Consulting firms are in a dilemma about how to organise their data & analytics capabilities most effectively: Should they have a stand-alone business unit, or should this expertise be embedded right across the organisation?..., writes Fiona Czerniawska of Source Global Research.

The former has the advantage of signalling to clients that it’s an area of investment for the firm concerned, but the latter will help the firm differentiate itself on a tactical basis, by using evidence-based insights in proposals and in all its consulting work.

But this debate isn’t confined to consultants: It’s mirrored in their clients too. Earlier this year our Intelligent Analytics report featured several case studies from clients. “We're a small central team,” said the head of analytics in a French insurance company. “Up until a year ago, we reported to the CIO, but now we're part of the marketing and digital innovation function team.” And moving out of the IT function is only part of the organisational shift going on: There’s also pressure to take a more integrated approach to the way analytics work is carried out. One oil company talked about the need to bring together its different data “lakes,” which had evolved as the business had grown in scale and complexity: “The opportunity—and challenge—going forwards is to consolidate our different data lakes. People are starting to recognise the benefits of doing this: They’re waking up to the potential value of the data we have amassed.” Another head of analytics commented, “Many business units perform very specific groups of tasks well, but, in order to grow, the organisation as a whole needs to step back and take a more holistic view of what its true business needs really are. Analytics can provide them the insight to do this.” The analytics function is coming out of the IT function and is establishing a separate identity for itself. Some organisations are going further, spreading analytics capabilities across different business units in an attempt to embed this new way of working.

There are parallels here with the evolution of the IT function back in the early 1990s. Simultaneously frustrated by the latter’s chronic inability to keep pace with the needs of the rest of the business and attracted by the easy opportunities new technology seemingly offered, business units started to take matters into their own hands, appointing their own IT experts. These autonomous mini IT functions were fast moving and flexible, but they didn’t pay attention to central guidelines and increased overall costs and complexity. Unable to solve the issue, CXOs outsourced it. Marry in haste, repent in leisure; this saying could equally be applied to the mammoth, 10-year deals of the period. Today’s picture is far more nuanced: Outsourcing still happens but often on a smaller scale; organisations buy in more, and buy in more specific services, which they oversee more actively.

Where analytics is concerned, it seems likely that we’ll see more data scientists being embedded in the business. Central teams will still be set up and will continue to operate, but there’s a danger that their role will become marginalised and that the ensuing lack of engagement between the different pockets of expertise will lead to poorly coordinated work. The result will be a less integrated approach to analytics, and also a more expensive one.

But will CXOs go so far as outsourcing their analytics function? We think the answer is both yes and no. Yes, in the sense that they will be looking for an alternative way of working and will be looking to save money, but no in the sense that the type of work involved is very different. Analytics requires much more expertise and creativity than, say, applications management outsourcing does, where most of the work is process-driven. It’s also inextricably linked to business skills: You can’t write algorithms for customers you’re not familiar with, for instance. Most importantly, many senior executives see analytics capability as a core skill their people need, not as an administrative burden they should be relieved of. Dealing with the proliferation of analytics functions and the costs they generate means that CXOs will certainly be looking to save money, but they’re going to need a new approach to solving the problem. Which is where the new version of managed services we’ve been writing about will come in. By replacing some of the work traditionally done by expensive management consultants with smart technology and by supplementing this with proprietary data, consulting firms are finding new, better ways to offer clients ongoing support.

Our research shows that it’s a model clients find hugely attractive, so why don’t we apply it to analytics work?


Fiona Czerniawska is a leading commentator on the consulting industry and a co-Founder of Source who provide specialist research on the management consulting market to consultants and their clients.