Why professional services firms are solving the wrong AI problem — and the governance gap that’s quietly eroding their brands.
Every professional services firm I talk to is asking the same question right now: “How do we get our people to use AI?”
Fair question.
But here’s the one I hear almost nobody asking: “Who governs what AI creates once they do?”
Because your people are already using AI. The adoption problem is solving itself. The governance problem? That one’s getting worse by the day.
Who governs what AI creates?
Because here’s what’s actually happening. Your associates are already using AI. They’re using Copilot to draft client memos. They’re using ChatGPT to build parts of pitch decks. They’re using Claude to write case studies. And every one of those outputs uses whatever font the AI defaults to, whatever tone it feels like if noit trained properly, whatever image it can find.
AI didn’t create your brand problem. It accelerated it.
What “Ungoverned AI” Actually Looks Like
Let me paint you a picture that’s probably happening at your firm right now.
It’s a Tuesday afternoon at a 500-person accounting firm. A senior manager needs a team slide for a client pitch tomorrow. She opens Copilot and asks it to create a four-person team slide. Copilot delivers something that looks professional enough — clean layout, decent copy. She drops it into her deck and moves on.
Here’s what she doesn’t know: the headshots are pulled from an outdated SharePoint folder. The font doesn’t match the firm’s brand standards. The manager’s title changed two months ago. The copy is inconsistent and conflicting from the last 5 pitches. And the disclaimer that’s required on every client-facing document? Not there.
Multiply that by 500 employees, five days a week, across every document type your firm produces.
That’s not an AI problem. That’s a governance problem that AI made exponentially worse.
The Numbers Behind the Governance Gap
This isn’t theoretical. The data is uncomfortably clear.
We ran a brand compliance scan for a $1.5 billion management consulting firm, eight months after they completed a major rebrand. The result: 67% of their files had branding issues. Wrong fonts. Old colors. Outdated logos. Unapproved trademarks and imagery. Eight months and millions of dollars into a rebrand, and two-thirds of what they were putting in front of clients was off-brand.
That was before AI-generated content was widespread. Imagine what those numbers look like now that every associate has a content creation tool, the vast majority with zero brand guardrails.
The broader industry data tells the same story
77% of companies report issues with brand consistency. Marketing teams spend 32% of their time just policing brand compliance, that’s not strategy work, that’s babysitting. And here’s the kicker: companies with consistent branding outperform their peers by 20%, according to McKinsey. The governance gap isn’t just an operational headache. It’s a revenue problem.
“Brand-Aware” Is Not the Same as “Brand-Governed”
This is where the conversation goes wrong at most firms, and I think it’s worth being precise about the distinction.
You can make AI “brand-aware.” You can load your brand guidelines into a prompt. You can build a custom GPT or a Claude Skill that knows your fonts, your colors, your tone of voice. And that’s genuinely useful, I’m not dismissing it.
But brand-aware is advisory. It’s a suggestion engine. It can tell the AI what it should do. It cannot enforce what happens after the AI creates something. It cannot prevent an associate from ignoring the suggestion and saving the document with the wrong font. It cannot scan the 50,000 files already sitting on your SharePoint to find out which ones are off-brand. It cannot push an updated logo to every employee when your brand evolves.
Brand governance is something fundamentally different. It’s enforcement. It’s measurement. It’s accountability at scale.
Think of it as a four-layer stack:
Layer 4 — Content Creation. This is where AI wins. Copilot, ChatGPT, Claude — they’re all excellent at generating content.
Layer 3 — Brand Awareness. This is where you can teach AI your brand rules. Your fonts, your colors, your voice. And it works, until someone ignores the suggestion, tweaks the output, or just saves the deck without checking.
Plus, not every piece of content needs to be AI-generated in the first place. Some content should be standardized, pre-approved, and pulled from a single source of truth, not recreated from scratch every time someone needs it. And let’s be honest about cost, the pricing you’re seeing today on these AI tools is subsidized by venture capital, not by sustainable business models. That won’t last forever. Brand awareness is a real step forward but it’s solving one small piece of a much bigger problem.
Layer 2 — Governance and Enforcement. This is the layer most firms don’t have. It’s the ability to scan every document across your organization and tell your CMO exactly what percentage is on-brand. It’s rules that are enforced automatically, not recommended hopefully. It’s the difference between “we think we’re compliant” and “here’s the data.”
Layer 1 — Content Distribution and Lifecycle. This is the layer most firms completely overlook and it might be the most critical one, that’s also easiest to accomplish and make the most impact at your company. You can create perfect content all day long, but if people can’t find it, they’ll build their own. If they can find it but it’s outdated, you’ve just put last quarter’s messaging in front of a client. Governance isn’t just about what gets created. It’s about making sure the right content is easy to find, always current, and automatically replaced when something changes. When your leadership team changes, when a disclaimer gets revised, when your brand evolves, the entire organization needs to be working from the same source of truth, not whatever someone saved to their desktop six months ago.
Layers 4 and 3 (creation and awareness) are the hot topics right now, and rightfully so. I believe this technology is going to be transformational for professional services firms.
But I hear almost nobody talking about Layers 2 and 1 ) governance and distribution). Here’s the thing: that’s where firms can get the most improvement in productivity and brand consistency the fastest. These aren’t just nice-to-haves. They’re the essential foundation that Layers 3 and 4 need to actually work.
Think about it this way, how can you expect AI to create good documents if the templates it’s pulling from are outdated? If the brand assets in your system are three versions behind? If the “approved” content employees are feeding into AI prompts is something they saved to their desktop during the last administration? AI learns from what your people give it. If your foundation is messy, AI just scales the mess.
Why LLMs Can’t Close the Governance Gap on Its Own
I want to be specific about this, because vague claims don’t serve anyone.
There are five things that brand governance requires that AI tools, even the most advanced ones, structurally cannot provide:
Deterministic enforcement. AI is probabilistic. Ask it to apply a brand rule and it will get it right most of the time. But in accounting, consulting, financial services, and law, “most of the time” isn’t acceptable. When your brand standards say the legal disclaimer appears on every client-facing document, the approved logo is in the correct position on every slide, the partner’s credentials are listed exactly as the firm requires, that has to happen every time. Not 90% of the time. Every time. Governance systems check the actual document data. It either matches or it doesn’t. No interpretation, no variance, no “close enough.”
Organization-wide scanning. Even the best brand-aware AI only governs what it creates. What about documents built without AI? Presentations inherited from other teams? Files modified manually after AI generated the first draft? The 50,000 legacy documents on your file server right now? At most firms, even aggressive AI adopters, AI handles maybe 20-40% of content creation. That means 60-80% of your documents are invisible to any AI-based governance approach. Partial coverage is not governance.
Compliance analytics and dashboards. When your CMO or managing partner asks “How compliant is our organization?” the answer can’t be “we think we’re in good shape.” They need actual numbers. Compliance rates by office, by department, by document type. Trend lines showing whether you’re getting better or worse. The challenge is that even if every employee at your firm used AI for every document (which they won’t and shouldn’t) you’d still need a way to measure, and report on compliance across the entire document ecosystem. The presentations built manually, the inherited decks, the files sitting on SharePoint from before AI existed. Governance requires a complete picture, not just visibility into what one tool touched.
Content lifecycle management. When marketing updates a logo, who notifies every employee using the old version? Who flags the stale content? Who pushes the updated asset directly into the tools people use? This is a distribution and version control problem and the majority of brand compliance failures are caused not by bad content creation, but by employees using the wrong version of the right content.
The Firms That Win Won’t Be the Ones That Adopted AI First
Here’s my honest take, as someone who’s spent years building brand governance and productivity tools inside Microsoft 365:
AI is not the threat to your brand. AI is genuinely great for professional services firms. It accelerates content creation, it reduces busywork, it helps your people focus on higher-value activities. Again, I’m pro-AI.
But AI at the individual employee level without governance is just a faster way to go off-brand and make your life as a marketer harder. It’s a content firehose pointed at your clients with nobody checking what comes out the other end.
The firms that will differentiate themselves over the next two to three years won’t be the ones that adopted AI the fastest. They’ll be the ones that governed it the best. The ones that can say, with data, not hope, “Yes, we’re on-brand. Across every office, every document type, every employee. Here’s the dashboard to prove it.”
That’s the gap. And it’s only getting wider.
I’m curious — what does your firm’s governance plan for AI-generated content look like? Is anyone in your organization actively measuring brand compliance, or is it still a qualitative gut check?
