I helped build an agency that eventually sold and exited.
We weren’t massive, and we weren’t chasing vanity growth. But we built something sustainable. Valuable enough that someone wanted to acquire it.
And here’s what’s interesting.
If I were building that agency again today, in an AI-accelerated world, there are many things I would do exactly the same.
And there are a few things I would do differently.
Yes, the tools have changed. AI has reshaped execution.
But the fundamentals of what creates durable agency value haven’t shifted as much as people think.
The real question is this:
If your agency disappeared tomorrow, what would your clients actually miss?
Because that answer reveals where your real value lives.
What I Would Absolutely Do the Same
Business Model Judgment
Our agency specialized deeply, and that specialization sharpened everything.
Working inside the same types of businesses repeatedly gave us more than marketing expertise.
It gave us business context. We understood how revenue was actually generated, where margins were thin, where growth stalled, and where complexity quietly eroded profitability.
We could see when a company didn’t have a marketing problem. It had a model problem:
Sometimes the issue wasn’t awareness. It was an overcomplicated offer.
Sometimes it wasn’t lead flow. It was misaligned pricing.
Sometimes it wasn’t positioning. It was trying to serve too many audiences at once.
Specialization gave us the confidence to name that.
That’s a different layer than execution.
AI can generate options. It can optimize tactics but it cannot sit inside a founder’s business and diagnose friction. It cannot understand how this business works in this season under these constraints.
Clients need (and want) someone willing to say: this is what we prioritize, this is what we stop, this is where you reposition, this is where you’re misaligned.
That kind of judgment is built through proximity and thought partnership.
And that layer still matters. If anything, it matters more now.
Strategic Sequencing
We were also strong at designing the arc of strategy.
Not just deciding what to build, but understanding how the work needed to unfold so each move strengthened the next.
Clients need direction: they need someone who understands how positioning influences pricing, how pricing affects operations, how messaging drives demand, and how all of it ties back to revenue.
That kind of sequencing doesn’t come from templates.
It comes from lived experience.
And it’s not disappearing.
What I Would Emphasize More Today: Risk Mitigation
Five years ago, we focused heavily on growth.
Today, I would focus equally on risk mitigation.
AI hasn’t just increased speed. It has increased volatility:
- Markets shift faster
- Messaging saturates more quickly
- Competitive advantages erode sooner
- Copycats appear overnight
- Reputations can be damaged in a single misstep.
The environment is less forgiving.
AI can generate messaging and accelerate campaigns but it cannot sit in a tense leadership conversation and feel the undercurrent in the room.
AI cannot anticipate how a move might trigger backlash. It cannot read internal political dynamics or protect brand equity during a strategic pivot.
Agencies that endure will protect downside. Risk mitigation isn’t defensive thinking, it’s strategic maturity.
And if I were building today, I would develop that muscle earlier and more deliberately.
The Thing We Did Exceptionally Well
Clients told me repeatedly that they looked forward to our strategy calls.
Not because we brought more deliverables, but because we brought perspective.
Working across similar companies gave us cross-client pattern recognition:
- We could see shifts forming before they became obvious.
- We noticed ceilings appearing across multiple businesses.
- We sensed when a channel was beginning to fatigue or when a positioning angle was flattening.
That meta-view created confidence.
Clients weren’t paying for slides they were paying for insight drawn from lived comparison and the ability to step back and say, “Here’s what we’re seeing across the landscape, and here’s what that likely means for you.”
AI cannot replicate lived comparison and practice judgement across real businesses with human nuance.
That is still where agencies win but only if they operate at that level.
The Real Test
So I’ll ask it plainly.
If your agency disappeared tomorrow, would your client struggle because they’d miss the production or because they’d lose the thinking?
If it’s production, you’re exposed.
If it’s thinking, you’re defensible.
Where This Leaves Us
I don’t believe AI is the end of agencies but I do believe it’s clarifying something.
It’s revealing which agencies were built on output, and which were built on insight.
The first group will feel squeezed.
The second will become rarer and more valuable.
If you’re building a Small But Mighty agency today, this isn’t about panic.
It’s about recalibration:
- Sharpen your specialization
- Elevate your judgment
- Design with risk in mind
- Extract the patterns you’re seeing across clients
- Move up the value chain intentionally
If clients would only miss the deliverables, you’re vulnerable.
If they miss the discernment, you’re building something durable.
