A piece about a restaurant labor model went around this week. Operators shared it. A newsletter summarized it. It read as news, and it was published on September 5, 2024.
Two years and twelve days old. Nobody in the chain checked the date. Not the newsletter, not the operators who shared it, and not me on the first read, and I have forty-five years in this business and a searchable archive of my own positions sitting open on the desk.
That is the part worth stopping on. If the date got past me, it got past everyone it reached.
Follow how it got to you. A trade publication runs an operations story. A vendor newsletter summarizes the trade story. A feed post summarizes the newsletter. An operator shares the feed post because the headline is about something on his mind that week. At no point in that path does the publication date travel with the claim. The trade piece has a date on it, in grey, under the headline. The summary does not carry it. The feed post does not carry it. By the fourth hop the claim has no age at all, and a claim with no age reads as current, because current is the default state of anything that just arrived.
That is Editorial Capture doing what it does. What reaches an operator about his own industry is supplied by parties whose business is circulation, not accuracy of timeline. Circulation rewards a story that reads as happening and does not reward one that reads as finished.
And it collides with Constant Expiry. Every operating fact in this business has a shelf life. Conditions move, the read goes stale, and an operator running last year’s conditions is running against a market that no longer exists. Date-stripping is the mechanism that defeats that discipline, because you cannot check the expiry on information that arrives with no date on it.
Now the story itself, which matters, because unlike almost everything that reaches you in this business, this one has audited results attached.
Shake Shack rebuilt how it deploys labor. They had been allocating hours against sales dollars, which breaks the moment your mix is uneven, since a handspun shake and a chicken sandwich do not consume the same labor per dollar of revenue. They moved to allocating against activity instead. That is a correct change, it is also arithmetic, and the trade wrote it up in September of 2024 with the CEO calling it a game changer.
Then 2025 happened, which was the first full year under it, and three more quarters after that.
The cost side worked. Fourth quarter 2025 labor came in at 25.4% of sales, a 150 basis point improvement. Restaurant-level margin expanded 120 basis points to 22.6% for the year. Compliance with the labor guide went from roughly half their restaurants to consistently above ninety percent. That is real execution and it is better than most of what you will see in this industry.
The demand side did not. Full-year same-store sales grew 2.3% against blended pricing near five percent. Fourth quarter traffic was up half a percent. First quarter of this year traffic was up 1.4%, inside a quarter the company told analysts weather cost them 240 basis points, which makes the disturbance nearly twice the size of the result. April went to negative 0.6%.
Read those two paragraphs together. Two years of the most disciplined labor redeployment in the segment produced a cost recovery on volume the company already had. Nobody decided to come more often. That is Repair Work executed at a high level, which is honest work and worth doing, and it is a failure in exactly one respect: it got delivered to you as a revolution.
Then watch which result gets stated precisely. The labor improvement is 150 basis points. The margin expansion is 120. The traffic result is positive. One side gets decimals and the other gets a direction, and the reader carries away an impression of the whole business built out of the one part that was measured to two places. That is Measurement Asymmetry in a public disclosure.
The streak works the same way. Twenty-one consecutive quarters of positive same-store sales. A streak counts quarters, not growth. Comps of 2.3% against pricing near five percent means you can hold that streak indefinitely while selling fewer things to fewer people, and the number will never turn negative to tell you so.
The hospitality claim does not survive the same read. The public position is that the model is not about cutting labor, that cost reduction is an outcome rather than a goal, and that the point is putting the right people in the right roles at the right times to drive hospitality. The same coverage reports the company operating with fewer labor hours. Both halves get sold at once and only one of them shows up in basis points.
So look at what is offered as evidence for the hospitality half. Wait times down from about seven minutes to under six. Retention up nearly forty percent since 2023. The retention number is real, it is a People result, and it is the most valuable thing in the entire disclosure. The wait time is service speed. You execute service. You produce hospitality. They are not the same act and they do not come off the same instrument. Neither figure tells you what one Guest decided about coming back.
Now the part nobody in that chain is paid to say.
The trade publication runs on access and on vendor advertising. Prosecuting the timeline costs you the next interview and prosecuting the instrument category costs you the advertiser who sells it. The counsel class needs the material, because a two-year-old deployment change repackaged as current is a free deck and a free conference session, and if it is old news the deck is worthless. The operators with enough tenure to catch a 2024 date are in their buildings at eight o’clock at night, not publishing. And calling it out means naming a party, and every party in that chain is a potential client, a referral, or a host. Reposting has never once cost anyone an invitation.
Here is what verifying it required. Read the publication date. Pull two earnings calls. Find the labor line, the margin, the comp, and the traffic. Check the hospitality claim against the labor-hours statement in the same coverage. Ninety seconds with instruments every person in that chain already has.
For most of my career, checking a claim like that meant knowing where to look, having the archive, and spending an afternoon. That cost is gone, and the checking rate did not move. Which tells you the bottleneck was never the work. It was that checking costs something, however small, and repeating costs nothing.
So the architecture is the other half of this, and it is small enough to start on Monday.
Four of the five sources of knowledge about your restaurant expire. Method knowledge, which is how to do it, is the only one that can be manufactured before it reaches you, which is why it is the only one for sale. Condition knowledge is what is true in your market right now. Operating knowledge is what your operation is actually doing. Return knowledge is what it turned out to be worth to the people paying. Positional knowledge is what you have decided the operation is for, and it is the only one on the list that is supposed to hold across periods.
Run the Extraction Test on each. The test is one question: can this exist without your operation running. Operating knowledge fails it, because it does not exist until you run a period. Return knowledge fails harder, because it exists only inside a Guest’s decision and reaches you only through a person who was standing there. Those two are irreducibly local and irreducibly current, which makes them the only two where you hold an advantage over everyone selling to you.
And every read you produce gets three things attached at the moment you produce it. The period it covers, which is the window of operating it describes rather than the day you wrote it down. The conditions it was produced under, which is what makes it comparable to the next one. And the expiry, which is when it stops describing your operation. A read with no stated expiry never expires. It just quietly stops being true.
The most useful sort in the business follows from that, and almost nobody runs it. Any improvement lands in one of two places. It either recovered margin on volume you already had, or it produced volume you did not have. Recovery runs out, because there is a finite amount of margin sitting in a loose spec or a bad labor model and once you have collected it, it is collected. New demand has no ceiling, because a Guest who came back and brought two people changes the volume your whole cost structure runs against, and that Guest is available again next period. An operation that has been collecting recovery for two years and reading it as growth will keep spending on the recovery side long after the recovery is gone, because the numbers kept improving right up until they stopped.
Read the prosecution: https://hacksterism.jeffreysummers.com/the-revolution-you-just-read-about-ended-two-years-ago
Read the architecture: https://physics.jeffreysummers.com/every-read-you-own-has-a-date-on-it
One move this week. Open the last ten industry items in your saved list or your inbox and write the original publication date next to each one, not the date it reached you. Then count how many you had been treating as current. That number is your exposure, and it is the only number in any of this that describes your operation rather than somebody else’s.
Then the rule, which costs nothing: no operating claim gets used until you have its date and its period. Not its headline. Its date.
— Jeffrey
Digging Deeper
Every term used above is defined in my Knowledge Base:
https://kb.jeffreysummers.com/
Terms used: Editorial Capture, Constant Expiry, Measurement Asymmetry, Repair Work, AI As Amplifier, Fundamental Knowledge, Extraction Test, The ReRead, Causal Read, Value Congruence, Operating Helix, Default Gravity, Sameness Machine
The architecture taught in full, fundamental by fundamental:
https://physics.jeffreysummers.com/
The Road 1 arbitrage prosecuted where it lives in the wild:
https://hacksterism.jeffreysummers.com/
How my thinking shapes the work:
https://jeffreysummers.com/
Sources Cited In This Piece
Restaurant Business, September 5 2024 — original labor deployment model story, CEO “game changer” characterization — https://www.restaurantbusinessonline.com/operations/shake-shack-reinvents-labor-deployment-model
Shake Shack Q4 2025 earnings call — labor 25.4% of sales, 150 bp improvement, restaurant-level margin 22.6% up 120 bp, labor-guide compliance above 90%, Q4 traffic +0.5%, full-year comps +2.3%, blended pricing — https://www.fool.com/earnings/call-transcripts/2026/02/26/shake-shack-shak-q4-2025-earnings-transcript/
Shake Shack Q1 2026 earnings call — comps +4.6% with +3.2% price/mix and +1.4% traffic, 240 bp weather impact, 21st consecutive positive quarter, April comps -0.6% — https://www.fool.com/earnings/call-transcripts/2026/05/07/shake-shack-shak-q1-2026-earnings-transcript/
NRN, operational improvements 2025 — “not about cutting labor,” right people right roles right times, wait time under six minutes from about seven, retention up nearly 40% since 2023 — https://www.nrn.com/fast-casual/operational-improvements-lead-to-robust-2025-for-shake-shack
Fast Casual — company operating with fewer labor hours after the move to activity-based labor — https://www.fastcasual.com/news/shake-shacks-operational-excellence-fuels-earnings-beat/