Editor's Note
Welcome back to Savory Bites. Fresh intel for restaurant and hospitality operators who want to stay ahead of the tech reshaping our industry.
The whole argument for scale is that scale buys you a better price. Someone finally checked 93 million data points to see whether that's true, and it mostly isn't.
Then: Chili's just posted its 21st straight quarter of growth. Five moves are credited with getting them there, and two of them are things the brand stopped doing. Those two are the ones you can copy this week without spending anything.
Let's get into it.
Let's get into it.
— James, Founder of Savory Bites
This Week In The Industry
Buying in Bulk Doesn't Get You a Better Price. Somebody Finally Checked.
Buying in bulk gets you a better price. Every operator believes it. Most have built their purchasing around it.
It isn't true.
When two restaurants buy the same product, from the same supplier, in the same month, the bigger buyer paid less 43.3% of the time. It paid more 33.9% of the time. The rest of the time there was no difference at all. That's from an analysis of 93 million purchasing data points published this month by Supy.
Sit with what that undoes. Purchasing power is the whole reason a chain is supposed to beat you on cost. On this data it's close to a coin flip.
A real gap does exist. It just has nothing to do with size. Between two restaurants buying the identical product from the identical supplier in the same month, the higher-paying one typically pays 32% more. On one product in six, somebody pays at least double.
The cause is structural, and the article is blunt about it. No restaurant can see what any other restaurant pays. With no reference price, the price you're used to becomes the price that feels right.
Across 1.27 million invoices, 22.7% contained at least one line billed at a price that didn't match the agreed price. Of the lines billed above the agreement, nearly half were 20% or more over.
More than one invoice in five is wrong, and when it's wrong it's expensive. The reason nobody catches it is the other 77.3%, where everything is exactly as agreed. Three quarters of the work finds nothing, so spot-checking quietly stops being a habit, which is why the answer is checking every line automatically rather than checking harder.
On stock counts, restaurants counting fewer than eight times a year had accurate records on 24.5% of items. Monthly counters hit 50.6%. Weekly counters reached 76.2%.
Three things to do this week.
Get a second quote on your ten biggest-spend products, from a different supplier, inside the same month. Not spread across a quarter, or you're measuring price drift instead of the gap between what you pay and what someone else pays. This is the only way to build the reference price the market won't give you.
Count your twenty highest-value items weekly and everything else monthly. That's the pattern the report found among the businesses with the best numbers, and it describes a smaller job than the full monthly count it replaces. Treat the accuracy gap as correlation rather than a promise: weekly counters are probably different operators, not just operators on a different schedule. The direction still isn't ambiguous.
Check billed price against agreed price on every line, not a sample. Whatever tool does it, the rule is that a person handles the short list of exceptions rather than reading sixty lines hoping something jumps out.
One disclosure, because it matters. Supy sells restaurant purchasing software, and this analysis is their marketing. The findings point at problems their product solves. The data is published free at supy.io/the-spread with the methodology attached, and their monthly price index covers the UK, Australia, the UAE and Saudi Arabia rather than the US. Read it with that in mind. The bulk-buying finding is still the most useful thing published this month.
Feature
Chili's Just Hit 21 Straight Quarters of Sales Growth. The Two Cheapest Moves Behind It Are Both Subtraction.
Brinker reported sales up 5% and Chili's extended its streak to 21 consecutive quarters, on continuing positive traffic.
Owner.com, a restaurant marketing company, published a breakdown of five moves behind it: cut the menu, eliminate friction, create value, install loyalty infrastructure, invest in labor. They put the combined effect at $1.4 million in additional average per-location sales over three years. It's a vendor's framing of somebody else's turnaround, so hold the attribution loosely and take the list seriously.
Three of those five cost money or systems. Two of them are things Chili's stopped doing, and those two are the ones available to you this week for nothing.
Menu cutting is the piece independents resist hardest, usually for the same reason each time. Every item on the menu is there because somebody once asked for it. Removing it feels like telling a customer no. What it actually does is shorten prep, shrink your inventory count, reduce waste in the categories where waste is most expensive, and make every remaining plate come out faster and more consistently.
That connects directly to this week's lead. A shorter menu means fewer SKUs to price-check, fewer items in your weekly count, and a purchasing list small enough that getting a second quote on your top ten is an afternoon rather than a project. The menu is upstream of the cost problem.
Eliminating friction is the vaguest phrase on the list and the easiest to test. Stand at your own door at 7pm and count the steps between a guest arriving and a guest eating. Then count the steps between your kitchen deciding a plate is ready and that plate reaching a table. Most operations have accumulated a step that made sense once and now costs ninety seconds every ticket.
Worth setting against last week's story. Subway's answer to declining sales was seven more hours a week and every delivery app. We said then that the franchisees reporting gains from late hours were telling the truth, and they were. But their gains came from hours where demand already existed. Chili's spent three years removing things instead, and has 21 quarters behind it. Adding load only works where there's demand to catch. Removing it works anywhere.
If you do one thing: pull your item-level sales for the last 90 days and find the bottom 20%. Ask what each one costs you in prep time, inventory lines and station complexity, not what it earns. That's the list Chili's worked from.
Read more → https://www.restaurantbusinessonline.com/operations/how-chilis-won-21-straight-quarters-perfecting-basics
News Bites
🎃 Augtober Has Arrived and the Urgency Is Wearing Out Starbucks and Dunkin' have set fall drop dates and Krispy Kreme, Panera, Caribou, Ziggi's and PJ's are already out. The caution raised in Modern Restaurant Management is about frequency rather than timing: when consumers get a new holiday, flavor or limited-time offer every few weeks, they stop paying attention to any of them. If your calendar carries a promotion every month, the promotions have stopped being events. Source: Modern Restaurant Management
🏗️ The Hottest Restaurant Real Estate Is Somebody Else's Old Restaurant Ten years ago operators wanted a raw box to design from scratch. Now second-generation space is the premium product, because build costs and interest rates have made starting from nothing the expensive option. Last week Dutch Bros agreed to pay $105 million for up to 65 closed Salad and Go sites, with 7 Brew contesting it. The same logic applies at one unit. A corner that already has a hood and a walk-in is worth more than one that doesn't. Source: Modern Restaurant Management
🥗 What Actually Killed Salad and Go Restaurant Business published the post-mortem on the drive-thru chain that closed all 70 locations this month. The diagnosis is aggressive growth paired with a loss of focus on quality. We covered the closure and the real estate sale last week. This is the part worth reading if you're deciding whether to open a second or third location, because the failure wasn't the concept. It was the pace. Source: Restaurant Business
Tech Spotlight
A New POS Launched With a Six-Month Free Trial
Menu Tiger launched a cloud POS this week handling orders, payments, menus, inventory, sales and staff in one system, rolling out as a soft launch with a six-month free trial.
Normally a POS launch isn't worth your time. This one is, because of what the lead story asks you to do. Checking billed price against agreed price on every line, and counting twenty items weekly, are both jobs that need a system. Doing them on paper is how they stop happening by week three.
Six months free is long enough to find out whether the inventory side actually holds up under your volume, which is the only part that matters and the part demos never show. Run your top twenty through it before the trial ends and you'll know.
Ellie's Corner
Every week this space is dedicated to something we're building at Ellie Carte. An AI phone ordering and restaurant management platform built for independent restaurants and hospitality operators.
The lead this week is about numbers operators run on without knowing they're wrong, because nobody measures them. There's a revenue-side version. Your phone has a missed-call list, so you technically have the data. What it doesn't tell you is which of those calls was an order, what it was worth, or that six of them landed in the same twenty minutes on Friday. A number with no context never makes it onto the P&L, so it never gets managed. Ellie answers every call and logs what the caller actually wanted.
👉 Learn more at elliecarte.com
Never leave the table hungry.
Subscribe to Savory Bites → https://www.savorybites.ai/subscribe
Till next week — stay sharp, stay fed. 🍽️
Savory Bites | Fresh intel, served weekly