In boardrooms across the world, executives routinely encounter a peculiar phenomenon: decisions that no one claims to have made, outcomes that no one intended, and accountability that evaporates into organizational structure. Dan Davies’ *The Unaccountability Machine: Why Big Systems Make Terrible Decisions – and How the World Lost Its Mind* offers the most compelling framework I’ve encountered for understanding this dysfunction. Yet for all its diagnostic excellence, the book ultimately disappointed me in the arena that matters most to practitioners: what to actually do about it.

Davies, a former Bank of England economist and investment analyst, taps into the discipline of management cybernetics – particularly the work of British theorist Stafford Beer – to explain why modern institutions systematically generate outcomes everyone involved claims not to want. His central concept, the “accountability sink,” describes systems where decision-making power becomes so diffused through processes, metrics, and rulebooks that identifying responsibility becomes impossible. The insight is immediately recognizable to anyone who has watched a promising initiative die in the gears of corporate bureaucracy or seen critical feedback disappear into a metrics dashboard without changing behavior.

In my opinion, the book’s greatest strength lies in its early chapters, where Davies methodically builds a theoretical framework for analyzing organizational pathology. By introducing Stafford Beer’s Viable System Model (VSM), he provides executives with a genuinely useful diagnostic tool.

Beer’s model conceptualizes organizations as information-processing systems with five necessary subsystems. System 1 represents operations – the people actually doing the work. System 2 handles coordination and conflict resolution between operational units. System 3 optimizes current operations. System 4 scans the environment and plans for the future. System 5, the “identity” function, makes the fundamental trade-offs between present optimization and future adaptation, defining what the organization ultimately stands for.

What makes this framework powerful is not its elegance (though it has that) but its explanatory capacity for dysfunction I’ve witnessed often in consulting engagements. When companies struggle to adapt despite obvious market signals, it’s often because System 4 (strategic planning) has little bite or become disconnected from System 3 (operations). When employee feedback about customer pain points never reaches decision-makers, the problem isn’t individual negligence – it’s a broken information architecture that Beer’s model makes visible.

Davies demonstrates particular insight in explaining the “Requisite Variety” principle – the idea that a regulator must have at least as much complexity as the system it regulates. In practical terms, this means that any management team lacking information-handling capacity commensurate with their organization’s complexity will eventually lose control. The implications for organizational design are profound: we cannot manage modern complexity by simplifying management information. We must either reduce genuine complexity or expand management capacity to match it.

Perhaps the book’s most valuable contribution for senior executives comes in Davies’ dissection of how accounting methods – and more broadly, our obsession with quantification – actively distort decision-making. This analysis cuts deeper than the familiar critique of “managing to the metric” because Davies traces the problem to fundamental issues in how organizations convert complex reality into information.

The critique centers on what happens when companies adopt a single overriding objective – typically shareholder value maximization – and then design measurement systems around it. These systems don’t just narrow focus; they actively filter out feedback that doesn’t align with the chosen metric. Transfer pricing creates phantom profit centers that distort resource allocation. KPI cascades transform strategy into a game of hitting numbers rather than creating value. Performance management systems incentivize gaming the metrics rather than improving underlying performance.

I’ve seen this dynamic play out repeatedly in client engagements. A retail chain measures store managers on sales per square foot, then wonders why customer satisfaction plummets as managers cut service staff. A software company tracks lines of code written, then faces maintenance nightmares from rushed, poorly-designed systems. A healthcare system optimizes for patient throughput, then confronts preventable readmissions because discharge planning received no attention.

Davies argues persuasively that these aren’t implementation failures but category errors. The problem isn’t that we chose the wrong metrics; it’s that we believed a single metric could capture organizational purpose. When accounting becomes the primary information system rather than one input among many, it transforms from a tool for understanding reality into a substitute for reality itself – one that systematically excludes the qualitative, the contextual, and the human.

This connects to Davies’ broader argument about Milton Friedman’s shareholder value doctrine. By collapsing organizational purpose into a single dimension, Friedmanism didn’t just change corporate priorities – it created permission to dismantle the very information-processing systems (Beer’s System 2 and System 4 functions) that would reveal the doctrine’s limitations. Companies engaged in what Davies memorably describes as “amateur brain surgery, hacking away bits of their regulatory and information-handling system, to see if they could do without them.”

Although I think this perspective is perhaps a little extreme and understates the value that has accrued from the Friedman approach, for anyone who has sat through a private equity-backed “optimization” that stripped out middle management and replaced human judgment with automated decision rules, this analysis will resonate. The resulting organizations may be more “efficient” by narrow financial metrics while being fundamentally less capable of adaptation, learning, or accountability.

After 200 pages of incisive diagnosis, Davies’ final chapter on “What Is to Be Done?” lands with a thud. He proposes three main reforms: corporations should adopt purposes beyond shareholder value maximization, we should limit the use of debt and leveraged buyouts to discipline management, and we should invest in rebuilding organizational information-handling capacity.

These recommendations aren’t wrong – they’re simply inadequate to the scale and sophistication of the problem Davies has diagnosed. The first two are essentially political projects requiring regulatory change or cultural transformation at the level of capital markets. While I’m sympathetic to the goals, Davies provides no roadmap for how executives operating within current constraints might move toward them. The advice amounts to “convince society to change capitalism,” which, however desirable, offers little guidance for Monday morning.

The third recommendation – rebuilding information systems and middle management – comes closest to actionable guidance but remains frustratingly abstract. Which specific practices should executives adopt? How do you make the case for these investments to boards focused on quarterly earnings? What does good information architecture look like in a 21st-century organization versus Beer’s 1970s manufacturing examples?

Davies seems aware of this gap, occasionally acknowledging that “there is no algorithm” for applying cybernetic principles. But this concession feels like an abdication. The book needed either concrete case studies of companies successfully implementing these principles or at minimum a framework for diagnosis and intervention that executives could adapt to their contexts – maybe that could become a companion book as a follow up!

Moreover, the recommendations display a curious naivety about organizational change. Davies implies that the primary barrier is intellectual – if executives simply understood cybernetics, they would implement better systems. This ignores the political economy of modern corporations, where power derives from control over resources and information, where management consultants profit from “transformations” that preserve existing power structures, and where meaningful accountability threatens entrenched interests.

Verdict: Worthwhile Reading Despite Its Limitations

*The Unaccountability Machine* succeeds brilliantly as diagnostic and fails as prescription. For C-suite executives, board members, and organizational designers, it offers an invaluable lens for understanding why well-intentioned people trapped in dysfunctional systems produce outcomes no one wants. The cybernetic framework, particularly the Viable System Model, deserves a place in the conceptual toolkit of anyone responsible for organizational design.

The accounting critique alone justifies the book’s place on executive reading lists. In an era where ESG metrics, OKRs, and dashboard analytics proliferate, Davies’ warning about the dangers of metric reductionism could not be more timely.

But readers expecting a roadmap for organizational transformation will be disappointed. The gap between Davies’ sophisticated diagnosis and his underdeveloped prescriptions suggests that understanding why systems fail may be easier than designing systems that succeed. That doesn’t diminish the value of understanding – it just reminds us that diagnosis, however strong, is only the first step toward cure.

Book review: “The Unaccountability Machine – why big systems make terrible decisions” was originally published in Just another blog for kicks on Medium, where people are continuing the conversation by highlighting and responding to this story.

It has been a while since I have read an enterprise architecture practice book and a very long time since I read one I enjoyed. For whatever reason, this genre of book tends to be either high framework or unstructured case studies wrapped up up in a concept or two.. “Architecture for flow: adaptive systems with domain driven design, Wardley mapping and team topologies”, by Susanne Kaiser, may not have the most promising of titles, but balances the content mix between concept and application well and is well worth a read if you are involved in enterprise wide technology transformation.

Do as really need another book in this space? I am certainly not anti-TOGAF, ISO-42010 and all of those stakeholder views from “classical” enterprise architecture, but I find that as frameworks they are typically one step too far removed from the context of the problem domain to do anything but frame and describe the architecture, so fail to support the architect with appropriate patterns and directional support. Similarly, despite – perhaps somewhat unfashionably today – still being convinced of most of the underpinnings of agile architecture approaches, I find they too often underplay the role of more systemic thinking models. Kaiser’s book deftly addresses both concerns by building on the open agile architecture model and thinking around domain driven design with the superbly simple, yet frustratingly challenging Wardley Mapping approach and the now classic Team Topologies models for organising and optimising for flow in software engineering.

I was really impressed by two facets of this book:

1. Kaiser managed to bring together the models into an authentic whole, which shows how you can use Wardley Maps to understand the strategic landscape of an organisation, identify differentiators for product and service design and then decompose these into an actionable set of architectural components using domain-driven design techniques and – crucially – to then make sense of what to build, what to outsource, what to architect for continuous change and what to invest in for the future
2. In parallel, she creates a coherent bridge between Wardley’s culture mindsets ( “explorers”, “villagers” and “town- planners”) and Skelton and Pais’ team topologies, recognising that they are related but adjacent concepts rather than just smashing them together – arguably, this could create a more useful framing for scaled agile organisational operating models than the eponymous “SAFe” equivalent!

There is not much to criticise in this book, in my opinion, but I think that there is real potential to build on this with further work and writing to expand the applicability of the organising model and the domain driven design more comprehensively out of the IT “enterprise” architecture niche to the true business domain – Wardley maps are already firmly out there well beyond the technology space for systems thinkers and strategists willing to invest some time, but there is much from “DDD” and team topologies that could and should be permeating the commercial and business design and strategy units of organisations. A fantastic companion book would be a workbook with more scaled examples of application; the contrived model single product organisation running events brings the models to life but is still quite a way from most messy enterprise multi-product brownfield operations!

I very much look forward to the continuation of this series!

Originally published on GoodReads: https://www.goodreads.com/review/show/8191230914

Architecture for flow: a book review was originally published in Just another blog for kicks on Medium, where people are continuing the conversation by highlighting and responding to this story.

Originally posted on GoodReads here

When I was scanning the contents table for this new book on the emerging field of agentic AI, the sections on the “memory” that modern organisations need to build – semantic, episodic, procedural – really stuck in my head. It’s a small thing, but it caught my eye. Not because it’s a unique perspective (it’s not), but because it signals something rare in the current flood of AI books: a genuine attempt to anchor agentic AI in the messy reality of enterprise architecture.

Bornet, Wirtz, and Davenport don’t appear to be the breathless hype types, like too many of the authors on this topic. They’ve been working in the (sadly too often not so) “intelligent” automation and robotic process automation (RPA) field for two decades, and their perspective is deeply grounded in the trenches – the kind where business cases are written in Excel, not philosophy. And that’s both the strength and in my view the subtle limitation of the book.

The strength is obvious: this is a practical, structured guide to how agent-based automation might actually work inside large, often dysfunctional organisations. The book presents a useful maturity model for agentic systems, walking through stages from simple rule-followers to autonomous collaborators. It’s peppered with real-world examples – some a bit too glowing to be fully trusted, but still illustrative. For CIOs, CTOs, and transformation leaders trying to navigate the gap between aspiration and execution, there’s value in this operational framing.

I particularly appreciated the clear process models and architectural considerations. The authors treat agents as a new class of enterprise “worker,” and explore what kind of infrastructure (technical and organisational) is required to support them. The book is packed full of useful two-by-two decision matrices, maturity models and phased roadmaps. There’s an implicit call to action here: stop slapping AI on the side of your processes and start thinking about how autonomous systems reshape your whole work design AND here is how you can get there.

Where the book is less compelling is in its treatment of the broader theoretical and ethical landscape. You won’t find much discussion of agency in the philosophical sense, nor a deep dive into the sociotechnical implications of humans and agents working together. It’s not that kind of book, although they do try to address these concerns – and that’s okay; I would go elsewhere for an elevated assessment of these concerns.

What this book is, is a playbook. A solid, well-structured, occasionally acronym-heavy playbook that helps leaders frame the messy middle of AI adoption. It’s not speculative, but it is strategic – in the “how do we make this work on Monday” sense of the word.

If you’re in the business of advising clients, leading tech strategy, or just trying to wrangle coherence out of a dozen half-baked AI initiatives, this book is well worth the time. Its focus is on what to do – and for most in the business, that’s exactly what is needed right now.

Book review - Agentic Artificial Intelligence: pragmatic and structured playbook was originally published in Just another blog for kicks on Medium, where people are continuing the conversation by highlighting and responding to this story.

I had an opportunity over Christmas to catch up on some overdue reading and, of course, it had to be AI.

First up was “Co-intelligence” by Professor Ethan Mollick. It was a worthwhile and well structured read. Being a techno-optimist, I had already bought into the basic premise of the book, that AI will “augment” our own intelligence and that we will need to adapt to collaborating with AI.

It is a sign of how fast the industry is moving and how quickly we adapt as consumers, that this book, published only in April 2024, in parts feels a little bit “dated”. In the first part of the book, Mollick outlines 4 principles for working with AI:

• 1. Always invite AI to the table
• 2. Be the human in the loop
• 3. Treat AI like a person (but tell it what kind of person it is)
• 4. Assume this is the worst AI you will ever use

For anyone who has, in the meantime, spent a bit of time with ChatGPT, Google Gemini or any of the other LLM chat interfaces this will be well-trodden and understood territory, although it remains good advice!

I enjoyed the second part rather more; thinking of AI as a collaborative agent and then working through the different roles it might have in that collaboration is both intuitive and effective. AI “as a person”; AI as a “creative”; as a “co-worker”; as a “tutor”; as a “coach”. Obvious perhaps, but Mollick provides useful insights and practical thoughts on how considering these different roles can inform how we can get the most out of it.

A thoughtful and optimistic assessment of the status quo.

Next up, Superagency, by Reid Hoffman, another leader in the field…

Book review — Co-intelligence; living and working with AI was originally published in Just another blog for kicks on Medium, where people are continuing the conversation by highlighting and responding to this story.

As a father of two young daughters, aged 8 and 11, I’m always on the lookout for experiences that can spark their curiosity in the world around them. Recently, we visited the Science Museum’s Power Up exhibition — a celebration of the evolution of gaming and its impact on technology. As we explored the interactive displays, I couldn’t help but feel a sense of nostalgia. I was reminded of my own childhood in the 1980s, sitting at a BBC B computer, painstakingly typing out game programs from a magazine, line by line, just to play them.

Back then, gaming wasn’t just about entertainment; it was a gateway into understanding how technology worked. In fact, many of us who were drawn into the world of computers and programming did so because of the games we played — or rather, the games we built or modified ourselves. My father, a teacher, had a discount on the BBC B, which made it accessible to us, and that simple machine became my first introduction to the world of code. Typing in those programs taught me about syntax, debugging, and the satisfaction of creating something that worked.

Fast forward to today, and the Power Up exhibition highlights how gaming continues to be a powerful tool for inspiring young people to engage with the underlying technologies. Whether through the complex hardware that powers gaming consoles or the intricate software that drives immersive experiences, games encourage exploration and innovation. For my daughters, playing and interacting with the retro and modern games at the exhibition sparked questions about how these systems work. This is the magic of gaming — it breaks down barriers between entertainment and learning, making technology fun and accessible.

In an age where digital literacy is increasingly important, gaming offers a natural pathway for young people to develop problem-solving skills and an interest in STEM (Science, Technology, Engineering, and Mathematics) subjects. Exhibitions like Power Up not only celebrate the history of gaming but also showcase how far we’ve come — and how much more there is to explore. From coding simple games to building complex AI-driven experiences, the connection between gaming and technological development is undeniable.

For me, the BBC B was just the beginning. That early exposure to coding led me down a path that shaped my career in technology. As a CTO today, I see gaming as a tool not just for entertainment but for empowerment — something that can unlock a world of possibilities for the next generation. If my daughters’ excitement after the exhibition is anything to go by, the next wave of innovators may very well be starting their journeys in front of a screen, just as we did.

If you haven’t already, I recommend visiting the Power Up exhibition at the Science Museum. Whether you’re a parent, educator, or simply a lover of gaming nostalgia, it’s a nice reminder of how games have shaped and continue to influence our relationship with technology… even if it is mostly an opportunity to play on some of your favourites “oldies” on the original machines!

On Gaming Nostalgia and the Link Between Games and Technology Engagement: A Visit to the Science… was originally published in Just another blog for kicks on Medium, where people are continuing the conversation by highlighting and responding to this story.

Having loved Tim Marshall’s previous books on geography, I was delighted to see the familiar branding at the airport ahead of a 13 hour flight to Singapore… and I wasn’t disappointed: a truly eye-opening perspective on the politics of space and the urgent issues that we will inevitably face.

As in previous books, Marshall has a gift for bringing the topic to life in an especially concrete way, connecting the physical with the conceptual so that some pretty complex geo-political challenges become straightforward common sense results of the environment we inhabit (or in this case may inhabit in the future).

I found the juxtaposition of the more recent Chinese and US space missions with their very different philosophical underpinnings fascinating, but I found Marshall strongest when he was bringing to life how many questions of “astropolitics” are unanswered with the legislative and political frameworks we have today. Perhaps naively, I had assumed a level of consensus and — perhaps more importantly — codification in our space endeavours which appears not to exist, even at a pretty primitive level.

Will it be Star Wars, Star Trek, or something darker? Who knows… but Marshall does a brilliant job here of bringing the issues, from space debris to the exploitation of the moon’s resources, to jurisdictions on Mars, and the speed with which they might become genuinely material for humanity.

Definitely worth reading. Now.

Book review — The Future of Geography, by Tim Marshall was originally published in Just another blog for kicks on Medium, where people are continuing the conversation by highlighting and responding to this story.

I picked up “Build, An unorthodox guide to making things worth making,” by Tony Fadell following a podcast recommendation by a colleague who is focused on product management.

It sounded like a nice easy summer read: It was. But it was also excellent.

Fadell obviously has incredible pedigree as the co-creator of the iPod, iPhone and Nest and his more recent activities as the founder of the Build Collective, so his thoughts were always going to be worth a scan, but I found the book to be refreshingly straight forward in tone and advice.

Conversational, authentic and packed full of Fadell’s life experience from early failures, the massive successes and with more reflection (for example on the painful Google acquisition of Nest) than I had expected, I found this to be a well rounded introduction to the entrepreneurial journey with lots of great actionable recommendations for every stage.

Definitely recommended reading for anyone in the business of building products and services, even if there really isn’t much in here that I would categorise as “unorthodox” – I found it energising.

Originally published on gooodreads here: https://www.goodreads.com/review/show/5779733667

Book review – Build, an unorthodox guide to making things worth making” was originally published in Just another blog for kicks on Medium, where people are continuing the conversation by highlighting and responding to this story.

Review originally captured on GoodBooks here

This is an excellent primer, showcasing the current state of good practice in corporate sustainability.

Professor Taticchi and his collaborators have created some nicely detailed and well researched case studies from across Europe, a range of industries and with transparently different levels of maturity when it comes to sustainability.

Most striking for me were the timelines from early ESG programmes (in the early 2000s) to (potentially) more impactful integrated thinking and strategy today and in particular the journey that many of the organisations have been on to avoid their programmes to be a “bolt on”. One has to hope that the next phase of acceleration will happen considerably faster.

The simplicity of the case studies and the straightforward language was refreshing on such a hyped topic and the final chapter, “10 Golden Rules to Lead the Sustainable Transformation” — though short — was pithy and compelling. Working at an organisation that is both on its own sustainability journey, but also providing a wide range of sustainability consulting and sustainable IT solutions, it was reassuring to see some strong overlap in our thinking towards alignment of corporate mission and sustainability as well as accepting (even embracing) that monetisation and competitive advantage represents a powerful motivation alongside “doing the right thing”.

If you are interested in finding out what a range of companies in different industries are really doing and how they are approaching their business around sustainability, then this is a great place to start.

Sustainable Transformation Strategy, by Paolo Taticchi et al: a book review was originally published in Just another blog for kicks on Medium, where people are continuing the conversation by highlighting and responding to this story.

I am a huge fan of Dan Roam’s work. “The back of the napkin” is still one of my favourite productivity books, and one that I recommend to new consultants (and indeed to anybody struggling to get the message across) a regular basis.

Somehow, I had missed the publication of “the pop-up pitch“ and was excited by the subtitle, “the two hour creative sprint to the most persuasive presentation of your life!“

Against the promise of its predecessor, and this enticing subtitle, I am afraid the book does not deliver. However, that does not mean that you should not read it! As always, Roam provides a simple framework and easy to follow guidance to creating a storytelling narrative for pretty much any flavour of pitch presentation. His focus on imagery and simple messaging is spot-on.

I would like to have seen more comprehensive examples, but it is an easy read with powerful takeouts.

Book review – The pop-up pitch was originally published in Just another blog for kicks on Medium, where people are continuing the conversation by highlighting and responding to this story.

Book review: “The Big Con: How the consulting industry weakens our businesses, infantalizes our governments, and warps our economies”

Originally published on GoodReads here.

As someone who has worked in professional services for my entire career and as a consultant for much of this, I was never going to agree with all of the findings of “The Big Con: How the Consulting industry weakens our businesses, infantilizes our governments and warps our economies” by the totally brilliant Mariana Mazzucato (I would recommend every book of hers that I have read so far).

That notwithstanding, this book should be mandatory reading for anyone involved in the industry: it is incredibly well researched and provides a systematic look at both the demand and supply side challenges of the growth in consultancies and the way they can impact in particular government departments, contributing through the economic system they support — despite mostly overwhelmingly good intentions — to a hollowing out of capability in these organisations. “To steer, you have to also row” is an insightful analogy and something I strongly recognise: the most impactful programmes I have worked on, leading consulting teams in both the private and public sector have been delivered more as a “peer” with some specific insights, skills, to an individual or team with a strong capacity and desire for self-sufficiency.

It is hard to argue with the systemic challenges that Professor Mazzucato highlights and she is nuanced in her positions, differentiating between traditional policy and strategy consulting and wider professional services organisations, providing augmentation and IT services.

I only really disagree with one part of the analysis; the assertion that professional services organisations are unwilling to take on end-to-end responsibility for delivery of an outcome. Although this may well be true for strategy consultancies (and even there, examples can be found), for larger scale operational delivery programmes and outsourcing, I would argue that the desire is typically there, but the willingness or ability of the procuring institution to a. enable the consultancy to control all of the parameters that allow it to own the outcome and — perhaps more critically — b. price for the risk appropriately without losing the tender are actually the key factors driving the current problematic commercial contracting structures in government.

This is, however, a minor niggle given the overall analysis. The concluding chapters, present some thoughts on how to address these problems at a systemic level, and I have to admit I would have liked a tick more on this and a little less of the painstaking analysis, as compelling as it is; hopefully there will be a companion book — given Mazzucato’s role at the UCL Institute for Innovation and Public Purpose, she is perfectly position to do drive this agenda further..

Unsurprisingly, I would hope that this will see not just a strengthening of the demand-side organisations own capacity for policy innovation and scaled delivery excellence, but also a recognition that the there *are* huge pools of outstanding talent and capacity within the industry that are willing to find different approaches to delivering value and tapping into these in ways that create an alignment of both interests and outcomes between the actors in this messy and complex problem space!

This book should be mandatory reading for anyone involved in the consulting industry

Book review: “The Big Con: How the consulting industry weakens our businesses, infantalizes our… was originally published in Just another blog for kicks on Medium, where people are continuing the conversation by highlighting and responding to this story.