Has AI Already Made Consultants Redundant?

Episode #372 Has AI Already Made Consultants Redundant?

AN INDUSTRY RIPE FOR DISRUPTION

Of all the industries that are on the chalkboard for major disruption by AI, the consulting sector must be high up in the batting order, along with accounting and legal firms.

So, wouldn’t you think that right now, consulting firms would be keener than ever to demonstrate their value, and extend their longevity? Anything they could do to prolong the mystique that surrounds what they do would probably be worthwhile, right!?

Imagine my surprise, then, to read about the report that Deloitte submitted (with a straight face, apparently) to an Australian Federal Government department, which had been largely generated by AI. There it was in all its glory, resplendent with phantom references, typos, and other AI hallucinations.

And they charged the client $440,000, a sum of money, which no doubt quickly succumbed to the gravitational pull of the partners’ bonus pool… all courtesy of the Australian taxpayer.

Let’s face it, when you hire a consulting firm from the big end of town, you expect to be buying high-end analytical skills, and a level of process discipline that ensures the highest quality of output.

I’m not one to delight in others’ misfortune, and you probably know me by now: I don’t jump on the ‘collective outrage’ bandwagon. But this public relations nightmare may be a distant early warning for the consulting industry – the shape of things to come in an AI-driven world.

If nothing else, it begs the question, Do we still need consultants, if consultants are just using the same AI tools that, despite their obvious drawbacks, are available to all of us?

I begin with a quick recap on the Deloitte situation; I explore an interesting article from The Economist, and I give you my bold predictions: three areas where consultants will stay ahead of the AI game, and three areas where they’re likely to be under threat.

DELOITTE SUFFERS SELF-INFLICTED WOUNDS

I don’t like to be critical of anyone based purely on what turns up in the media. I’m acutely aware of the fact that I’m not in their shoes. I’ve also been on the wrong end of media reports often enough in the past to know how prone they are to basic factual errors.

But in this case, neither Deloitte nor their hapless clients are disputing the facts. Even the most generous interpretation of the evidence casts Deloitte in an incredibly unfavorable light (not to mention the breathtaking incompetence of the government department that commissioned the report, and then greenlit the next phase of the Deloitte engagement after being fed the tripe that was the initial report).

The Australian Financial Review produced a number of excellent articles analysing the Deloitte train wreck. I am not going to go into them here, but there’s a range of interesting angles, and the subeditors seem to have had a lot of fun coming up with the headlines:

• Deloitte to Refund Government, Admits Using AI in $440k Report;
• The Cautionary Lessons of Deloitte’s AI Sloppiness;
• Deloitte Blamed ‘Transcription Issue’ for Flawed AI Report; and, my favourite
• AI Scandal Not Enough to Stop Canberra’s Addiction to Deloitte

I must say, I found the commentary on this is slightly disheartening. There are all sorts of experts coming out with helpful tips on how we should adapt our approach to AI based on this cautionary tale from Deloitte.

There seems to be a focus on regulating how AI is used. By that, I mean “small-r” regulating, as in, establishing corporate rules for AI. This includes being upfront about how and where AI is being used; keeping a human in the loop; and having better fact checking processes.

Well, okay, but for me, this misses the point.

One of the AFR articles I just cited suggests that the consulting industry is going to remain relevant, because there’s no replacement for human cognition.

One of the main selling points for consultants in the past has been their ability to leverage data sources and knowledge from their global client footprint, which is something a single client firm can’t achieve on its own.

It looks suspiciously to me as though AI is already leveling this playing field…

For example:

• How much did Deloitte’s consultants rely on their own smarts?
• How much did they rely on the accumulated wisdom of their global knowledge base?
• How much did they rely on the IP of their internal processes?
• How much did they rely on their boilerplate solutions?
• How much did they rely on the information in front of them: the customer’s actual context; and
• How much did they simply subcontract their work to Claude and Chad (that’s Chad-GPT, of course).

It’s true that there’s no replacement for human cognition… yet!

But, armed with the right data, companies are going to be able to lean more heavily on the cognition of their own people. Why pay a middleman to prompt an AI GPT when you can do it just as effectively yourself… for free?

CONSULTING IS SHOWING SOME STRESS FRACTURES

Consulting has long been a bellwether for corporate health and economic confidence. It signals availability of capital, companies’ appetite for growth, and a willingness to invest in strategic change across all sectors of the economy.

Even in lean times, when consulting spend appears to contract, there are still plenty of companies hiring consultants to help them cut costs, restructure, and improve operating efficiencies (the irony of which isn’t lost on me: some companies spare no expense in the quest for a lower cost base).

But in the last 12 months or so, some of the top consulting firms in the most lucrative markets have been shedding staff and slowing their talent pipeline, by drastically reducing graduate intakes.

We’re probably not going to see the impact of this in their financial results for another year or two, but it’s clear that the decline has commenced.

Long before the Deloitte debacle became public, an interesting article appeared in The Economist looking at the erosion at the very peak of the consulting business. It was titled How McKinsey Lost Its Edge. The article makes a number of points about the decline in McKinsey’s business model, much of which has nothing to do with AI.

But the conclusion the author makes on AI towards the end of the article is foreboding. Here’s a slightly modified quote from that article:

“Eventually, the AI revolution may come for McKinsey as well. Strategy projects combine deep cogitation with a good deal of grunt work. Consultants must comb through spreadsheets and prepare reams of PowerPoint slides in order to make their recommendations look as robust as possible. Before long a bot may be able to do a good chunk of that. After all, plenty of consultants already sound like that.

At first, that may be advantageous for McKinsey. McKinsey’s technology chief says that AI will make its teams faster and better. Sooner or later though, clients may start to wonder why they are paying such enormous sums when much of the work is being done by technology that they could use themselves.”

I think that the slowdown at the premium end of the consulting market is the canary in the coal mine, and the consequences will be far-reaching.

Many top leaders in industry were shaped by their early consulting roles. Having done this myself, I can speak really highly for the rapid growth that a stint in a consulting firm can produce. And in fact, many of the world’s top executives are alumni of the MBB firms –McKinsey, Bain, and BCG.

At the time the article was published, 36 of the Fortune 500 CEOs came from the MBBs.

So, what happens when that talent pipeline dries up? Today’s partners were once consultants, plucked from the best universities in the land. They learned their trade from their peers. They became engagement managers and associate partners before they got the keys to the executive washroom.

However, if the future is a world where AI does the low-end, repetitive grunt work, how do people learn the ropes so that they can ascend to a level where human cognition still has currency?

And if the top consulting firms don’t have this development pipeline, will their senior people provide sufficient value to justify their astronomical fees?

WHY WOULD YOU ENGAGE A CONSULTANT?

Where are consultants still likely to offer value for money?

We produced an episode about consultants around three years ago now, Ep.220: Management Consultants: Are they worth It? This was before generative AI, in a long-forgotten time before Claude and Chad were even born.

This was a bumper episode, where I covered a massive amount of ground. Having reviewed it this week, I can confidently say that it’s a complete guide for how to understand and engage the different types of consultants.

In this particular podcast, I gave a breakdown of the consulting market, with an overview of which consulting firms do what type of work.

The market has several tiers:

• The MBBs (McKinsey, Bain, and BCG) are the pinnacle in terms of both capability and price;
• The Big Four (Deloitte, EY, KPMG, and PwC);
• There are the consultants that are more technology-oriented, like Accenture and Fujitsu and IBM;
• There are highly specialised firms, like those that do economic modeling and forecasting;
• And then there are the millions of small consulting companies where the individual is the company.

I also explored the three compelling reasons why you might consider hiring a consultant.

1. Lack of capability
2. Lack of capacity
3. The need for independence

If you lack capability, particularly in an area that’s either a one-off requirement, or in an area that isn’t part of your core expertise, consultants can be highly valuable. They don’t just bring extra smarts; they also bring access to the intellectual property that they’ve acquired, by working with other clients like you in many different industries and markets. Used wisely, this is a great reason to bring in a consultant.

It’s fair to say that AI is making this particular value proposition a lot less compelling.

If you lack capacity, consultants can offer the opportunity to scale, without having to increase your underlying cost base. The ability to flex your capacity with capable resources, who can focus on a specific task for a finite period, can be an effective way to solve a problem… as long as there’s a clear path to value delivery at the end of it, that is!

In a world where AI dominates, this is going to remain a potentially effective investment.

If you’re looking for an independent opinion, consulting firms can still bring that to the table (bearing in mind of course, that many consultants just feel the direction of the prevailing winds and tell their clients exactly what they want to hear).

Aside from the fact that independent advice can be extremely useful for things like financial audits, where a board or investor community needs to have confidence on the numbers being produced by management.

Independent peer review of major investment proposals is another great way to use expert consultants to provide independent advice.

Boards love to appoint independent consultants to report directly to them, and give feedback on what’s going on in the parts of the company that they can’t necessarily see. Consultants can provide a level of comfort that management is actually behaving themselves.

THE VALUE PROPOSITION IS DECAYING

I think that AI is clearly going to erode that value proposition for using consultants to bring an injection of capability. Much of this work is going to be driven internally soon enough, because the type of knowledge that the top firms used to call IP will be truly commoditised.

But the use of consultants for other reasons, like augmenting internal capacity and providing independent scrutiny of management, will probably survive the pressure of AI, at least for now.

In Ep.220, I gave my top 10 traps for young players. I’m not going to go into those here, but I’d highly recommend you review those before you consider engaging a consultant. It could save your company a shit-ton of cash.

In fact, it’d probably make that $440,000 Deloitte haul from the federal government seem like chump change, if you get a consultant engagement wrong.

Instead of that, I want to outline the top three areas where I think consultants are still going to hold their value, and three areas where their value may already be under threat from AI.

3 AREAS WHERE CONSULTANTS WILL HOLD THEIR VALUE

Let’s start with three areas where I suspect that hiring a consultant will still potentially pay dividends.

1. Operational efficiency.

When you want to drive operational performance improvement, as I’ve done on several occasions in the past, you need experts from outside to come in and place some scrutiny over what your people are doing: to watch the people work; to talk to them about what they do on a daily basis; to work out how much of the current inefficiency is attributable to genuine obstacles, and how much is just an excuse to cling onto the status quo.

You need humans to do this type of work, at least as it stands today.

2. Digital disruption and technology integration.

Ironically, the same consultants that face being disrupted by AI will be the ones who guide their clients’ organisations on how to implement it. When it comes to digitisation, it’s a complex ecosystem of people, processes, systems, and politics.

This is a space where external advice from smart people is still going to be required in order to build effective change-management strategies. I think this type of advisory work is also pretty safe for the moment.

3. Comfort blanket consulting.

There are still many boards, particularly in listed companies, that want the comfort of a big brand to help them justify the really big decisions. Whether it’s engaging a SHREK firm to conduct a search for a key executive role, or a McKinsey team to help develop your strategy.

The comfort blanket of a big brand to support management actions is sometimes as much an insurance policy as anything else.

It reminds me of the incredibly effective marketing campaign that IBM ran in the 1980s, targeting Chief Information Officers and Chief Technology Officers. The tagline was…

“No one’s ever been fired for buying IBM.”

3 AREAS WHERE CONSULTANTS WILL STRUGGLE

Let’s look at the three areas where hiring a consultant is likely to lose its lustre as AI becomes more ubiquitous.

1. Audit and compliance.

The vast amounts of manual labour that are currently required to undertake financial audits is likely to be shifted to AI. Not only will this remove a lot of the work traditionally given to consultants who undertake this compliance work (typically the Big Four), but it’s also going to reduce dependence on offshoring.

The concept of having a hundred qualified accountants in an office in Manila, even with the cost advantage of labour arbitrage, will begin to look less attractive, as AI learns how to crunch the numbers.

2. Capital efficiency and investment strategy.

When I think about what goes into a really good investment analysis and decision-making process, it’s fundamentally about financial hurdles, assumptions, and risks. Leaving aside the product strategy and competitive elements, there isn’t much that AI won’t be able to replace.

Analysing historical data to provide accurate risk assessments, trawling through assumptions to price their impact and give an assessment of their robustness, applying investment rules and modelling financial return profiles, and comparing multiple potential investments to rank their attractiveness.

What used to require highly skilled analysts will be easily built into an AI agent.

3. Strategy.

Strategy development has always been a bit of a black art.

Companies bring in the consultants who live at the top end of the gene pool to help them do this work, but in reality, a lot of this effort is completely wasted.

Consultants start with templates. And yes, they look at your market position and bring a lot of intelligence to the table, but you already have a lot of this data available to you. And working out how to get AI to interpret it will be a key job that executives need to contemplate themselves.

The fact that most strategies defy sensible execution should already be sounding alarm bells for most companies. And, not withstanding the point I made earlier about the desire for a well-known consulting brand to provide a comfort blanket, strategy can be driven by management teams internally with the help of AI.

Let’s face it, the most sophisticated strategists, like McKinsey, may help you to conduct war-game simulations to replicate competitive forces… but with AI, you’ll be able to run a thousand what-if scenarios in a matter of seconds.

I think that strategy, the linchpin of the premium consulting firms, is definitely under threat from AI.

THE CANARY IN THE COAL MINE

There will almost certainly be a need for consultants to provide services in a range of areas well into the future; but some of the highest-margin, most resource-intensive work will no doubt be subcontracted to AI.

As with many industries, the toughest question is going to be, “What happens to the people who used to do this work?” Well, the very best people used to ascend through the hierarchy to become partners.

But what happens when you can no longer grow people from within, because the entry-level work no longer exists? Does this erode the foundations of the consulting skyscraper?

One thing’s for certain: management teams are going to have to be much more astute about what they are actually buying from consultants. Maybe government departments can get away with this type of mismanagement, but now, the world is becoming more attuned to the shortcuts the consultants are taking.

Hiring a consultant who borrows your watch to tell you the time used to be almost a cliché in business, but in the future, it could just end up being a career-limiting move, more than anything else.