NAMED OFFERINGS · NEW Fractional CMO from $2,500/mo. Plus the 30-day AI Adoption Sprint at $25k. See both paths
Innovative Group · Ground Truth · July 2026

Non Mihi Soli.

Not for myself alone. Why expertise trapped in one head is a business risk, and how to systemize what your experts know into a repeatable system the whole team can run.

Non mihi soli means "not for myself alone." It is old shorthand for a simple idea: what you know is worth more once other people can use it. Last time I argued that your best content is locked in your experts' heads and the job is to get it out. This is the step after that. Capturing what one person knows only pays off when the knowing becomes a system anyone can run. A skill that lives in a single head is fragile. Written into a workflow a teammate can follow, it turns into something the whole program can lean on.

AudienceFounders, marketers, web and growth operators
AuthorAshlesha KhondAshlesha Khond, Innovative Group
PublishedJuly 27, 2026
The thesis

Systemize the judgment first, share it second, then step back.

The most valuable thing an operator builds is the system that lets the next person do the work the same way. Knowledge written into a checklist, a tracker, or a documented workflow turns one person's memory into team throughput. Systemize the judgment first, share it second, then step back and let the system run.

The scarce skill is turning the way you work into something anyone else can pick up and run.

01

The bus factor is a risk worth measuring

Every team has a bus factor: the count of people who could vanish before a critical process grinds to a halt. When that number is one, the program sits one resignation, one illness, or one overloaded quarter away from losing something it cannot quickly rebuild. In my last column I made the case for getting expert knowledge out of people's heads and onto the page. The reason it matters this much is that undocumented knowledge is fragile by default, and the fragility has a price you are already paying.

That price shows up long before anyone leaves. The McKinsey Global Institute found that the average interaction worker spends nearly 20 percent of the workweek, close to a full day, looking for internal information or tracking down the colleague who holds it (McKinsey Global Institute, 2012). That is the tax a team pays when what it needs lives in people and never makes it into a system. I came to marketing from engineering, where a single point of failure is something you log and design out on purpose. A process only one person can run is the same defect wearing a job title.

A process only one person can run is a risk with a deadline you cannot see.
Try thisPick your most important recurring deliverable and ask one question. If the person who runs it were out for a month, could someone else pick it up from what is written down? A "no" is the first thing on your list to systemize.
02

Judgment is the part worth writing down

The instinct when documenting a process is to list the steps. Steps are the easy part. The part worth capturing is the judgment underneath them: why each step exists, what you are checking for, and what you do when the situation does not match the happy path. A checklist that records reasoning survives a change of staff and travels to hands that were not in the room when the decision got made.

Early in my career as an engineer I logged a lot of defects, and the entries that helped the next person recorded the cause and how to recognize it again, so the fix could be repeated by someone who had never seen the original failure. Documentation works the same way. The knowledge that makes a company worth citing in an AI answer is the same knowledge that makes a process worth trusting: reasoning you can see. It is the bar we hold our own AEO work to, and it is why a good runbook reads less like a recipe and more like a short explanation of how an expert thinks.

Write down the reasoning behind each step. The clicks will change, and the reasoning is what lets the next person adapt when they do.
Try thisTake a process you already documented and read it as a brand-new hire would. For every step, check whether it says why. Where it only says what to do, add the one sentence that explains what you are actually checking for.
03

A system someone can run beats a document nobody opens

A document that captures everything and then sits in a folder no one opens has close to zero leverage. A system earns its leverage by living where the work happens: the tracker that flags what is overdue, the template that starts a task half-finished, the recurring calendar hold, the definition of done a task cannot pass without meeting. The knowledge stops being something you remember to go and find and becomes the path of least resistance for getting the job done.

That is how the teams in Innovative Group's digital marketing and technology practice keep multi-account programs moving. The workflow carries the standard, so quality does not rise and fall with whoever happens to be on the task that week. Building those trackers and workflows is most of my own job, and the lesson that keeps proving itself is unglamorous: a system a tired person can follow on a bad day beats an elegant one that depends on everybody being sharp.

Shelfware waits to be found. A system meets the work where it already happens.
Try thisTake your best-documented process and move one piece of it into the tool where the work actually happens. Turn the top checklist item into a required field, a template, or a recurring task. Embedded beats archived every time.
04

The agent era rewards the knowledge you have already systemized

There is a timely reason to do this now. The software market is reorganizing around AI agents that execute whole workflows, and an agent can only run a process that has been made explicit. On July 1, 2026, Gartner projected that up to $234 billion of enterprise application spending will be exposed to "agentic arbitrage" by 2030, as agents complete tasks across systems and the interface stops being the point (Gartner, 2026). The same analysis makes the quieter point that matters here: better outcomes from AI require systems that can retain deep institutional memory over time.

Most of that memory is not written down anywhere an agent could use it, which is the catch. Gartner also predicts that more than 40 percent of agentic AI projects will be canceled by the end of 2027, citing unclear value and inadequate controls (Gartner, 2025). An agent handed a vague, undocumented process fails the way a new hire would, only faster and at scale. The teams that get real leverage from agents are the ones that already turned their expertise into explicit, governed workflows. Systemizing your knowledge has quietly become the prerequisite for automating any of it, which is the groundwork a Next Best Action program lays. Once a workflow runs on its own, the discipline moves to measuring what it produces, the kind of tracking we document in the Insights library.

An AI agent can only run a process you have already made explicit. Everything still in someone's head stays manual.
Try thisBefore you pilot an agent on any workflow, write that workflow down to the point a competent stranger could run it by hand. If a person cannot follow it, an agent will not either.
05

Sharing the method is where the leverage compounds

Systemizing your own work is step one. The leverage arrives when you share the method so other people can run it without you in the loop. Holding onto hard-won know-how feels like protecting your value, and it quietly caps it, because knowledge that never leaves you can never scale past your own hours.

This is the move from being the person who does the work to being the person who makes the work repeatable. A fractional marketing leader earns their keep on exactly that: the value is the system left behind, the trackers and playbooks a team keeps running long after the engagement ends. If your best operator's knowledge only produces output while they are personally in the room, the program has bought labor that stops the day they do. Turning that knowledge into a system others can run is what converts it into durable leverage. When no one owns that translation yet, it is the kind of engagement we take on at IG.

Knowledge kept to yourself lifts one career. Knowledge built into a system lifts the whole team.
Try thisName one thing you are the only person who can do. Book thirty minutes to write down how you do it, hand it to one teammate, and have them run it once while you watch. The gaps they hit are your real documentation backlog.
Appendix

What I refuse to call done.

  • A process only one person on the team can run.
  • A tracker or workflow with no named owner.
  • A "system" that lives in someone's memory instead of a page a teammate can open.
  • A document written once and never updated after the process itself changed.
  • Knowledge captured for an audience of one: the person who already knew it.

Not for myself alone

Non mihi soli. The phrase is a reminder that competence kept private is competence half-wasted. The knowledge that makes you good at your job is worth more the moment it stops depending on you being in the room. Write down the judgment, embed it where the work happens, and hand it to the people around you. That is how one person's expertise becomes a program that keeps shipping, and increasingly, the raw material an AI agent can actually run. The work outlives the week you did it, and it outlives you being there to do it again.

Next in Ground Truth: the synthesis. What changes when the same person has worked as an engineer, a program manager, and a marketer.

Frequently Asked Questions

The bus factor is the number of people who would have to leave before a project or process stalls. A bus factor of one means a single person holds knowledge nobody else can act on, so the work is exposed to any absence, planned or not. It matters because it turns ordinary events, a resignation or a long holiday, into operational risk. Lowering it means writing down what that person knows and giving the process a second owner.

Put the documentation where the work happens and capture the reasoning, not only the steps. A process embedded in the tracker, template, or recurring task people already touch gets followed, while a file in a shared drive gets forgotten. Record why each step exists so someone can adapt it when the situation shifts, and give it a named owner responsible for keeping it current.

A standard operating procedure is a documented, repeatable set of instructions for completing a recurring task the same way every time. A useful SOP covers the trigger that starts it, the steps in order, the judgment behind the tricky ones, and the definition of done. The goal is that a competent person who has never run the task can complete it correctly by following the document.

Systemize the repeatable work before you add headcount. Shared templates, trackers, and definitions of done let one standard run across many accounts without depending on any single operator's memory. Reserve senior time for judgment calls and let the documented workflow carry the routine execution. This is also what makes the work legible to AI agents, which can only run a process that has been made explicit.