Fee models in consulting

I've had a lot of feedback and questions on my post on sharing my rate card in the last couple of years, a large subset of those questions can be summed up as "How do big consultancies charge their clients?".

Most independent consultants have a relatively simple fee structure, which is known as "time and materials". A consultant has a rate for a unit of their time (an hour, a day, a week) and the amount of time they charge for each project is a function of how many units of time they work multiplied by the pre-agreed rate. They then add in things like expenses and other "materials" (usually without a margin on top) and add that to their time. Send invoice, done.

Large consulting organisations rarely charge that way. They still use time and materials as the base unit of cost, but the way the client pays for it differs quite a lot. This post is my attempt to answer this question and describe the journey of how they got there. The content of this post is broadly focused on "strategy" consultancies like McKinsey & Company, Bain and BCG.

The olden days

If we look back into the far-off, long-forgotten olden days of consulting (before the millennium lol) large strategy consultancies used a model similar to the one we described for independent consultants.

A client had a problem they needed solving, a Partner at one of these firms came along and said "That'll take three people about six weeks, which is $N of their time, but it might be more than that". The fees were open-ended and projects took as long as they took. This was great for the consultancies, it allowed them to account for difficult to predict projects or scope creep from clients (which, as we know, always happens).

Over time, their clients became less and less happy with this model, as it made predicting the cost of these projects very difficult. They also perceived it as incentivising consultants to take as long as possible to do the work. The longer they worked, the more they got paid.

Because of this dissatisfaction, two market forces emerged to put this model under pressure:

1. The consulting market grew, and larger consulting projects were undertaken. This meant more of your client's balance sheet was being paid to consultants, and therefore more of it was "unpredictable" because of these dynamic, open-ended pricing models. CFO's and public markets don't like large, growing, and unpredictable costs on a balance sheet. This put a lot of demand side pressure on consultancies to change their fee models.
2. As the market grew, competition increased. Some players in this market recognised the effect of (1) and saw that "fixed fee" models were a competitive advantage. Clients started favouring those arrangements in growing numbers. This put a lot of supply side pressure on the rest of the consulting ecosystem to follow this model.

This happened in different parts of the consulting ecosystem in different parts of the world at radically different times and with radically different effects, but for "high-end" strategy consulting it was pretty universal by the 2010s. It happened much earlier in large-scale technology consulting and outsourcing (like TCS, InfoSys, Accenture).

Fixed fee, but not "fixed fee"

While known as "fixed fee" these pricing models aren't usually as simple as you might expect. Rarely, if ever, do contracts say things like "We'll complete Project X for $N no matter what the scope" they look much more like clients buying "blocks" of time and materials.

In the above example, Foo Inc has reduced the opportunities for cost to grow out of control, as they know what they'll pay for deliverables X, Y and Z and if they want A and B they know how much that will cost them too. If they want something else entirely different, it'll still cost them some unpredictable amount, but that's acceptable, and they know how they'll mutualy price that if it happens, so they have some amount of control over it.

Fees at risk and sharing upside

This mix of pure time and materials and "fixed fee" has stuck around, but there's a new model whose adoption has accelerated in the last 10-years: consultants putting their fees at risk, and sometimes sharing in their clients' upside from their engagements.

It's easiest to explain this with another example:

This is a powerful idea for Foo Inc, as these consultants now have "skin in the game". Their fees are tied to the success of the work they're doing. It means that, in the worst case scenario (if the new market launch goes badly) Foo Inc has significantly reduced their cost, by paying the consultants much less.

However, the price they pay for that downside risk reduction is that if the new market launch goes well, Foo Inc pay Consultants Ltd a lot more money. As a leader at Foo Inc, you could rationalise that as okay, on the understanding that the new market will generate far more revenue than this additional cost.

But why would consultants do that?

On the face of it, this new model adds significant new risk into the consulting business model, but there are (once again) market forces that have forced this change.

As a direct result of the ever-growing dollars spent on consulting in the past 30 years, a resentment has grown towards consultants at their clients. This resentment is commonly driven by a perceived lack of quality in the work that's completed vs the cost; these consultants come in, get paid a lot of money and then leave with no consequence that their work added no value. They get paid the same for bad work as good.

This fees at risk and sharing upside model is a direct response to this resentment. This way, the consultants (in theory) have a significant incentive to make their work valuable and impactful for their clients.

The new existential threat

Recently, a new, much more existential threat to strategy consulting fees has emerged: data and analytics. To understand why this is such a problem, we need to first understand something fundamental about how strategy consulting works, let's again look at Foo Inc:

The reason this is such a threat to strategy consultants is because the humans who do "the work" tend to do things computers are pretty good at doing, they:

1. Take some context-specific data and variables
2. Apply some set of rules or pre-existing knowledge to those variables
3. Produce some output that's consumable by humans

If we think about this in very simplest terms:

Activity	Input context	Rules	Output
Consultants	Client-specific data, external data, proprietary data	Human knowledge from other similar engagements	PowerPoint, Excel
Data Models	Client-specific data, external data, proprietary data	Codified models and rules	PowerPoint, Excel

If we think about strategy consulting this way, you can see why companies McKinsey & Company, Bain and BCG are investing so heavily in analytics.

For example, McKinsey & Company's acquisition or Quantum Black (and it's subsequent meteoric scaling) might prove one of the smartest decisions that company ever made. You can't be displaced if you're the one doing the displacing.

All of these strategy shifts in pricing results in one conclusion: pricing is one of the Hard Problems™ in consulting. There are 1,000's of smart people who've spent decades looking at this problem, and they still rely on pretty high-level, simple solutions to this problem. What's also true is that the seismic impact of analytics will change these models again, and this kind of large-scale consulting will look quite unfamiliar in 10 years.