2026-04-22

What We Cut in the First 90 Days

A typical $10M–$50M operator carries 30–40% operational waste. The first 90 days of an Axis Method engagement cut waste through removal, not addition. Here is what gets cut, and why.

By Cedric Corbett   ·   Axis MethodOperational wasteFractional COO

The first 90 days of every fractional COO engagement at OAG are a removal exercise. We don't install. We don't add. We cut.

That sequencing is deliberate. Operational waste compounds because companies keep adding, new tools, new processes, new approval steps, and rarely retire what they replaced. By the time we arrive, a typical $10M–$50M operator carries 30–40% waste in dollars, calendar time, and cognitive overhead. The fastest way to operating capacity is to cut, not to add.

This post is the catalog of what gets cut in those 90 days. I'll be specific about how the cuts get identified and why each one matters.

Week 1–2: The SaaS audit.

The first thing I ask every new client to send me is the credit card statement for the last 12 months, every recurring software charge. Not the AP-system view, not the contracts file. The actual statement.

The reason: contracts file what you signed up for. Credit card statements show what you're actually paying for. The delta between the two is usually 15–25%. The most common findings in week one:

  • Tools nobody uses anymore. The marketing automation platform from 2022 that got replaced in 2024 but never canceled. The SEO tool the previous head of marketing used. The reporting dashboard the founder built once.
  • Per-seat charges for inactive users. Former employees still on the seat list. Contractors who finished their project. Sales reps from the team that got reorganized.
  • Duplicate tools doing the same job. Two CRMs because one team migrated and the other didn't. Two scheduling tools because two functions each picked their own.
  • Annual contracts on tools you'd cancel monthly. Tools that priced you to "save money" by going annual, that you wouldn't have paid for at all if billed monthly.

For most clients, the SaaS audit cuts $3K–$15K per month in week one. Just from canceling.

Week 3–4: The decision-routing audit.

Here we map every recurring decision in the business and ask: who actually makes it? The output is a graph of "decision routes." The pattern that always shows up: 60–80% of decisions route through the founder by default, and most of those shouldn't.

Specifically, three categories get cut:

  • Decisions that have an obvious correct answer. "What pricing do we give this 3-unit SKU customer?", there is a written pricing matrix; the rep should reference it without asking.
  • Decisions that should belong to the function, not the founder. Hiring a junior role. Approving routine vendor renewals. Choosing between two operationally-equivalent vendor options.
  • Decisions that nobody actually needs to make. "Should we do the X initiative?" when nobody has resourced X. Deciding by inaction is fine; the meeting to decide is waste.

We don't fix routing in this phase. We just measure it. Founder-bottleneck count is one of the three numbers from the operational waste audit, and seeing the count is usually enough motivation for the founder to cooperate with the routing fix in the next phase.

Week 4–6: Process cuts.

Now we cut workflows. The criterion: any process that produces output nobody acts on, or that adds approval steps the risk doesn't justify.

Common cuts in this phase:

  • Reports nobody reads. The weekly P&L summary going to 12 people, of whom 2 actually open it. We cut the distribution to 2 and shorten the report.
  • Approval steps for low-stakes decisions. Three-signature approval on $500 expenses when the embezzlement risk is zero. We raise the no-approval-needed ceiling.
  • Standing meetings without an agenda owner. The "weekly leadership sync" that became a calendar fixture three years ago and now mostly recaps emails everyone already read. Either cancel it or assign someone to run it with an agenda.
  • Multi-step handoffs that exist because the systems don't talk. If the sales-to-ops handoff requires a manual ticket and a Slack message and an email, that's three steps. We collapse to one, usually by fixing the underlying integration in the StackOS phase later.

Week 6–8: The headcount audit.

This is the most uncomfortable cut, and the one most engagements get wrong. The goal is not to fire people. The goal is to identify roles whose work has been replaced or rendered unnecessary by the cuts above.

Common findings:

  • The role that exists to maintain the SaaS we just canceled.
  • The role that exists to run the report we just cut.
  • The role that exists to route the decisions we just devolved to the function.

The pattern: we don't fire those people. We move them to higher-leverage work. There is always more high-leverage work than there is bandwidth. The audit identifies who is currently doing low-leverage work; the next phase is repositioning them, not removing them.

Engagements where the founder skips this audit (out of conflict aversion, usually) are the ones where the operating system doesn't compound. If the same headcount is doing the old low-leverage work plus the new high-leverage work, the new work just doesn't happen.

Week 8–12: Stabilize the highest-leverage broken thing.

By week eight we've cut 30–40% of waste through removal alone. Now we fix one thing, the single highest-leverage broken process. Always a hand-off, almost never a tool.

Common targets:

  • Sales-to-ops hand-off. The gap between "deal signed" and "work begun." Usually 3–14 days of latency, with information loss in the middle.
  • Ops-to-finance hand-off. The gap between "work completed" and "invoice sent." Often 5–30 days, costing real working capital.
  • Hiring-to-onboarding hand-off. The gap between "offer accepted" and "productive on the team." Routinely 30–90 days at companies without a documented onboarding playbook.

One process, fixed. Not all of them. The point of stabilize is to get one piece running cleanly so the rest of the operating-system installation has a stable platform to build on.

What's left at day 90.

Three things:

30–40% lower operational waste, measured the same way it was measured at week one. Same SaaS spend audit, same decision-routing graph, same process-bloat count.

One process running clean, the highest-leverage hand-off, fixed and documented in interim form.

A diagnosed company, the leadership team understands where waste was, why it accumulated, and how the next stage of the engagement will install permanent operating discipline.

Day 90 is not the end of the engagement. It's the end of the cutting phase. The next 90 days are documentation. The 90 after that are hand-off. The full Axis Method is six to nine months. But the cuts in the first 90 are what makes the rest possible.

Doing this without us.

Most of these cuts are doable solo. The hardest part is not the analysis, it's the conviction to actually cancel the SaaS, kill the meeting, raise the approval ceiling, move the person. Founders know which cuts are right; they pay a fractional COO partly to be the third party who makes it socially possible to execute them.

If you want help, that's what the engagement is for. If you want to try it yourself, the audit reading guide gets you most of the way.

Related

Keep reading.

Talk through it.

If any of this is applicable to where you are, book a scoping call. No pitch deck.

Book a call →