In this interview series on inspiring digital technology leaders, we take an inside look at how enterprise technology leaders make systems work in some of the largest and most complex business environments in the world. Meet the Individual, Understand their Challenges, and learn about Their Vision of the Future.
An Interview with Dan Onions
Introduction
In this very insightful interview we learn about the impact of getting the early stages of a project right to increase the chances of success, and how the focus on stakeholders, and continual engagement and cooperation will create an environment of opportunity and prosperity. We also learn that we should be solving systemic problems, rather than letting the short term focus overwhelm us, and that the successful enterprises of the future will understand how to create a stable platform for rapid change.
Dan Onions is an experienced Technology Leader with a background in Enterprise Architecture and Delivery Leadership in Large Enterprise programmes, who is now helping build an Analytics Startup in the FinTech sector. Dan has had considerable experience in complex environments, including UK Government, Defence, Finance and Consulting Services, and is used to working in leadership roles on programmes with budgets in excess of £100m. Dan is also the founder of DASH — a unique agile project management app that puts results and stakeholders at the centre of a very visual model of your work planning, enabling you to focus on achieving results through better communication and relationships.
Bitesize Takeaways:
Be more stakeholder-centric, embrace business-led change, incubate change initiatives, and build up from a stable core platform
Computing will be more distributed, with more processing at the edge
Platform-as-a-service will become dominant for software development
The hardest problems to solve are systemic
Systems influence behaviour, and behaviour influences the system
Systems oriented design thinking can help society as much as enterprises
Complexity is driven by integration and adaptation of systems over time — short term thinking only makes it worse
Make time for strategy — get the ‘system’ right, and the rest will be easier
Being strategic should never be used to avoid decisions and being operational should never be used to avoid planning
Change is a constant challenge, yet most enterprises are change fatigued
A model office approach is the most effective way to engage Stakeholders
Change is hard to win support for & too easily undermined.
Stakeholder management is a key reason that many projects fail
Technology and business should be treated as one and the same
Teams must work together consistently, despite different perspectives
Agile can be productive for development, but we’re still bad at estimation
Forcing Tech use because of a Brand or a Trend is a proven way to fail
Meet the Individual
Q: Tell us a bit about what you currently do:
I am founder of DASH, a project management approach for use in critical early phases of projects where decisions get made. In any project, in any industry, early phases are about dealing with uncertainty and complex options on what to do and how to do it. In most organisations, this is ten-times as hard because decisions need agreement from multiple stakeholders, each with a different perspective. DASH stands for Decision Agility with Stakeholders and it’s about using agile project management to manage decision making with stakeholders.
Early phases of projects have many labels, such as: strategy, requirements, feasibility, architecture, analysis or design. Very often these projects are stopped before they achieve any of their objectives, because they don’t get the agreement needed. So much of the conditions that make a project successful are defined in the early stages. My goal at DASH is to help drive projects to a successful outcome by helping people make the right decisions and get agreement with stakeholders, so they move forward to implementation and beyond.
I am also Delivery Director for Quantexa, an analytics software start-up with a technology that creates a single view across data sources using entity resolution and network analytics. My role is to enable successful delivery for clients and partners. Analytics and artificial intelligence can be game-changing for organisations work if they can learn how to apply it and I am determined to make that a reality.
My roles have a common theme: enabling change through better decision making.
Q: What motivates and inspires you to get up and go to work everyday?
I have always been attracted to solving difficult problems, when others have given up or cut corners I always tried to find a way, and then do better than most thought possible. I get a thrill when I’m onto a compelling solution and helping people achieve. Ultimately, I like my work to make a meaningful difference, so if I can make an organisation perform better I get a kick out of it. Organisations are the bedrock of society, so the way they work matters.
I get inspired working with people that have a thirst to learn. It’s amazing to work with younger people with no strong preconceptions of work — with the right guidance they can achieve incredible things, because they were not told something was not possible. Mind-set and belief is the biggest factor in what makes someone talented. I have been lucky to work with very talented people over the years, and have since seen many ex-colleagues become very successful in their own businesses.
Q: Tell us about your journey to get to where you are today?
My first computer was an Amstrad (CPC6128), which was one of the first with a floppy disk drive, very popular in the mid-late 80s. I did some programming on it but really got into computers when I got a PC in the early 90s. I earned extra money building database software for small businesses in the village where I grew up.
At university, I did a degree in Maths and Computation with a year out at IBM’s Ease of Use department which focused on human-computer interactions. The year at IBM was a big learning experience for me, we used to do R&D work with web technologies and Java. The experience taught me a lot about being user focussed when designing software. After university I was determined to work in a smaller company, so I joined a start-up technology consultancy called Rubus and got involved in building web-based applications. I was very focussed on delivering projects in teams — and did several large-scale projects where I took on as much responsibility as I could and learned a huge amount about software architecture and engineering.
I have always challenged myself to become better at something each time I do it, and with projects too.
I started DASH because of challenges I saw where IT organisations and great technical people were not hitting the mark; either not focussing on customer need and often not getting their message across. DASH is all about balance between influencing and thinking.
Q: If you were to do anything other than what you do now, what would it be?
I have always had an interest in economics and modelling to drive better government policy. Economics is a big subject — and I am no expert — but part of it is looking at society as a system of people, where behaviours are driven by way the system works and the system is in turn influenced by people’s behaviours.
In my various roles I have used business architecture to design the way organisations should work — the idea of a target operating model, defined not just as a high-level picture, but with proper design principles and tests applied to show how it works. I think this kind of design thinking has a great potential to be used at government level to better society. This is a step away from technology, but at some stage I would like to work more on this.
Q: What books or podcasts do you recommend for Enterprise Technology Leaders?
“Good strategy, Bad strategy” is a great no-nonsense book about how strategy needs to be specific and actionable. In my view strategy can be both big vision and practical, and this book goes some way towards helping to illustrate this point.
There is a recent podcast called “Masters of Scale” by Reid Hoffman, founder of LinkedIn and venture capitalist. He interviews some big names in tech, such as Mark Zuckerberg and Eric Schmitt about how to apply start-up ideas at scale. It’s fascinating.
Understanding Their Challenges
Q: What do you personally spend most of your time dealing with?
I spend a lot of time supporting multiple projects, right the way through the lifecycle. I am spending more than half of my time on projects that are mobilising or going through proof of concept activities.
One of the challenges with data analytics projects is making sure all the dependencies are in place, such as environments and data — with some clients it’s relatively easy but with others the process is less predictable, especially when they are less familiar with analytics projects.
Life in start-ups is often about getting stuck into everything. Prioritisation can be hard because there is always so much to do. The balance that can be hard to achieve is between the immediate and strategic. I always try to create time for the strategic — get the systems right so the operations are smooth and can be scaled.
Q: What is the most important thing you focus on to make the organisation successful?
I like work to be as enduring as possible so establishing strategic direction helps me get future use out of day-to-day work. For instance, if I need to get a process in place I will do it quickly, but it will be a step towards what we really need, so we are not always reworking. Being strategic should never be used to avoid decisions and being operational shouldn’t mean not planning where you are going.
I tend to lead by communicating vision and trying to get key stakeholders pointing in the same direction, which is especially important when working with larger organisations with very different perspectives. In a startup, strategy needs to be very quickly followed by action.
Q: What are the most expensive mistakes you have seen made in organisations, and how did they happen?
Big organisations find it hard to change even when it’s clear they need to. When you find rare individuals in an organisation who are prepared to push change, the momentum they create is so valuable. The biggest mistakes I have seen is where this energy is squandered because of in-fighting over differences at the detail level, which can normally be worked out later.
The hardest problems to solve are systemic — where teams can seem to be performing well in isolation, but the system overall produces poor outcomes. It takes a change of the overall organisational system to create a meaningful improvement, but this is often a huge challenge because teams have lost trust in each other and so focus on keeping control. Often it takes a real crisis to force change.
I have seen a number of mistakes relating to selection of enterprise technology. On one particular project, early in my career, the organisation had done an enterprise deal with a vendor. We were implementing a large web application and there was huge pressure to use products from this vendor even when they were only a fit on the surface. In particular there was a workflow product that we had to build around to make it usable, which meant we ended up building something much more complex, which ended up being removed.
On projects, there is a tendency to avoid difficult decisions or difficult stakeholders in order to keep progress looking smooth, but often means deferring the pain until later. Stakeholder management is the poor cousin in project management and yet it’s a key reason that many projects fail.
Q: What drives the complexity in your Enterprise Systems?
It’s the same story as any enterprise, complexity is driven by integration and adaptation of software products to meet requirements.
Enterprises are reasonably complex by nature, even in a start-up we face choices that involve trade-offs. Ideally, a single product meets all enterprise systems, but there is the normal trade-off between cost and features, and then another trade-off with integration complexity when you go for separate applications. Going for a single product often seems simpler, but it can create hidden cost if the integration points are not understood. It’s much better to plan upfront and work out how systems need to integrate in advance.
The challenge in large organisations is they add new IT systems because they often can’t face turning off the old. This is often because change projects naturally look to reduce delivery risk, so adding a bolt-on system is lower risk than replacing a critical system. It’s reasonable for there to be projects that only add capability, but there needs to be programmes that actively manage integration and drive application consolidation.
Q: What are the most frustrating challenges when designing, building and managing enterprise systems?
One of the key challenges during the design phases is getting different disciplines to work together productively. You have business analysts, architects and developers all with different views on the end result and how deliverables fit together. If not carefully managed then there ends up being mismatches when you get into the build phase. What can help is small multi-disciplinary teams to encourage people to cross boundaries.
Agile development practices have become very productive for managing development, but there are still huge challenges in estimating projects upfront to meet the need to fit with budgets.
I think many larger organisations are change fatigued, with so many previous attempts at change people lack the motivation to try again. In these circumstances, one of the biggest challenges is gaining the trust of stakeholders that change is achievable.
Q: What has most helped to drive the success of your change programmes?
Change is one of biggest challenges for an organisation, because you are often asking people for faith that everything will be ok in the new world. One of the most powerful ways to engage with stakeholders is to use working demonstrations of software that is configured with their own data to show them how it will work. I often call this a model office approach to transformation — where you regularly get end-users into an environment where they are running through scenarios so they can experience how life will be after the change, but also to confirm requirements. This approach works across the lifecycle, in the early stages this is a demo, but then it becomes acceptance testing, training and then rollout. The idea to gain trust and create dialogue with users.
Q: What has most negatively impacted your change programmes?
I have found battles between business and IT departments are normally the biggest issue. I have seen good IT initiatives that have failed because the business is not ready, and I have seen great business-led initiatives that have been stopped by IT. Meaningful change can only work if everyone is pulling in the same direction. Change is hard to win support for and too easily undermined.
Q: What is currently holding you back from having the tech/engineering credibility of a Google/Facebook/Amazon?
It is amazing what these companies have achieved by hiring talented people and creating an environment where they flourish. I think what holds back most traditional companies is not understanding the technology enough to make the investments.
I spent some time recently encouraging a large company to invest in developing a software technology for distributed computing, but although everyone could see the need, ultimately, they were not confident enough to put in the resources. It would have meant a variation on the way they normally operate. The advantage the tech giants have is the business leaders understand the technology — they became successful by betting on change and so are willing to do it again.
We have very strong tech capability in my companies, which are start-ups that have solved some very challenging engineering problems. Like most companies we benefit from the ideas and open source software so we are standing on the shoulders of giants. I hope we can build something of significant enough scale and value to be widely recognised. My companies are still young! Hopefully we will get it right!
Their Vision for the Future
Q: What are the 3 key changes that Technology Leaders should focus in the modern world?
I think a lot of technology leadership is about fundamentals. If I were to pick three areas it would be as follows:
Develop the next generation of talent. Becoming a senior technologist takes a lot of learning from practical experience, but that learning can really be accelerated by providing the right guidance.
Technologists need to be prepared to follow the needs of business leaders. It’s as important to follow as it is to lead, and focussing on the needs of business stakeholders helps build trust to achieve greater things. Most technology is in support of the business, and technology leaders need to be seen as part of the business leadership team where they can add value.
Focus on business change. Infrastructure has dominated the thinking of many IT departments, but with cloud this has become commodity, which means the focus should be on business value. Most business change is technology enabled, so there is demand for better services. Companies that do not make this shift risk being open to competitors that do.
Q: What is one thing about the future of technology that you believe in, but most people generally don’t know, or disagree with?
With cloud, we are seeing a big shift towards centralised computing. However, soon after we are going to see the opposite effect: computing will become more distributed and with-it software will need to be smarter.
There is a huge growth in the internet of things and smart devices, so as the number of devices grow the amount of data being collected will create a need for smarter distributed computing where more data processing and analytics will happen at the edge of the network without always relying on a cloud connection.
Outside of cities internet connections are still quite poor, especially in less developed countries, even though demand for services is there. Distributed software will solve this through intelligent sharing of information through local peer-to-peer connections.
There is a lot of hype about artificial intelligence (AI), but most enterprises will not directly implement AI, they will buy products that have existing analytical models. Machine learning needs many data points to train a decent model, so it needs to be applied to areas that generate loads of data and are highly repeatable.
Q: How can we build better technology organisations?
Be more stakeholder-centric: we need to better engage with key stakeholders within the organisation, and understand their desired outcomes. Change initiatives at all levels need to become better at connecting with stakeholders and help drive change forward.
Embrace business-led change: when business managers want to change their energy should be embraced and directed. It’s all too common for IT departments to shut down ideas they do not control, which loses trust and frustrates motivated business users.
Incubate change initiatives: the change portfolio should be subject to incubation where ideas are developed and tested in a supportive environment. The objective is to reduce the risk and get support prior to a full implementation project. Incubation should be able to happen for both formal projects and business-led initiatives — the best projects can be brought into more formal management once fully defined and risk is reduced.
Stable core platform: if data is accessible via a core platform (using open APIs) then it reduces the cost of IT projects looking to innovate. In fact, Facebook allows its developers to innovate and change by having stable infrastructure on which they can rely.
Q: What do Enterprise Systems of the future look like?
Platform as a service will become the dominant way software is developed and deployed, using containers. The modularity of containers and technology agnostic nature of them is such a strong deployment model that it is bound to dominate. This means that software vendors will support them more and more as the platform to run their software. Mobile devices will start to use containers in some form to distribute apps in a common way, since the model is highly attractive in a distributed environment.
Software is going to become more distributed. At the moment most software, even on mobile devices is reliant on centralised infrastructure. Devices will get smarter at peer to peer connections, which will enable more seamless integration across IoT and smart mobile devices.
I think there is potential for using AI within enterprise integration packages to automatically map and distribute data across the enterprise. This means less effort to reuse data.
Dan kindly provided his time, knowledge and insights in this interview as part of my research for an up-coming book: Mastering Digital: how technology leaders, architects, and engineers build the Digital Enterprises of the Future
You can contact Dan via LinkedIn. You can find out more about the DASH App and approach at: http://dash.pm/