that’s exactly what we fix at Valiotti Analytics — building systems where data flows, definitions align, and dashboards actually tell the truth. If that sounds like your next step, let’s talk!

If your data work mostly happens with IT, you’re probably not doing strategy. You’re doing infrastructure.

I always say that and my recent conversation with Mike Drake was another proof to its rlevance across teams. We were talking about fractional CDOs, when they actually add value, and why so much data work quietly underdelivers. Here’s the pattern we kept coming back to:

Strategy doesn’t live where tickets live, it lives where outcomes live.
CEO. CFO. CMO. Revenue. Risk. Growth. Allocation.

When data is treated as an IT function, a few things happen almost automatically:
→ The work optimizes for systems, not decisions
→ Dashboards pile up, but ownership stays fuzzy
→ Teams argue about numbers instead of acting on them

That’s not a tooling, but a positioning problem.
The real job of a fractional CDO isn’t “better dashboards” or “one more model.” It’s turning data into a shared business asset:
– Owned by leadership
– Aligned to decisions
– Designed to compound over time

A few things Mike and I aligned on quickly:
→ Execution is getting commoditized. AI and cost pressure made sure of that.
→ SQL is cheap now. Thinking isn’t.
→ Dashboards should be disposable. Foundations shouldn’t.
And if you don’t historize from day one, you don’t get a second chance later.

The path I keep seeing work looks like this:
IT-centric data → shared business capability → revenue leverage. That’s where strategy actually lives. And that’s the only place it gets paid for.

If you’re a business leader trying to understand where your data work is helping — and where it’s quietly holding you back — comment CDO below.

I’ll send you a link to a quick quiz I built that helps identify data bottlenecks, growth zones, and gives you a concrete 7-day action plan to move things forward.

And Mike, thanks again. Always a good sign when a call reorganizes a few neurons instead of just filling a calendar.

Fractional Data Leader | From Chaos to Clarity | Uncomfortable Conversationalist | Liberator of Hostage Data

Nothing fuels overengineering like a decision nobody wants to make.
I’ve seen this pattern more times than I can count.

When a team can’t agree on what actually matters, the solution quietly becomes: “Build everything.”

So data engineering turns into an insurance policy.
Extra pipelines “just in case.” Abstracted models “for future flexibility.” Another orchestration layer because maybe we’ll need it later. A schema that can support ten businesses when you barely run one.

It looks sophisticated. Feels responsible. And it’s almost always a symptom of avoidance.

Because here’s the uncomfortable truth:
Most data complexity doesn’t come from scale.
It comes from indecision.

When no one wants to commit to:
→ which metric actually drives the business
→ which questions deserve fast answers
→ which use cases can wait or die
…the platform absorbs the uncertainty.

And the platform always pays interest.

Overengineered systems don’t fail loudly.
They rot slowly:
– longer lead times
– fragile dependencies
– higher cognitive load
– teams afraid to touch anything
All in service of optionality no one ever exercises.

Good data engineering starts much earlier than tools and architecture.
It starts when someone is willing to say: “This is the decision we care about now.”

So if your data stack feels heavier every quarter, don’t ask “what tool are we missing?” Ask: “What decision are we still avoiding?”

👋 Hi, I’m Nick, fractional CDO. I help teams replace “future-proof” complexity with clarity, ownership, and decisions. If your data stack feels heavier every quarter, let’s talk.
Not ready to talk but still need help? This book might be a place to start → https://lnkd.in/dxS6W92D

Most dashboards don’t fail because of bad data or bad charts.

They fail because they were built too early.

Someone asks for “a dashboard.”
The team opens the BI tool.
SQL gets written. Charts get polished.

And only at the end does anyone ask the uncomfortable question:
“Wait… what decision is this supposed to support?”

That’s why prototyping is non-negotiable if you want dashboards that actually get used.

Prototyping forces the hard thinking before the build:
→ What questions are we really trying to answer?
→ Who is this for, and what do they need to decide?
→ Which metrics matter, and which ones are just noise?
→ What level of detail is useful vs overwhelming?

A simple low-fidelity prototype (even boxes and labels) does two critical things:
1 — It exposes vague thinking immediately.
2 — It lets stakeholders react to structure and logic, not colors and charts.
By the time you open your BI tool, 80% of the decisions should already be made.

That’s exactly why I put together this practical, no-nonsense cheatsheet for dashboard discovery and prototyping.

It’s a repeatable framework to gather clean, complete dashboard requirements — without endless meetings, vague requests, or dashboards that get ignored.

The guide walks you through:
– Running focused user interviews
– Defining business questions first
– Aligning on metrics and breakdowns
– Mapping data sources
– Translating all of that into a clear, stakeholder-friendly layout

If your dashboards aren’t getting used, don’t add more charts.
Prototype better → https://lnkd.in/dxyFYdUy

Read this book before you start your next analytics project.
Because it teaches you one lesson that will save you hundreds of hours and hundreds of thousands of dollars:

Your first question matters more than your first dataset.

Most failed analytics projects start long before the dashboards do.
They start with vague, unanswerable questions:
“How do we grow?”
“What’s working in marketing?”
“Can we be more data-driven?”
These aren’t questions.

They’re weather forecasts with a KPI attached.

A real analytics question is specific, falsifiable, and tied to a decision:
→ “Which acquisition channel produces customers with the highest 90-day LTV?”
→ “Which step in the funnel creates the biggest revenue leak?”
→ “Which product segment is worth prioritizing next quarter?”

Ask better questions → get better insights → make better decisions.
Every transformation starts there.

This principle is so fundamental that it’s the backbone of my new book, Your Fractional CDO.
Not tools, not architectures, not AI dashboards with glowing neon charts.
Just decision-making clarity.

In the book, I break down how to rewrite messy business questions into clean, analytics-ready ones — the exact process I use with founders, CMOs, and product teams.

Once you do this, everything else gets easier:
You know what data you actually need, you stop building dashboards “just in case”, you avoid 3-month detours into perfect ETLs that solve nothing, and you finally get insights that drive action instead of confusion.

Check out Your Fractional CDO on Amazon → https://lnkd.in/dU-kiEUf

If you think AI will “solve analytics” in 2026, you’re focusing on the wrong problems. Here are my predictions. 3 things I’m confident about.

1 — People trust people. Not models.
This hasn’t changed. It didn’t change before AI, it won’t change after.

When you sent a report to a director, they trusted it because you owned the numbers. You could recheck them. You were responsible.
No one wants to blindly trust LLM output. There is always a human who validates, verifies, and stands behind the result.

That’s why a new role emerging this year: AI-enabled analyst.
Not “AI replaces analyst”. Analyst who uses AI and owns the outcome.

2 — Data is still the foundation.
No matter how good LLMs get, everything still depends on data quality: clean structure, clear logic, well-designed models.

Yes, LLMs can help design schemas and even suggest warehouse structures. But you can’t just say:
“Hey AI, fix my data.”

You still have to explain:
→ what the business actually does
→ how metrics should behave
→ which decisions depend on them
Good prompting still requires a brain, context, real business understanding. AI doesn’t remove thinking. It exposes the lack of it.

3 — The chat interface is already annoying.
Let’s be honest: business leaders don’t want to chat with data. One question leads to ten more. No CXO is spending their day interrogating a chat window about metrics.

Chat is great for exploration, not for decisions.
What’s coming instead:
→ You see a chart.
→ A metric drops.
→ You click it.
→ AI runs the analysis in the background.
→ Returns 2–3 hypotheses actually worth checking.

And then — the crucial part — someone still owns the conclusion.
Same role as point #1.

The boring truth: AI accelerates analytics. Responsibility doesn’t scale automatically.

What changes do you think we’ll see?

--------

👋 Hi, I’m Nick — Fractional CDO. I write about building analytics teams that actually work, not just demo well. Follow for practical takes on data, AI, and decision-making.

Most dashboards fail for a simple reason:

They answer questions nobody asked.

We jump straight into charts, colors, and layouts… but skip the one step that actually makes a dashboard useful:
Understanding how people will use it.

Here’s the part so many teams miss:
A useful dashboard isn’t built around metrics. It’s built around decisions.
Before opening Figma, Tableau, Power BI, or anything else — ask your users:

→ What decision do you need to make faster?
→ What is the pain with your current process?
→ What do you consider a “good” or “bad” week?
→ Which slices, filters, or comparisons do you need every time?
→ Which metrics do you actually trust?

If you can get clear answers to these five things, the dashboard basically designs itself.
No more 12-widget monsters. No more “Can we add one more chart?” No more reports that look great… and get ignored.

Because the truth is simple:
Dashboards fail when they’re built for the tool, not the user. Dashboards succeed when they’re built for the decision.

If you want a shortcut:
I put together a Dashboard Interview Cheatsheet — the exact framework I use with clients to turn vague requests into dashboards that actually get used.

Happy to share it: https://lnkd.in/dNSPbAzx
Let’s make dashboards people actually use.

A lot of companies think they’re “behind” on data.

In reality, they’re just out of sequence.

I wrote Your Fractional CDO for leaders who feel pressure to “level up” analytics, but aren’t sure what level they’re actually on. The book walks through a simple but uncomfortable truth:

Most teams don’t fail because they lack data. They fail because they try to operate like a Stage 5 company while still living in Stage 2 or 3.

That leads to:
→ overbuilt dashboards
→ premature tooling
→ stalled initiatives
→ and a constant sense of “why isn’t this working yet?”

The goal isn’t to be “data-driven.” It’s to build the next layer that actually fits your business right now.

Normally, Your Fractional CDO is available on Amazon. But I’m giving free copies to a small number of founders / execs who feel stuck between “we know data matters” and “we don’t know what to do next.”

If that’s you, comment “CDO” and tell me what stage you think you’re in — or where things feel forced. If the book can help you move forward cleanly, I’ll send you a copy.

When was the last time a dashboard helped you make a decision without someone saying, “let me double-check that”?

If the answer isn’t “recently,” you’re not alone.
I see this all the time:
– the dashboard is open
– the numbers look reasonable
– and yet the decision stalls

Someone asks for another cut. Someone pulls up a spreadsheet. The meeting ends with “let’s come back to this.”

Usually, nothing is technically broken. What’s missing is clarity around three things:
→ who owns the decision
→ which metrics actually matter
→ whether the numbers are trusted enough to act on

So I put together a short, practical check to make this visible. It’s called the CDO Healthcheck and takes under 5 minutes.

No tools. No vendors. No buzzwords. You just answer a few questions based on how things work today — not how they’re supposed to work. At the end, you get a clear read on where friction is coming from and what to fix first.

If you want the link, comment "CDO" and I’ll send it over.
If dashboards feel heavy instead of helpful, this will tell you why.

If I had to give just one piece of data advice, it would be this:
Build the Boston Public Library.

This idea clicked for me over the holiday break, spending a few quiet hours there. No rush. No noise. Just space designed for thinking.

A good library does a few things quietly and exceptionally well.
Books are easy to store — because there’s a system. Knowledge is easy to access — because things are labeled. Learning feels natural — because the space removes friction.
The room itself helps you think.

Now look at most data environments.
Files stacked like books on the floor. Queries written once and forgotten. Sheets that “kind of” match, but never fully agree. Dashboards that exist, but no one sits down with them.
That’s what a library looks like without shelves.

In a real library:
– every book has a place
– similar topics live near each other
– you know where to look before you start searching
– history is preserved, not overwritten

In good analytics:
– every metric has a definition
– related data lives together
– you know where truth lives
– past decisions stay explainable

Libraries don’t optimize for storage. They optimize for understanding.
Great analytics works the same way. It’s not about having more data. It’s about making learning feel inevitable. Because mess spreads quickly. But order compounds quietly.

So if your dashboards feel heavy, your metrics feel fragile, or your team keeps asking the same questions again and again — you don’t need a new tool.
You need shelves, labels, and silence.
Build the library.

------

👋 I’m Nick, Fractional CDO. I write about building calm, decision-ready analytics systems. Follow if this is the kind of data thinking you want more of.

Most companies don’t waste money on data tools.
They waste it on the wrong data hires.

Last year (had to do it!) I released a book called Your Fractional CDO — and if you’re a founder, CxO, or anyone involved in decision-making, it can save you a lot of time, money, and frustration.

It’s a practical 101 to data analytics, written in plain language. But one section matters more than most: data roles.

Your first data hire can either accelerate your business or quietly burn months of runway. Not because they’re bad — but because each role solves a different layer of the system, and there’s a natural order to building a data team.

Here’s the short version:

Data Engineer — pipelines, integrations, storage, identity stitching. They build the foundation everything else depends on.

Data Analyst — investigations, insight generation, decision support. They tell you what changed — and why it matters.

Analytics Engineer — transformations, models, metric definitions. This is where CAC, ROAS, LTV, and retention become trustworthy.

Data Scientist — forecasts, experiments, advanced modeling. Powerful, but only once the basics are stable.

CDO / Head of Data — strategy, architecture, prioritization. The role that ensures data work drives ROI, not noise.

In the book, I break down each role in detail: what they actually do, when you need them, what problems they solve, and the risks of hiring them too early or too late.

I also share a simple framework for assembling your first data team based on your company stage and the decisions you’re trying to make.

If you’re hiring your first (or next) data person, read this first.
Your Fractional CDO is out now on Amazon: https://lnkd.in/dczU_G34

Transform your AI & data ambition into action | xQuantumBlack, xMcKinsey | Global top 100 Innovators in Data & Analytics ’24 | Founder & Director, Cambiq

AI and data are creating some of the most exciting opportunities organisations have seen in years.

With the right approach, they can unlock better decisions, smarter operations and more rewarding ways of working.

I help organisations establish, nurture and scale their use of AI, AI agents and data, turning potential into real outcomes and sustained enthusiasm.

If you’re wondering how you, your business or your team can get up to speed, get started and confidently drive value from AI and data, I can help. 
Whether you’re a large or small organisation, I bring pragmatism, creativity and a strong bias to action to support you to succeed.

If you're wondering how to get started with AI and data, I offer executive education sessions to help bring to life the opportnity for you and your team. We can tailor this to your industry and business.

If you want to do more, I can help you:
1. Develop a clear AI and data strategy that drives growth, efficiency and trust, from early ideation through to business cases and board-ready narratives
2. Create and embed responsible AI and data governance that reduces regulatory risk while making it easier for teams to use data well
3. Build practical adoption strategies so leaders and teams feel confident, capable and motivated to use AI and data
4. Set up a new AI or data program, or turn around an existing one, to maximise the chance of lasting success

I’m the founder and Director of Cambiq, where we work with organisations to turn their own data into commercial and operational advantage through AI strategy, governance, education and delivery.

My background combines McKinsey consulting, senior industry leadership and strong technical foundations. That mix helps me shape strategy, navigate complex stakeholder environments and, most importantly, drive real adoption on the ground.

Inspiring growth and fostering a culture of learning are passions I bring to everything I do. I’m genuinely optimistic about what AI can enable when people are supported to understand it and use it well.

I work flexibly, either on a project basis or as an interim executive embedded within your team. If you’re looking for a partner, advisor or speaker to help your organisation build confidence and capability in AI and data, I’d love to connect.

You can reach me at hello@cambiq.com

Clare Kitching

Everyone wants the AI lift.
Few want to stand in the cold and build the foundations.

Most organisations want outcomes fast. 
Better productivity. 
Lower costs. 
Smarter decisions.

So they jump straight into AI solutions.

What they forget is this: 
AI amplifies what already exists.

Good data gets better. 
Bad data gets louder.

Broken workflows get automated. 
Confusion spreads faster.

The unglamorous work matters: 
Data fundamentals. 
Workflow redesign. 
Clear accountability.

That’s the difference between a pilot that impresses and a capability that lasts.

Curious to hear...
If you had to invest in just one foundation this year, which would it be?

♻️ Repost to help someone get their foundations in shape.
🔔 Follow Clare Kitching for insights on unlocking value with data & AI.
👋 Want more from me? Sign up for my newsletter here: https://lnkd.in/giQ3b6Fi

Transform your AI & data ambition into action | xQuantumBlack, xMcKinsey | Global top 100 Innovators in Data & Analytics | AI & data strategy, governance and capability building

Everyone wants the AI lift. Few invest in foundations.

The fastest way to understand AI governance?
Stop thinking about data. Start thinking about decisions.

There's a lot of noise around governance in the age of AI.
And leaders cut through some of that noise by understanding the split between inputs and outcomes.

Data governance protects the inputs.
It’s about accuracy, privacy, lineage, access and compliance.
It answers: Can we trust the data going into our systems?

AI governance protects the outcomes.
It deals with fairness, explainability, drift and human oversight.
It answers: Can we trust the decisions made on outputs of AI systems?

Both matter.
Both are essential.
But they serve different purposes.

What I’ve seen inside organisations this year is that AI governance doesn’t sit neatly with one owner.
Legal, risk, product, ethics, engineering and the business all hold a piece.

That’s normal, because the outcomes affect so many parts of the organisation.

The companies making real progress connect the two:
They build strong data governance first. Clean foundations mean AI systems fail predictably, not catastrophically.

Then they layer AI governance on top as the mechanism that gives business leaders confidence to actually use what's been built.

Have a look at the two side by side to see where each plays a role.

What's your experience? Where does AI governance live in your organisation, and is it working?

♻️ Repost to help someone get their governance layers set.
🔔 Follow Clare Kitching for insights on unlocking value with data & AI.

Microsoft just opened the door to AI mastery.
 
No fees. No barriers. No excuses.
If your 2026 goals have anything to do with AI, start here.

These courses come with certificates. 
They're built by experts. 

And they're waiting for you right now.
 
📌 save this, you'll want it later

// START HERE: FOUNDATIONS
 
1️⃣ Career Essentials in Generative AI 
↳ A clear, beginner-friendly path into how GenAI works, where it's used, and why ethics matter. 
🔗 https://lnkd.in/gk5rpgeU
 
2️⃣ Artificial Intelligence for Beginners 
↳ A full 12-week intro to AI: NLP, neural nets, deep learning and hands-on labs. 
🔗 https://lnkd.in/dfyFHeJN
 
// APPLY IT: IMMEDIATE USE
 
3️⃣ Generative AI Productivity Skills 
↳ Learn to work smarter with AI: better prompts, faster workflows, stronger output. 
🔗 https://lnkd.in/gngmKuq6
 
4️⃣ Adopting Copilot for Microsoft 365 
↳ Master Copilot in the tools you already use: Word, Outlook, Teams, everything. 
🔗 https://lnkd.in/gUiktsXS
 
// LEAD IT: STRATEGY & MANAGEMENT
 
5️⃣ AI for Managers 
↳ How managers can use AI for coaching, decision support, and better team execution. 
🔗 https://lnkd.in/ggnf5yQy
 
6️⃣ Transform Your Business with Microsoft AI 
↳ A strategic, leader-level program on planning and scaling AI across the enterprise. 
🔗 https://lnkd.in/gz3bMRaN
 
// BUILD IT: TECHNICAL DEPTH
 
7️⃣ Using Azure OpenAI Service 
↳ A technical guide to Copilots, LLMs, prompt design, and Azure's AI ecosystem. 
🔗 https://lnkd.in/g2-es725
 
8️⃣ Build Apps with Azure & Power Virtual Agents 
↳ Build real AI apps and conversational agents using Azure AI + Copilot Studio. 
🔗 https://lnkd.in/gdCnTBZE
 
9️⃣ Create Custom Machine Learning Models 
↳ Train, deploy, and fine-tune your own ML models with Microsoft's tools. 
🔗 https://lnkd.in/guAVcKDC

Which one are you starting?

♻️ Repost to help someone upgrade their AI skills.
🔔 Follow Clare Kitching for insights on unlocking value with data & AI.
🍿 Want more from me? Sign up for my newsletter here: https://lnkd.in/giQ3b6Fi

Data isn't the hard part. 
Understanding each other is.

Ontology. Lineage. Semantic layers. Vector databases.

I've been in data for over 15 years, and sometimes even I feel like I'm decoding a foreign language. 
We've turned simple ideas into jargon that makes non-data people tune out.

Here's what these terms actually mean and why they matter for AI:

▶️ Ontology 
A shared definition of your core business concepts and how they relate. 
It gives AI clear concepts to reason about instead of guessing.

▶️ Entity 
A real world thing like a customer, product or event. 
It helps AI tell the difference between people, products and moments in time.

▶️ Metadata 
Data that explains other data. 
It tells AI what something means, how fresh it is and whether it can be trusted.

▶️ Physical layer 
Where data is stored and processed. 
It shapes how fast, scalable and reliable AI workloads can be.

▶️ Logical layer 
How data is organised conceptually, not physically. 
It shields AI from raw technical mess.

▶️ Semantic layer 
A business friendly layer with agreed definitions and metrics. 
It stops humans and AI arguing over what a number actually means.

▶️ Schema 
The formal structure of what data exists and what type it is. 
It gives consistency so AI knows what to expect.

▶️ Data modelling 
How entities and their relationships are designed. 
It reduces confusion in how AI interprets data.

▶️ Data virtualisation 
Accessing data from many sources without copying it all. 
It lets AI work across systems seamlessly.

▶️ Vector database 
A database that searches by similarity, not exact matches. 
It enables richer retrieval and context for AI.

▶️ Data pipeline 
How data flows from creation to consumption. 
It keeps AI fed with timely and relevant inputs.

▶️ Orchestration 
Coordinating when and how pipelines run. 
It keeps jobs reliable and in the right order.

▶️ Data quality 
How accurate, complete and consistent the data is. 
It directly affects confidence in AI outputs.

▶️ Observability 
Seeing what data systems are doing and spotting issues early. 
It helps catch drift and weird behaviour before damage is done.

▶️ Data lineage 
Where data comes from, how it changes and where it’s used. 
It adds transparency and explainability to AI decisions.

None of this is magic. 
But together, it’s the foundation AI stands on.

What other terms would you add as essential?

♻️ Repost to help someone get their idea into action.
🔔 Follow Clare Kitching for insights on unlocking value with data & AI.
💎 Get more from me with my free newsletter here: https://lnkd.in/giQ3b6Fi

15 Essential Data Concepts: Data foundations AI depends on

To everyone who thinks the data team just makes dashboards…
Here’s what we really do:

When I tell people I work in data, they often think I just make dashboards.
Some imagine I spend my days in excel.
Others think I'm just waiting around for their "quick" data ask.

But very few people really understand what it means to work in a data team.

So, here’s what we actually do 
🧹  We collect, clean, and organise messy data from dozens of sources.
🏗️  We build reliable pipelines that keep the business running.
🏃‍♀️‍➡️  We turn numbers into insights, and insights into action.
📢  We communicate complex findings clearly and simply. 
🪟  We translate between business needs and technical realities.
📐  We manage projects, set standards, and protect data quality.
💡 And yes, sometimes… we do make dashboards.

In short:
A data team doesn’t just report on what’s happening,
we help the business understand why and decide what’s next.

We’re engineers, analysts, storytellers, translators, governance gurus, and problem-solvers.

And most days, we’re just trying to make the data a little cleaner and clearer than we found it.

If you work in data, what’s one thing people often overlook about your role?

♻️ Repost to give your data team the love they deserve.
🔔 Follow Clare Kitching for insights on unlocking value with data & AI.
💎 Get more gems from me for free in my newsletter here: https://lnkd.in/giQ3b6Fi

We obsess over intelligence.
The real risk is behaviour.

Most AI conversations still follow the same storyline: 
Model gets smarter. Outputs get better. Jobs disappear.

That mental model is already outdated. And it breaks once you step into agentic AI.

Thanks to my friend Luís Rodrigues for the visual that makes this obvious.

Models are just one aspect. And honestly, they’re becoming the least differentiating part of the stack.

We’ve moved past tools that answer questions. We’re building systems that plan, act, remember, recover, and improve over time.
That changes everything.

Because once AI becomes operational, the real work begins.

ML foundations turn data into signals. 
Deep learning finds patterns at scale. 
GenAI creates text, images, audio, video.
AI agents execute.

Agentic AI coordinates intent at scale.

And suddenly we’re no longer “using AI”. 
We’re managing behaviour.

Memory. Feedback. Cost. Risk. Control. Failure modes.
None of that is glamorous. But that is where value and trust are actually created.

The question to start asking is:
“What happens when this fails, forgets, or confidently does the wrong thing?”

That answer will decide whether AI becomes leverage or liability.

♻️ Repost to help someone see the bigger picture.
🔔 Follow Clare Kitching for insights on unlocking value with data & AI.
🍭 Get more from me with my free newsletter here: https://lnkd.in/ghBtk6jR

AI that delivers the biggest returns is usually the least exciting to explain.

Executives don’t need more AI hype. 
They need better mental models for AI.

AI feels like a solve all technology. 
In reality, it’s a set of very different approaches.

Most organisations miss value by skipping straight to the top. They jump to GenAI or agents before exhausting simpler, proven options.

Rules and classical ML already power many high ROI use cases. 
They can be easier to deploy because there’s proven patterns to follow.

I've seen value compound when you match the right tool to the job,
and risks compound when you don’t.

Here's an overview of how to think about AI:
▶️ Rules give clarity and auditability.

▶️ Predictive models give reliable ROI.

▶️ Deep learning gives power at a cost.

▶️ Generative AI gives leverage with oversight.

▶️ Agents give autonomy, and that demands discipline.

AI isn’t one decision. 
It’s a portfolio you actively manage.

Start with the simplest approach that works. 
Add complexity only when it earns its place.

Most teams skip the middle and jump to the top.
That’s why results feel fragile.

Be explicit about where human judgement stays firmly in the loop.

Which layer do you think most organisations miss value from today?

♻️ Repost to help someone get their idea into action.
🔔 Follow Clare Kitching for insights on unlocking value with data & AI.
💎 Get more from me with my free newsletter here: https://lnkd.in/ghBtk6jR

AI isn’t a single technology you adopt.
It’s a set of building blocks you choose from.

AI feels sudden.
But what looks like an overnight breakthrough is really decades of layers quietly stacking up.

Thanks to my friend Luís Rodrigues for the image inspiration, it captures that story better than most explanations I’ve seen.

Here’s how we got here…

⏭️ At the bottom: Classical AI
Rigid rules. No learning.
Think eligibility checks in insurance or compliance rules in banking.
Useful, but fragile.

⏭️ Machine Learning
Data replaces rules.
Patterns replace assumptions. 
We see better sales forecasting, fraud detection or customer churn prediction.
Better decisions, still narrow.

⏭️ Neural Networks
We stop telling machines what matters.
We let them figure it out.
Messy inputs suddenly become usable. 
This has given us image recognition for quality control or speech to text for call centres.

⏭️ Deep Learning
Scale hits hard.
Transformers unlock language.
Vision finally works. Computers can spot defects humans miss.
This is where performance jumps, not inches.

⏭️ Generative AI
The public moment for AI.
Suddenly AI writes, draws, builds.
Not perfect. But good enough to turn heads and change behaviour.

That’s why everything feels different now.

⏭️ Agentic AI
Is this the shift people underestimate or overhype?
Generative AI gives answers.
Agentic AI takes action.
Memory. Planning. Tools. Execution.
Software that behaves less like a model and more like a junior employee.

Here’s the part worth remembering...
Every layer in this stack still matters.

Not every business needs agentic AI straight away.
Not every problem needs generative AI.

→ Rule based systems still power compliance and control.
→ Machine learning quietly drives forecasting and optimisation.
→ Neural networks and deep learning handle complexity at scale.
→ Generative AI accelerates thinking and creation.
→ Agentic AI orchestrates work when the foundations are strong.

The opportunity is in knowing & choosing which layer creates value for your problem.

AI isn’t a single bet.
It’s a portfolio.

Are you missing a layer of value in your business today?

♻️ Repost to help someone understand the layers of AI.
🔔 Follow Clare Kitching for insights on unlocking value with data & AI.

AI governance breaks down the moment no one owns the decision.


A couple of weeks ago I shared a simple way to think about data governance vs AI governance.
Governing inputs vs governing outcomes.

There was broad agreement on one thing. 
Data governance alone isn’t enough. AI governance is another layer, protecting outcomes.

But one gap kept resurfacing:
Decision accountability.

Who owns the call when the model is technically right, but the outcome is wrong?

Not more process. 
Not another committee.
A clearly named decision owner.

You can govern inputs perfectly and monitor outcomes rigorously. 
Without a named decision owner, neither will protect you when something goes wrong.

The first time this matters won’t be in a governance review. 
It'll be when a customer, regulator, or exec asks, “Why did the system do this?” And the room goes quiet.

At that moment, the organisation will look for an owner. 
If you haven’t named one in advance, the system itself becomes undefendable.

This decision owner needs to understand both sides. 
The data being used. And the AI guardrails in place.

I’ve seen this work well in high-risk use cases, even before GenAI. 
Everyone knew the stakes. 
And there was a plan for when things went wrong.

As AI moves into everyday decisions, many organisations already have models in decision paths without having named who owns the decision.

When accountability is missing, the biggest issue isn’t blame.
It’s that no one can explain why the outcome happened.

And when you can’t explain it, you can’t learn, adjust, or improve the system.

If an AI-driven decision caused real harm tomorrow, could your organisation name the decision owner in under 10 seconds?

Someone always owns the decision in practice. 
The question is whether you’ve acknowledged it yet.

♻️ Repost to help someone develop their decision accountability.
🔔 Follow Clare Kitching for insights on unlocking value with data & AI.
💎 Get more from me with my free newsletter here: https://lnkd.in/ghBtk6jR

Data governance is the unsung hero of AI success.
Overlooked when it works, blamed when it falters.

Here are 8 harsh truths of data governance, 
and how brilliant teams navigate them:

1. Good Governance is invisible
→ When it works, no one notices. When it fails, everyone blames it.
TRY THIS: Track and share small wins to make the invisible visible.

2. Perfect data does not exist
→ Quality will never be flawless, and that’s okay.
TRY THIS: Define “good enough” for each use case and move fast.


3. Executive love the idea, but hate the cost
→ Governance rarely gets real budget or airtime.
TRY THIS: Tie every initiative to business value or risk avoided.

4. 80% of people tune out at “governance”
→ To most teams, it sounds like rules, not results.
TRY THIS: Speak their language and show what’s in it for them.

5. Data ownership is never clear
→ Everyone wants insights, no one wants accountability.
TRY THIS: Get exec buy-in and celebrate great data owners publicly.

6. It’s not about technology
→ Most failures are people and process issues, not tool gaps.
TRY THIS: Focus on the people & processes before you shop for a platform.

7. Governance is not a project
→ There’s no finish line, it’s a daily discipline.
TRY THIS: Bake it into business-as-usual rhythms and reporting.

8. Governance can feel like a solo sport
→ You’ll often be the only one pushing it forward.
TRY THIS: Find allies in Risk, Finance, or Ops and share the wins.

The road isn’t always easy, but embracing these will help you succeed.

Which truth resonates most with your experience, and what strategies have you found effective?

♻️ Repost to help someone thrive on their data governance journey.
🔔 Follow Clare Kitching for insights on unlocking value with data & AI.

Harsh truths of data governance and how great teams manage them

If 2025 was the year of talking about AI agents,
2026 will be the year they reshape how work gets done.

According to McKinsey, 23% of organisations are already scaling AI agents in at least one business unit. That number will climb fast. 
In 2026, you won’t just hear about agents, you’ll be expected to make decisions about them.

So what are they?
An AI agent is not just a chatbot. It’s not traditional automation.

It’s software that can plan, decide and execute work end to end. 
Agents retain context, act across systems, and can operate without constant human prompting.

For executives, this changes the leadership task.
From approving tech to setting guardrails for decision making at machine speed.

Done well, agents unlock productivity, faster execution and better customer outcomes.
Done poorly, they introduce new operational and risk blind spots.

This is why leaders need a working mental model of AI agents now, not later.
What they are. 
What they are not. 
Where they create value. 
And where human accountability must stay firmly in place.

You don’t need to be technical to lead in this space. But you do need clarity.

This infographic is a simple starting point. 
A shared language for leaders who want to unlock value while staying in control.

Because the future of work won’t wait for us to catch up.

What’s the one part of AI agents your leadership team is least clear on today?

♻️ Repost to help someone get their head around agents.
🔔 Follow Clare Kitching for insights on unlocking value with data & AI.

One dashboard. Six executives.
Six completely different reactions.

Data teams aren’t just analysts anymore. 
They’re translators of business value.

The story behind your insights matters as much as the accuracy of your models.

When your CEO cares about revenue, show how your model drives growth.
When your CFO watches cash flow, quantify the savings.
When your COO obsesses over efficiency, track process gains.
When your CHRO champions engagement, link your analytics to retention.

The same churn model can justify headcount, shift budget, or kill a product line.
The art lies in choosing the one that lands.

When you start with their success metric, your analysis becomes fuel for action, not decoration on a slide.

How do you tailor your team’s insights to resonate with different leaders around the table?

♻️ Repost to help someone land their story with data.
🔔 Follow Clare Kitching for insights on unlocking value with data & AI.

Metrics for the C-suite: top KPIs for every leader

The real AI risk?
It’s already inside your company.

For many companies, AI didn't arrived through strategy or careful planning.
It slipped in sideways.
Through curiosity.
Through enthusiasm.
Through a little bit of chaos.

People are trialling tools.
Teams are experimenting.
Models are being woven into processes.

And in many organisations, none of it is being monitored.
This is how risk spreads. Not through malice, but through momentum.

The answer isn’t to slow down.
It’s to put simple structure around what already exists.

Here are three moves that change everything:
1/ Map your AI landscape
See the tools, models, and use cases before they multiply.

2/ Give each system an accountable owner
Someone responsible not just for the tech, but for the outcome.

3/ Put basic guardrails in place
Clear approvals, simple documentation, and a risk check before anything goes live.

These small steps cut legal and reputational risk, protect your brand, and help teams innovate with confidence.

People move faster when they know the boundaries.

Most companies are already using AI.
But not every company is using it safely.

What’s the biggest governance gap holding back your AI work right now?

♻️ Repost to help someone manage their risk.
🔔 Follow Clare Kitching for insights on unlocking value with data & AI.

Think your data governance covers AI risks? 
Think again.

Data governance covers access, quality, privacy, lineage. 
So surely AI governance is already handled, right?

Not quite.

That assumption is why so many AI programs quietly drift into risk.

What people think AI governance is:
→ Data governance with a new name 
→ Long policies nobody reads 
→ Banning tools and adding approval layers

But AI introduces something data never did on its own:
Decision making, behaviour, and real world impact.

What AI governance is actually about:
→ Who owns real world outcomes, not just datasets 
→ Monitoring what happens after launch
→ Simple checks before release, not heavy gates everywhere
→ Enabling human intervention
→ Helping teams know what’s ok to use
→ Protecting data and IP from misuse

Data governance is necessary. But it’s not sufficient.

If you rely on data governance alone, you govern inputs, not outcomes.

Which part of AI governance do you think gets most overlooked today?

♻️ Repost to help someone accelerate their AI journey.
🔔 Follow Clare Kitching for insights on unlocking value with data & AI.
🍭 Want more from me? Sign up for my newsletter here: https://lnkd.in/giQ3b6Fi

What people think AI governance is about vs. What AI governance is actually about

taplio

The Battle

Nikolay Valiotti

Clare Kitching

Followers

Posts

Engagements

Likes

Performance ScoreNikolay Valiotti's LinkedIn Performance

Performance ScoreClare Kitching's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Nikolay Valiotti

Clare Kitching

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Nikolay Valiotti's LinkedIn Performance

Performance ScoreClare Kitching's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreNikolay Valiotti's LinkedIn Performance