Chris Sporleder is a managing partner at Rewire, a data and AI consulting firm with offices in Amsterdam, Heidelberg, and Tel Aviv. He’s responsible for developing client relationships in Germany, Austria and Switzerland. Rewire’s DACH operations provide end-to-end data and AI transformation, from strategy through implementation and change management.
In our latest post, Chris explains how businesses can benefit from strategic data & AI consulting. How can business leaders get the most from AI and data consulting projects? What should leaders look for in a consulting partner, and what factors will ensure optimum consulting-client collaboration? Chris also reveals the most common challenges businesses face when embarking on new data & AI initiatives, and offers key tips for avoiding them:
My journey started in data and analytics. My first role was with a utility company where I analysed mainframe data and even helped calculate the company’s first wind park. I was fascinated by analytics, and that passion led me to SAS, a leading data analytics company, where I was involved in building out their professional services division, which ultimately led me into consulting.
With the advent of cloud computing, the consulting landscape shifted dramatically. Previously, we focused on explaining the value of data analytics, but cloud technology has enabled us to operationalise insights in ways that weren’t previously possible. This evolution has been incredibly exciting.
The internet was just emerging then, but many industries were already familiar with using data, such as insurance actuarial work. For those in other sectors, we emphasised modernisation and the potential of new algorithms to enhance insights. But with today’s GenAI advancements, there’s far less emphasis on convincing people of data’s value – now the focus is on how to execute and achieve impact.
Cloud computing was a major leap, especially in handling and integrating data from diverse sources. Thistransformation was also driven by IoT developments, connecting millions of devices and greatly expanding the data we could analyse, from customer behaviours to complex production processes.
Data management remains a challenge. We’ve evolved from data warehouses – low-tech but high-governance – to data lakes, which brought high-tech but low-governance, often creating data chaos. Now, with concepts like data mesh, we’re seeing a more balanced approach that combines technology with governance. Yet, engaging business leaders in data management is still an ongoing challenge.
You have to differentiate between types of consulting services – there isn’t just one type, but many different flavours and shapes. It’s a spectrum:
On one end, you have the ‘body leasers’ – you order three units of data scientists, and that’s exactly what you get, but then you have to direct their work.
On the other end, you have strategy consulting firms – the peak of the mountain, high-performing and fastmoving consulting firms.
In between, you have system integrators who lean more toward the first category, and boutique firms that position themselves in the middle or closer to the strategy end.
You need to identify the right type of consulting for your current state and engage with the appropriate firm. It’s rarely one provider from start to finish – you’ll likely need different types of services at different points in your maturity journey.
Choosing the right consulting partner is crucial. For instance, strategy consultants excel at high-speed, impactful short-term engagements but may require substantial internal resources to keep pace. In contrast, system integrators operate within very specific project parameters, and body leasing can lead to dependency on individual freelancers. Each approach has its place, but the fit depends on the project’s goals.
Yes, I do. However, you need to be aware that data and AI initiatives can originate from different points in organisations:
Sometimes you have an enthusiastic board member who believes in it unconditionally. That’s a brilliant starting point, and your strategy becomes more operational since you don’t need to convince anyone at the board level. You can focus on implementation and achieving impact. If you don’t have that convinced board member, you first need to:
Yes, that does happen and it’s not so much that the overall objectives fall apart, but rather the specific scoping of the solution that can be problematic.
What often happens is that there’s a complex challenge, like developing a ‘copilot’ based on generative AI. A copilot is typically not a single use case, but rather multiple use cases bundled together. Deciding how to slice and scope that effort requires very conscious choices:
These design choices are critical, and sometimes the solution being developed can end up not fully aligned with the organisation’s strategic direction. It’s very important to identify and fix any such misalignment very early on.
While less common in data and AI engagements compared to other types of consulting work, this challenge can still arise. The key is to maintain tight alignment between the solution being built and the overarching business goals throughout the project.
There are very common challenges, particularly around the expectation versus reality gap. The first step in capability building is understanding the art of the possible. There are a lot of expectations, especially now with GenAI, about what it can and cannot do. Spending time understanding the possibilities is crucial to avoid mismatched expectations.
A second major challenge is that organisations frequently underestimate the infrastructure and data management requirements needed to produce an AI use case. Transformations can fail in different directions: Some organisations overshoot by investing heavily in technology and data first. I’ve seen programs spend three years working on data without deploying a single use case. On the other extreme, some programs deploy three use cases without considering how to build a common data foundation or infrastructure.
Both extremes should be avoided. The right approach is to think about all dimensions and develop your infrastructure, data foundation, and organisational capabilities alongside the use cases that create impact. Recently, I posted about the end of the ‘lighthouses’ phenomenon. We’ve seen many organisations building exciting lighthouse projects by different groups, but with no common foundation or connection between them. While individual lighthouses may be stable, they don’t help the organisation scale data and AI because there’s no scaling plan behind them.
There are several key considerations:
This is specifically about data and AI rather than general procurement. Data and AI projects have significant dynamics across different dimensions, similar to the early days of online marketing – you often end up somewhere different from where you initially planned. This doesn’t mean you’re not creating impact; you may just take a different path to achieve it.
For procurement, this means you shouldn’t engage data and AI consultancies in the traditional way you would work with a system integrator. Strict, tight statements of work can actually hinder your success.
You need alignment across three domains: business, leadership, and technology. Business leaders provide insight into core needs, leadership drives organisational changes, and the tech team delivers solutions. These elements together form a strong foundation for impactful AI projects.
Essential. Successful AI initiatives integrate AI into processes, often involving workflow changes, capability-building, and sometimes even role redefinitions. Leadership commitment is key to driving these changes. You cannot implement data and AI at scale without senior leadership commitment. Without it, initiatives will remain siloed or isolated in specific departments. Everyone says AI will change the world and transform every process, but that will only happen if your most senior leadership not only supports it, but is convinced that this is a new enterprise capability.
Everyone says AI will change the world and transform every process, but that will only happen if your most senior leadership not only supports it, but is convinced that this is a new enterprise capability.
I see a few key trends emerging:
INFRASTRUCTURE AND TECHNOLOGY:
INTEGRATING CONVENTIONAL AI AND GENERATIVE AI (AGENT NETWORKS):
There will be a lot of discussion around how to integrate conventional AI models (e.g., for risk scoring) with the new wave of generative AI models (language, image, video). Creating end-to-end processes that leverage both will be an important next step.
DATA MESH:
While a full data mesh implementation can be highly complex and time-consuming, organisations can take a more pragmatic approach. Focusing on data product-centric thinking and selectively adopting certain data mesh components that can be implemented quickly, is a smart way to progress.
Many still underestimate the infrastructure needed to support AI. Additionally, trying to shortcut foundational work in data management or infrastructure will hinder sustainable AI growth.
Consulting offers unparalleled learning opportunities. Every client and project is unique, exposing you to varied business challenges and technologies. It’s an excellent choice for anyone seeking continuous growth and diverse experiences.
Rewire not only help clients identify impactful areas but also support them in capability building and integrating AI into business processes. The Rewire approach is a hybrid between strategy consulting and implementation. Our aim is to drive impact collaboratively with clients, adapting engagement models as projects evolve. Our team’s strength is in balancing high-level strategy with hands-on execution and capability building.
