Why Every Decision Ends Up With the Founder

When decisions have no designed path upward, they land on the founder. The team usually has the ability; what's missing is a written layer that says who approves what and up to what limit, and the first version takes an afternoon to build.

Why does every decision end up on my desk?

Decisions land on the founder when there's no escalation layer between the team and the top. Nobody can say who approves what, or up to what limit, so the only path that always works is upward. Writing down approval limits for each role builds that layer, and the first version takes an afternoon.

When decisions have no designed path upward, they take the only path that always works and land on the founder. The team usually has the ability to decide; what's missing is a layer that says who may decide what.

The Pattern

Picture a services firm at roughly ₹45 Cr. The org chart shows three layers, but the decision flow shows only one, because eleven people report to the founder in practice. The structure doesn't require that. It happens because nobody in the building can answer a plain question: who can approve what, up to what limit, and when does a decision move to the next level?

The symptoms are always the same. A pricing exception sits in a WhatsApp group for four days because nobody knows whose call it is. A delivery team sends a client complaint past the ops head straight to the founder, "in case they don't have context," and the ops head, bypassed twice, stops trying. A new hire learns inside a month that every real question ends at one desk.

The founder feels it as time: thirty to fifty decisions a week that were never meant to reach them. The business feels it as drag on everything else.

The Misread

Most founders read this as an ownership problem and reach for one of three fixes. All three fail, and for the same reason.

Delegation training tells the team to "own it." Two weeks later the team is back to asking permission, because "own it" comes with no definition. Nobody has written down what each person can approve, what budget they can clear, or how much client risk they can take on without checking.

Weekly alignment meetings work for about two weeks and then decay into status updates. The real decisions still happen between meetings, in hallways and direct messages.

Hiring a COO is the most expensive version of the same mistake. A COO with no decision structure to work inside becomes a more expensive single point that every decision routes through, so the bottleneck stays and only its title changes.

Two more fixes show up often in firms that cross Rs.50 Cr. A culture workshop assumes people would take more decisions if they felt more ownership, but ownership comes from authority plus accountability, and a workshop hands out neither. Written SOPs help with execution, yet the decisions that pile up at the founder are judgement calls, and judgement is exactly what an SOP can't hold.

Each of these fixes works on behaviour, and none of them touches the structure that makes the behaviour sensible.

The Mechanism

At ₹5 Cr with ten people, the founder holds every piece of context in their head, and being the decision point for everything costs nothing. At ₹50 Cr with seventy people, the same habit breaks down. There are now sixty decision moments a day the founder can't fully see, yet the business has no other defined layer, so the decisions keep travelling up. Revenue grew while the decision structure stayed the size it was at ₹5 Cr. The team has the capability; it lacks written permission.

Three forces hold the pattern in place:

  1. No written thresholds. Nothing states, for example, that the category head decides anything below ₹25,000. Every call is therefore a little ambiguous, and when things are ambiguous, the safe move is always to escalate.
  2. Shared ownership. When three roles own a category, none of them will carry a wrong call alone, so collective ownership ends up producing escalations instead of accountability.
  3. The training loop. Every quick answer the founder gives to a small, unnecessary escalation teaches the team that escalating pays off. The founder builds the pattern while trying to be responsive.

The team isn't being timid. It's behaving sensibly inside the system it was given.

Another way to see it is that the firm grew wide instead of tall. Revenue and headcount scaled, while the number of real decisions made each week without the founder in the room stayed close to zero. That count is a truer measure of a middle layer than the titles on the org chart, because a firm can have three heads of department and still have one point of judgement. Two signs make it visible. A single week of founder travel leaves three or four fires waiting for the founder's return, and promotions feel forced, because a new title comes with no new decisions attached to it.

The Method: A Decision Threshold Map

The fix is a two-page document with four defined parts, and it changes structure rather than attitudes.

Decision categories. List the categories that currently escalate, such as client exceptions, pricing, hiring, vendor payments, delivery changes and strategic commitments, and name each one explicitly.

Authority thresholds by level. For each category, define what a level can approve outright, what it must report upward without waiting, and what has to escalate. Set the limits by amount, by risk and by client impact. Give the authority to the level rather than to the person, so it survives when someone leaves.

Escalation triggers. State the exact condition under which a decision moves up, and the response time expected at each stage. Ambiguity produces escalation, and a clear trigger removes the ambiguity.

Routing paths. For each category, name who receives the escalation, through which channel and with what supporting information. A decision with a known destination stops drifting to the founder by default.

Two habits make the map hold:

Single ownership per category. One named person owns the decision and its outcome, rather than a team or a committee, because shared ownership removes the accountability that lets someone make an independent call.

Review instead of approval. The founder steps out of the approval chain for individual decisions and reads a weekly summary of decisions made without them. The founder reviews patterns instead of single calls, and that moves the founder from gatekeeper to designer of the system, which is the only seat worth holding at this size.

Treat it the way the business already treats money. Every firm at this size has financial controls that say who can sign a cheque for what amount, but almost none have decision controls written at the same level of detail. The founder can build the second set the same way the first was built.

The Audit Loop

A map that nobody checks slides back into "ask the founder." So the loop gets closed with a light tool, an escalation log. Every escalation records the date it was raised, who raised it, who it went to, how severe it was, the expected and actual resolution dates, and one field that matters more than the rest: was the protocol followed, yes, no or partly?

The log audits the map, and the people are never on trial. A cluster of "no" answers in one category shows that a threshold is wrong or a trigger is unclear, and the fix goes into the document. That cycle of correction is what turns a document into an operating layer that holds.

What the Install Looks Like

The change can move quickly because it's structural, so nobody has to be retrained and no reorg is needed. A founder who clears thirty to fifty small approvals a week can write the thresholds and name a single owner for each category in an afternoon. From that week on, the escalation log shows whether the volume reaching the founder is actually falling, and which categories still leak upward.

The Principle

If the team can't say, in one sentence, who owns decisions at each level and what moves to the next level, the business has a missing layer, and delegation advice won't supply it.

The businesses that grow without breaking have built a path for decisions to move without the founder touching every one. Better people help, but the path is what lets them work.

Why This Is the Real Work

I build operating models for a living, and this is the clearest example of something I keep coming back to: the operating model is the asset. A founder who is the operating system has built a company that can't be sold, can't be left and can't grow past the founder's own calendar. A founder who has put the layer in place has built something that runs when the founder isn't in the room. The escalation layer is a small document, and what it really buys the founder is choice.

In my experience, two pages like these change more about how a company runs than months of coaching do, so start with the map. If your week is full of decisions that shouldn't need you, look at the layer between you and your team before you look at the team.

This is the first in the MetMov Method Memos series: fortnightly working notes on the operating layers that let founder-led businesses run without the founder in every loop.

Pick a time that suits you.

Book a call