← Back to Insights

In over twenty years of working with family businesses, I have sat across the table from dozens of owners who have built something genuinely remarkable — survived recessions, managed payroll through lean years, navigated family tension, and come out the other side with a business that employs real people and creates real value.

Almost none of them have a succession plan. Or more precisely — they have a version of one that lives entirely in their head, has never been spoken out loud, and has certainly never been tested. "Someday" is not a plan. "My son knows what I want" is not a plan. The cost of that gap is far higher than most owners realize — and it starts accruing long before anyone retires.

The costs you don't see on a balance sheet

The financial consequences of a poorly managed succession are well documented — forced sales, valuation disputes, tax exposure, litigation. Those are the visible disasters. But the costs that concern me most build quietly over years, invisible until they're not.

I've watched talented next-generation leaders leave — not because they didn't want to be there, but because the signals were ambiguous. They couldn't tell if they were being developed or just tolerated. And I've watched families fracture slowly as siblings never given clear roles began interpreting every decision as a signal about their standing. Sunday dinners become careful. Holidays become tense. The business survives but the family doesn't.

Why owners don't plan

The reasons are understandable, even if the consequences aren't.

• Succession planning forces a conversation about mortality, which most people avoid.
• Naming a successor means making a choice — and making a choice means disappointing someone.
• Owners who have built something from scratch often struggle to imagine anyone else running it, including their own children.
• The business demands so much daily attention that "long-term planning" always waits for a quieter season that never comes.

None of this is a character flaw. It's human. But it doesn't make the cost any lower.

What a real succession process looks like

The businesses I've watched navigate succession well start early — often five to ten years before the intended transition. They treat it as a process, not an event. They develop next-generation leaders in real roles with real accountability, have hard conversations about readiness and fairness, and bring in outside perspective: someone who isn't emotionally invested in a particular outcome. They're not the ones who avoided the difficult conversations. They're the ones who had them early enough that the business, the family, and the relationships all survived intact.

A question worth sitting with

If you were hit by a bus tomorrow, what happens to your business? Not the legal answer — the real answer. Who runs it? Who holds the relationships with your ten most important customers? If the honest answer is "I'm not sure," that's not a crisis. It's an opportunity. But the window to use it wisely is shorter than most owners think, and the conversations are always easier now than they will be later.

← Back to Insights

Most of the business owners I work with want to use AI — they just don't know where to start. Right now, they're using it to rewrite emails or compare contract language. That's a reasonable beginning. But it's a long way from where AI can actually move the needle for a 50-person family business.

Here's what I've seen actually change, and what the next level looks like.

What actually changes

The highest-return AI applications aren't glamorous. Drafting routine communications. Summarizing long documents. First drafts of proposals, SOPs, job postings. Reports that used to take a half-day. A manager who compresses four hours of weekly admin down to forty-five minutes gets three hours back — every week. Multiply that across a team and you have real capacity without adding headcount.

Beyond the basics, the owners getting the most value are using AI as a thinking partner — pressure-testing ideas, exploring scenarios, identifying blind spots before a decision is made. They are not using it to make decisions for them. That distinction matters enormously.

The agentic shift — and why it matters for your business

The most significant development in AI right now isn't a better chatbot. It's the emergence of agentic AI — systems that don't just answer a question but pursue a goal. They plan a sequence of steps, use tools, make decisions mid-task, and report back when the work is done. The key distinction: ordinary automation follows rules. Agentic AI pursues outcomes.

Here's what this looks like in practice. A customer calls a heating and cooling company after hours. An agentic system recognizes the caller, checks their service history, sees the last visit was for a furnace issue, offers a follow-up appointment, books it, and sends a confirmation — all without a human involved. That's not a chatbot answering FAQs. That's a system completing an entire workflow end-to-end.

For a 50-person family business, the applications are less dramatic but just as real:

• Finance and accounts receivable. Agentic tools can read company policy documents, audit expenses, flag policy violations, follow up on unpaid invoices, and reconcile accounts — tasks that consume hours of staff time every week.
• Customer inquiry routing. An agent monitors incoming messages, categorizes them, drafts responses to routine questions, flags complex issues for a human, and routes requests to the right person — before anyone arrives at the office.
• Cross-system reporting. An agent connects your CRM, accounting software, and operations data and assembles a weekly summary, flagging anything outside normal ranges.
• Employee onboarding. When a new hire enters your HR system, an agent generates paperwork, sends check-ins at day one, seven, and thirty, tracks training completion, and surfaces gaps to the manager.

What doesn't change

The competitive advantage of a family business is almost always relational — the customer who's worked with you for fifteen years, the supplier who gives you priority because they trust you. AI cannot build or maintain those relationships. The hardest decisions aren't information problems — they're judgment problems. A human being has to own the outcome.

A practical framework for evaluating AI tools

• Does it save time on work that is currently costing us? If you can't quantify the time cost of the problem the tool solves, you probably don't have a real problem — you have a solution looking for one.
• Does it touch a customer relationship? If yes, the bar for quality is much higher.
• Who is accountable for the output? A human being needs to own what happens next.

Where to start

If you're early in thinking about this, my advice is simple: start with one use case, pick something with a clear before-and-after measure, run it for ninety days, and evaluate honestly. Don't try to transform everything at once. The businesses that will use AI best are not the ones that moved fastest. They're the ones that moved thoughtfully.

← Back to Insights

I have facilitated peer advisory groups for over a decade. I've seen groups that changed the trajectories of businesses and the lives of the people running them. I've also seen groups that slowly became something people dreaded — a monthly obligation they tolerated until they finally found a polite reason to quit.

The difference between those two outcomes is not luck. It's not the quality of the members, though that matters. It comes down to a handful of things that are entirely within a facilitator's control — and almost entirely ignored by groups that fail.

Most peer groups fail for the same reason

They become comfortable. Comfortable is the enemy of useful.

It happens gradually. Members stop bringing their real problems because the group has drifted toward safe conversation. The hard questions go unasked. The honest pushback gets replaced by supportive nodding. The meeting starts to feel less like a board of trusted advisors and more like a pleasant breakfast club.

What actually makes a group valuable

Confidentiality that is real, not assumed. Members will not bring their actual problems — the family conflict, the underperforming executive, the financial stress — unless they genuinely trust that what is said in the room stays there.

A facilitator who does not manage to please. The facilitator's job is not to make members feel good. It is to make the group useful. That means asking the uncomfortable follow-up question, slowing down the advice-giving, and occasionally naming what's happening in the room when the conversation has gone polite.

Members who are genuinely peers. "Peer" doesn't just mean similar title. It means similar level of pressure, similar level of accountability, similar willingness to be vulnerable.

Issues that are real, not curated. The most valuable meetings I've facilitated started with someone saying something they weren't sure they should say. A member willing to bring a genuinely unresolved problem gives the whole group permission to do the same.

Accountability between meetings. The groups that create the most lasting value are ones where members have made specific commitments and are expected to report back. The accountability is not punitive. But it is real.

The thing no one talks about

Running a peer group well requires genuine care for the people in the room. Not just facilitation skill. Not just a good agenda. Real investment in the problems members are facing, real curiosity about their businesses and their lives, and real willingness to stay in the conversation when it gets hard.

Members feel the difference between a facilitator who is going through a process and one who is genuinely present. The latter produces groups that people have been part of for eight, ten, twelve years — not because they have to be, but because they can't imagine making major decisions without them.