AI and professional services – what happened next…

A few weeks ago, the article “What AI Will – and Won’t – Do To Consultants” triggered a bit of feedback and spurred additional calls, interviews and research.  The key observations from that effort included:

• The professional services industry is definitely an industry that is concerned about AI’s effects.
• Some sectors (e.g., offshore package software implementers, MSPs, etc.) are especially worried while management consultants are less so.
• Smart service firms are actively seeking ways to utilize AI so that their people are seriously focused on adding value and not doing a lot of scut work.

The management consulting view

There are some specific points where AI and management consulting intersect. One key example of this involves the tedious, tactical, expensive work that adds little value and lengthens project timelines.  More specifically, management consulting personnel are often tasked with: collecting and synthesizing vast amounts of client data, conducting primary research to gather other data that was not directly available from the client, developing reports that describe what this data means, and, identifying the new strategies the client must embrace and the projects needed to bring the strategies to life.  Much (but not all) of this activity can be done by (or done more quickly by) new AI technologies.  Generative AI, for example, can complete some of the analysis and content generation.

I recently spoke with Casey Foss, Chief Commercial Officer with consultancy West Monroe to get her perspectives on this. She indicated that consultants will be more important than ever as clients will want access to a true, trusted advisor. Most AI answers (except hallucinated responses) aren’t that novel and the results could be less than trustworthy. Clients might not want a probabilistic AI answer or suggestion.  Clients will want access to subject matter experts (SMEs) – the consultants who possess real-world industry expertise and deep functional, prior expertise in similar matters. Foss summed it up well when she said “the real work starts when the mundane tasks are done”. I think that’s the quote of the day.

Foss identified an obvious challenge in this new AI powered consulting world. How will consultancies develop and build the next generation of leaders and SMEs? When newer staff collected and analyzed data, they often shared their observations with more senior consulting leaders. It is through this work that junior consultants gained insights into businesses, learned from highly experienced peers and honed critical consulting skills (e.g., interviewing senior client leaders, ideating new client strategies, conducting highly effective meeting, driving change in client organizations, etc.).

Foss also indicated that these activities create first-hand, up-close experiences that are often times best served in-person and help create the next generation of industry leaders.  You can’t underestimate how much can be observed and learned by walking shop-floors, client office hallways, and other client locations. Often, what consultants observe doesn’t match the client’s process diagrams, stated procedures, and anecdotal soundbites that management believes to be the case. Some of my most important insights come from these kinds of activities.

However, clients today may be less inclined to let consultants have this unconstrained access to their people and facilities. Some clients will argue they lack the time to spend with junior consultants or don’t see the value in this effort. Some clients want consultants to work remotely and avoid travel costs. But the biggest issue here is that clients don’t want to pay a lot of fees for junior consultants when all they really want is access to the consultancy’s top experts. The best approach may be to permit this in-person learning and combine this with AI inputs.

Alternatively stated, clients want to deal with the deeply knowledgeable SMEs and not pay to train more junior people who add very little to their project. Readers should note that consultancies shouldered a lot of this on-the-job skills development. Rates for these newer staff were often discounted and senior executives mentored these staff.

Junior consultants are really apprentices. They learn from the more senior leaders in their firm. It’s how they develop deep industry knowledge and gain practical client engagement expertise. This is the consulting industry’s lifeblood for talent development.

At this point in the conversation, Foss took the discussion into a new area: exactly how consultancies should be using new AI technology themselves. We discussed how AI-powered resource management tools from PSA and HR technology firms are making better staff assignments for employees and clients. I’ve seen how AI can infer specific skills needed for a given assignment based on information found in the proposal, work plans and/or other documents. I’ve seen these capabilities in some of the new third-party solutions on the market.   

Foss stated that West Monroe’s AI can assist in:

• Creating better, more complete proposals, work plans, etc.  More time can be spent doing phase 0 analysis to drive insights and prototypes.
• Collecting data and assessing it against proprietary data sets in West Monroe’s Intellio® Insights to create new marketing thought leadership
• Design and develop automated solutions for data, software, and products tasks (e.g., integration development, data conversions, report development, process automation, etc.)

Casey indicated that the latter work, frequently a key output of offshore service firms, can now be done in-country again. AI makes this work cost effective in-country and it avoids 12-hour time delays and communication challenges with offshore firms.  (Note: West Monroe is a business and technology consulting firm with a limited offshore presence in Costa Rica.).

Why AI can’t fully replace consultants

To understand how AI may/may not replace all of the work consultants do, you should check out this video from Sloan MIT Review. While the video is about detecting AI use by jobseekers, you can immediately see how easy AI gets tripped up, how limited its functionality is, and, how easily employers can spot its use if they know what kinds of follow-up questions to ask.

While watching this (it’s only 5 minutes long), I realized that great consultants know how to ask probing questions, tricky follow-up questions, and, request client personnel to explain the nuances of processes, policies and other management/business decisions. In my consulting training decades ago, we were taught to ask clients “Why?” five times in a row to divine the real, underlying forces affecting a given client. (Hint: It works!) AI tools can’t quite replicate that kind of relentless digging and they certainly lack the insights to know which area they should probe next and why.

If you thought AI was all-powerful, that video will change your perspective.

It’s not an hours billed game anymore

This AI issue is not just affecting management consulting and IT services. A recent article in the Wall Street Journal discussed the transformation occurring in the creative services space. The author noted common issues that affect numerous service firms/sectors. Glance at the following excerpt and note the similarities to consulting work:

Agencies have long billed marketers by the number of hours their employees spend producing client work, using rate cards to charge different amounts for contributions by people according to their role. 

Now, AI is eroding the number of people, hours and roles required to deliver for clients, and agencies may find the standard billing arrangement comes up short. AI is helping agencies rapidly produce personalized creative images, for example, or altering elements like color, position, lighting and language—tasks that were once highly manual. It’s also letting copywriters, who once may have needed several hours to write 50 variations of copy for a given ad, now generate 100 variations immediately, then choose to edit and curate them.

One of the sources quoted in that Wall Street Journal piece was Tracey Shirtcliff, a serial entrepreneur and CEO of SCOPE Better, a services CPQ (configure, price, quote) software provider. A few weeks ago, I had a chance to chat with her.

In our brief conversation she noted that “If law firms only provide an hourly rate, people won’t buy”.  I believe she’s right. Clients want outcomes not rate cards. Further to Shirtcliff’s point, if I find a prospective client wants to challenge the rates a services firm is charging, then it appears that the service firm hasn’t done a good job of differentiating itself from its competitors. In other words, if the prospect sees two firms that are essentially offering to do the same thing for roughly the same price, what else can they haggle over other than price?

And that gets us to the fundamental services question du jour: “Do clients want to pay based on time expended on the effort or based on the realization of project outcomes/deliverables?”  Answer: It’s the outcomes, obviously!

Every service firm today should be looking at AI and asking itself:

• “How can our firm use AI to clearly differentiate us from our competition?” (Competitive differentiation)
• “Will AI help us identify something that clients and prospects didn’t expect but will add value to their firm?” (Expanding the value proposition)
• “How can we use AI to help more client personnel than we previously did?” (Grow the value delivered)
• “How can AI help us move more quickly in delivering value to clients?” (Shorten time to value)
• “How will AI help us deliver more comprehensive, complete solutions to clients?” (Better results)
• “How will AI help us reduce the quantity and severity of adverse surprises that occur during the course of projects?” (Delighting the customer)
• “How can AI help our team members focus on the highest-value-adding work elements while it (AI) deals with the more pedestrian, generic, busy-work tasks?” (Greater focus on value-add activity)

But Shirtcliff suspects we’ll see some resistance from service firms not wanting to change or use AI as a change catalyst. Why? There are always firms who want to maintain the status quo with the defense “but that’s the way we’ve always done it”. I’d add that early pioneers with outcome pricing likely found it hard to implement this with clients years back.  I suspect that resistance has softened a fair bit by now.

I can validate Shirtcliff’s position that clients want an outcome and, whenever possible, they want certainty. They want to know what something will cost and hate to be surprised. To illustrate, I rarely ever provide estimates of hours to be worked to my own clients. I tell them the elapsed time the project will take, who will be on the project and the deliverables they should expect. The hours and mix are my problems to manage – if I do it right, we make money. If I mess it up, that’s on me. Clients rarely care about the time and rates per person as long as they feel confident a service provider will deliver the expected work products on time and to the estimated cost.

Why is this a concern these days? That’s simple: too many service providers won’t make the initial investment in time to meet with a prospective client and correctly scope out potential work or they try to get a prospective client to accept an ill-fitting standardized proposal. Either way, the unfortunate client that accepts a ‘proposal’ like this is going to be buried in a regrettable tsunami of change orders.

Shirtcliff and I had a further exchange of thoughts re: scope creep.  We discussed how it is often the result of a poor initial scoping of the project. It occurs frequently especially when the service provider didn’t invest enough time in identifying the true project requirements, used a poor scoping tool (e.g., a generic spreadsheet), or, intentionally low-balled the deal with a plan to provide a more accurate estimate after the deal was signed. Some of those scenarios may just be the kind of problems that Shirtcliff’s firm helps its users avoid.

AI tools that help identify ALL of the potential tasks involved in a project could be of great help to service organizations. AI might help some service firms craft better proposals and deliverables – and – see these get completed faster, as well.

At this point, Shirtcliff noted that AI has more uses than just for helping in the pricing and staffing of projects. AI tools should help sales leaders and project planners identify opportunities for the project team to identify heretofore unseen opportunities for the client to differentiate itself, improve its competitive standing, etc. She maintains that the best service firms give clients something special and/or provide them a pleasant surprise. I wholeheartedly agree.

My take

An article (or two in this case) can’t do the subject matter justice. There are many areas where services and AI can intersect and we’ve only scratched the surface for now. For example:

• AI could be terrific in spotting certain kinds of service, employment and other frauds.
• AI could very efficiently handle all kinds of project tracking administrivia.
• Etc.

AI use cases to identify potential service personnel flight risks already exist as do great AI-powered resource management tools. To see the latter optimize a massive staffing problem while simultaneously trying to optimize the employee travel, client timeline and 47 other factors in a matter of seconds or minutes is amazing.

And yet, with all of the great positive use cases for AI in services, there are still cautions that service firms should heed. That’s likely the subject for the next piece. Until then….