You Keep Buying The Same Process Under Different Names

Summary

Every change effort in this business fails for the same reason, which is that it was started at the wrong layer. An operator who decides he needs a rebrand, a turnaround, or a new concept has usually already skipped the only question that governs the outcome, which is what his operation is for and whether anything he currently designs follows from that answer. The objective written on the invoice does not change the causal order underneath it. Change entered below the cause layer produces a new surface on an unchanged engine, and the engine wins every time.

The question almost always arrives already answered. You do not call and ask what is wrong. You call and say you need a restaurant turnaround. Or a rebrand, a concept change, a repositioning, a refresh, a second concept in the same box. The diagnosis is in the request, and what you are shopping for is the method that matches it.

The industry is built to sell you that match. There is a rebrand process, and a turnaround playbook, and a reinvention framework, and a repositioning methodology, and each one arrives with its own steps, its own vocabulary, and its own invoice. The premise underneath all of them is that different objectives require different processes.

That premise cannot be true, and the reason it cannot be true is sitting in your own trade area. Take two operators on the same block, in the same band, in the same year, with the same rent structure, the same labor market, the same vendors, the same weather, and the same economy. One is holding. One is not. Every condition anyone would name as the cause is held constant between them. A list of identical conditions cannot explain a difference in outcomes. Whatever is producing the difference is inside the operation, and it is producing it the same way in both buildings, in opposite directions.

There is one process. It is the five fundamentals, run in causal order. What you are trying to accomplish changes the answers. It does not change the process, and there is no separate method hiding behind any of the names you were shopping for.

It Is Not Semantics Because It Sends You To A Different Room

The first defense against that is that it is a word game. A turnaround and a new build are obviously different work, so calling them the same process is a rhetorical trick.

Test it by where each framing sends you to work on Monday. If a rebrand is its own process, you start with the name, the look, the menu architecture, and the positioning statement, and you hire out accordingly. If there is one process, you start by stating what the operation is for, and you do not touch the name until you know whether the answer changed. Those are two different rooms, two different budgets, and two different people you call first.

Only one of those rooms holds a lever you still own. The design firm’s room holds a surface you can buy. The other room holds the read that decides whether the surface will hold, and nobody can sell you that read, because it is your own. The distinction is not semantic when it determines which of those two things you spend your money on.

That is the whole argument, and the principles below are the mechanism underneath it. The terms themselves live in my Knowledge Base, and the architecture gets taught in full on my other surfaces. What sits on this page is the read, and what the read makes me do when I am sitting across from you.

The Objective Renames The Work, It Cannot Change It

My framework rests on a natural law. Every outcome inside a restaurant traces to one of two causes, design or default, and there is no third cause. That is [The Summers Principle], and it is the reason the objective on your invoice has no power over the sequence.

Work it through. A change effort is an attempt to produce an outcome. Every outcome traces to design or default at a cause layer. The cause layers are the fundamentals, and their order is fixed: right decision-making demands the right perspective, right perspective frames the right product, right product attracts the right cast, right cast executes the right performance, right performance engineers the right profit. That cascade is closed. Nothing in it asks what you were attempting. The law admits no exemption for intent, and no separate physics for an operator who happens to have called his effort a rebrand instead of a turnaround.

Which means the five separate methods on the market are not five methods. They are one cascade, cut into pieces, with each piece sold to the objective that made the buyer show up.

And the read sitting at the top of that cascade is not a mood or a mission statement. It resolves to one question with two available answers. Either you are asking what this operation costs you, or you are asking what it produces. That is [Two Roads], and it is the only fork in my framework that changes every answer downstream of it, because the first question turns every decision into a subtraction to be managed and the second turns it into an investment to be judged. Two operators can run the identical sequence, use the identical vocabulary, hire the same firms, and end up in opposite buildings, because they answered that one question differently and neither of them ever said it out loud.

That is also the part nobody shops for. You called about a restaurant turnaround. Nobody calls to ask which road he has been on for the last nineteen years, and that is the answer that has been producing the outcome the whole time.

That is why I do not ask what kind of engagement you want. I ask what the operation is for, and I ask it first whether you called about a failing store or a fourth location.

The Order Is Causal, Not Chronological

Here is where most people take a true sequence and turn it into a false one, and I want to be precise because the error is expensive.

The cascade is a statement about what causes what. It is not a set of project phases. You do not get to suspend Performance and Profit for a quarter while you rebuild Perspective, because you open tonight either way, and the cast still has to work a Saturday while you are thinking. Any consultant who hands you a five-phase plan where the money and the execution wait their turn has given you a document, not a process.

What causal order actually requires is that you intervene upstream first, while everything downstream keeps running. You change the read. The design follows from the read. The cast gets built to the design. The execution gets held to the cast you built. The margin follows from the execution that held. All of it stays live, and the change propagates in one direction because causes do not follow their effects.

Run it the other way and you get the thing you have watched happen to other operators. Start at the margin and the profit chase eats the cast, the cast erodes the product, and the product empties the read until the operator is running on default and calling it experience.

That is why my engagements do not begin with a phase plan. A phase plan assumes I can stop your business while I work on it, and I cannot, and neither can you.

A New Build And A Turnaround Differ By One Thing

An operator opening in eleven weeks and an operator eighteen months into a decline believe they are in unrelated situations. They are in the same one, at different points.

Each fundamental has two halves. There is the diagnostic half, which reads what the current system is producing and names where it is running on default. And there is the design half, which builds the structure that produces what you actually intend. The turnaround runs both halves, because there is an existing operation producing existing outcomes, and something in it has to be named before it can be replaced. The new build runs only the design half, because there is nothing yet producing anything.

That is the entire difference. Not a different process. The same process with one half of it empty, which is also why the pre-opening window is the cheapest change effort you will ever run and the one most operators waste. Nothing has to be undone. Differentiation is a pre-opening decision, and the operator who defers it until the doors are open has converted a design problem into a diagnostic one and added a year of default to the bill.

That is why I take pre-opening work at the same depth as a rescue, and why I will not treat it as the lighter engagement. It is the same sequence with less resistance, and it is the only time an operator gets to run it clean.

A Rebrand Is A Product Decision Wearing A Strategy’s Clothes

The name, the look, the menu architecture, and the way the operation presents itself are Product. That is the terrain they live in, and it is a real terrain with real work in it.

What a rebrand cannot do is come with nothing above it. Product is caused by Perspective. If the read that produced the old presentation has not changed, then the new presentation is an effect with no new cause, and the unchanged cause reasserts itself in about two quarters. The signage is new. The operation underneath it is the one that produced the decline, and now it has a fresh set of expectations to fail against, which is worse than where you started, because you have spent capital teaching your market to look again.

There is exactly one case where a rebrand is the whole job. Your read is current, your operation is run, your standards are enforced, and what the building says you are no longer matches what you actually produce. That case is real. It is also rare, and it is the only case where running the upstream work returns nothing to change. A null result is a result, and getting it in writing is what makes the spend defensible.

That is why I will not scope a rebrand as a first engagement. I will run the read, and if it comes back clean, I will say so and tell you to go hire a designer, which is the shortest and least profitable version of that conversation available to me.

The Money Question Cannot Go First

Profit sits at the end of the cascade. That is what makes it the most misread fundamental in the business, because operators take last to mean least, and it means the opposite. Everything upstream is judged by what arrives there.

It also has its own upstream layers, and this is the part that gets flattened. Pricing, mix, cost architecture, capital allocation, and revenue design are all designed inside Profit. So Profit is not merely an effect you wait for. It is a terrain with structure of its own, and capital allocation lives in it. Which is why the question of what a change effort will cost and where the money comes from is a real question with a real home, and that home is not the front of the sequence.

When the money question goes first, it does not get answered. It gets used. It becomes the reason the read never happens, because you cannot justify spending on something whose return you cannot model, and the read is exactly the thing whose return cannot be modeled in advance. So the operator funds the surface, which has a quote attached, and skips the cause, which does not.

That is why I do not open an engagement with your numbers. Your numbers tell me what already happened at the end of a chain I have not read yet, and if I start there I will end up doing what you can already get for less, which is telling you your labor is high.

Most Of What Gets Called A Restaurant Turnaround Is An Execution Problem

This is the one that costs me work, and I am going to say it plainly.

Covers are down and checks are soft, so the concept must be tired. That is almost never what happened. What happened is that line checks stopped. Pre-shift became an announcement. The bench was never built, so every schedule is a negotiation. The one standard you were actually known for stopped being enforced sometime in the last year and nobody wrote down the date. None of that reads as dramatic, and all of it reads on the P&L exactly the way a tired concept reads.

Reinvention is the most expensive way in this industry to avoid a conversation about execution, and it is attractive for a reason that has nothing to do with strategy. A buildout is easier than firing your brother-in-law. Capital work feels like leadership. People work feels like conflict. So the operator reaches for the version of the problem that can be solved with a contractor and a designer, and the industry is more than happy to confirm the diagnosis he arrived with.

I will include my own trade in that. A process that cannot look at an operator and tell him the crossing is not his problem is not a diagnostic. It is a funnel with steps printed on it. If every operator who calls is a candidate for the method being sold, the method is not reading anything.

That is why the first thing I do is try to disqualify the engagement you called about. If the chain is broken downstream, you do not need me to change what you are. You need to run what you have, and I will tell you that before you have spent anything with me.

The Hacks Are Priced Where You Can Reach And The Fix Is Not

There is a question the top of this piece left open. If the named products are all one process cut into pieces, why do you keep buying them. The answer is not that you are gullible. It is price, and the price is doing exactly what it was built to do.

Look at what the market actually puts in front of you. A menu refresh. A loyalty app. A marketing retainer. A mystery shopper program. A playbook with your objective printed on the cover. Every one of them is small, separately approved, and expensed in the month it lands. The work at the top is one number on one page, and it arrives as a capital decision you have to sit down and look at.

So the comparison you are actually running is not the hacks against the fix. It is this month’s four hundred dollars against a decision, and the four hundred wins. It wins every time it is asked, and it gets asked forever.

Nobody ever adds it up, and your own P&L is what prevents it. Marketing expense sits on one line. Software on another. Consulting somewhere else. Promotional discounting disappears into revenue. None of it is aggregated against the architecture you never bought, so the total does not exist as a number anywhere in your business. The instrument is not hiding it from you. The instrument was never built to show it to you.

That is the affordable version of the avoidance the last section prosecuted. Reinvention is the expensive way to avoid a conversation about execution. The hack roster is the cheap way, and because each purchase is small it never triggers the review a single large number would trigger. An operator can spend a decade like this and never once make a decision big enough to notice he made it.

That is why I ask for three years of tactical spend before I look at your current numbers. Not to embarrass the spend. To put a figure on a page that has never existed on a page before, so the comparison you have been avoiding becomes one you can actually run.

What I Will Not Take On

I will not take a change engagement that starts below Perspective. Not because of principle for its own sake, but because I would be taking money to produce an effect while leaving its cause in place, and I would know that going in.

I will not sell you a method named after your objective. There is no reinvention product here, no turnaround package, no rebrand track. There is one sequence and it does not get renamed to match what you walked in believing.

And I will not run the sequence for you while you watch. The read is yours. I can build it, hold you to it, and refuse the comfortable answer, but an operator who outsources his own read has bought a document that expires the day I leave.

The Diagnostic

Five tests. Run them in order. The first four take an afternoon and cost nothing. The fifth takes a night with your own records.

Test One, the held conditions read. Name two operators within your reach, comparable band, comparable format, who have been open as long as you have. Are they holding. If they are and you are not, the conditions you have been blaming are held constant between you and them, and the cause is inside your building. If none of them are holding either, you may genuinely be reading a market that moved, and that is the one answer that earns a crossing conversation.

Test Two, the stopped-enforcing read. Walk the last twelve months and name every standard that stopped being enforced, with a date for each one. Line checks, pre-shift, the bench, the one thing you were known for. There is no declining restaurant in which this list is empty. If you cannot produce a single item, you have not looked, and the test has not been run.

Test Three, the liveness read. Name the capability you would build a change on. Then answer three things about it: who is paid to produce it right now, what written standard governs it, and the last date somebody was held to that standard. Names and dates, or it does not count. A capability everybody remembers and nobody currently produces will pass every test anyone else gives you and it will not survive a new format, because the new format will not carry the people who made the old one true.

Test Four, the purpose read. State in one sentence what your operation is for. Then read it back and ask whether a contractor could have answered it. If the sentence describes a format, a price point, or a cuisine, it is a build spec and not a purpose, and you do not currently have the thing that causes everything downstream of it.

Test Five, the tactics total. Pull three years of every dollar that left the business aimed at demand rather than architecture. Marketing retainers, agency fees, software subscriptions, loyalty programs, promotional discounting, refreshes, printed playbooks, consultants who left you a document. One number. Put it next to what running the sequence from the top would have cost you in year one. I do not know what your number comes to and I will not pretend to, but I have never watched an operator run that total and come away still believing the small purchases were the cheaper path.

How the score sorts. Failures concentrated in Two and Three mean your problem is execution and every dollar of change spend is avoidance. A clean Two and Three with a failed Four means the operation is run and the read is missing, which is the real crossing and the hardest work of the four. A clean Four with a failed One means you are running a current read correctly into a market that left, which is the only case where changing what you are is the actual answer. Clean across all four, and what you have is a presentation problem, which is a designer’s invoice and a good afternoon. Test Five does not sort you anywhere. It prices whatever answer the first four gave you, and it is the test that decides whether you act on that answer or spend another year not acting on it.

What You Do Monday Morning

Write one sentence naming what your operation is for. Not what it serves, not who it serves, not what it costs. What it is for.

Under it, write the last three operating decisions you made. Any three, as long as they are the real last three and not the three that flatter the sentence. A schedule call, a menu change, a hire, a price move, a vendor swap.

Then draw a line from each decision back to the sentence and see whether it actually follows from it. That is the whole exercise.

If all three follow, your read is live and running, and whatever is wrong with your business is downstream of it, which tells you where to work. If one or two follow, you have a read that operates when you are paying attention and defaults when you are not. If none of them follow, the sentence is not your read. It is a description you wrote for the exercise, and the thing actually causing your decisions has never been named, which means every change effort you fund from here is going to be an effect looking for a cause.

Keep the page. In ninety days, write the next three decisions under the same sentence and run it again. The change in that ratio is the only leading indicator of a read that is actually crossing.

The Closer

You have been shopping for a process, and the market has been happy to sell you one per objective. But there is only the one, and you already own the part of it nobody can sell you.

Forty-five years in this business and I have never once seen a restaurant change what it was by starting anywhere but the top. I have seen plenty of them buy a new surface and call it a strategy, and I have watched the old cause come back and take it apart on schedule.

And the question underneath all of it costs nothing to ask. What is this operation for, and which of the two roads has it actually been on. Ask it at the front end and it is a conversation and an afternoon. Answer it eighteen months into a decline and you are buying a second business at retrofit prices, with the room emptier than it was back when you could have afforded the work.

So stop buying the name of the thing. Run the sequence, in the order causation actually runs, and find out whether what you need is a crossing or a Tuesday.