Once the loop is running, this is where you make it pay: cost your recipes and plates, engineer the menu, hand the work to a team without losing accuracy, run more than one location, and, when the day comes, take all your data with you. None of it is required to get value from Stockcount. It is what is here when you want more.
Chapter IV
IV
Going further
Cost a recipe (and keep it costed when vendors change prices)
A recipe in Excel is a snapshot. A recipe in Stockcount is a live calculation. The difference shows up at month-end.
Our take
Recipe costs go stale. Tomato paste goes up forty cents and nobody emails you. The menu analysis you paid a consultant for in March is out of date by June. Stockcount fixes this by treating every recipe as a live formula: the cost of a dish updates itself the moment a new invoice changes an ingredient price. You build the recipe once, tidy it up now and then, and stop getting surprised at month-end.
The catch: you have to build it right the first time. Two things to get straight — the two kinds of recipes, and how to set yield.
Two kinds of recipes
Stockcount draws a line most inventory apps don't:
Dish recipes: what you sell to guests. Burger. Salad. Cocktail. One portion = one plate.
Prep recipes: what you make in advance to use later. Marinara. Vinaigrette. Pickle brine.
The difference matters for counting. A prep recipe turns into a single item on your shelf that you count as one line (say, five quarts of marinara), instead of counting the raw oil and tomatoes separately. Stockcount handles both kinds, but you have to tell it which one you're building.
Simple rule: if you count it on the shelf, it's a prep recipe. If a customer can order it, it's a dish recipe.
How to build a recipe
Recipes → New builds a dish recipe. On the iPhone app, the cost updates live as you add each ingredient. On the web, it's calculated when you save.
Open Recipes → New.
Name it, then add ingredients. Pick each one from your catalog (your list of ingredients), and set the quantity and unit.
Save. Stockcount works out the plate cost — what the dish costs you to make. If you've linked the recipe to a menu item with a price, you also get the margin (more on that below).
Prep recipes start somewhere else. Because a prep recipe becomes an item on your shelf, you build it from that item. On the web, open the ingredient and add a prep recipe to it. Or in chat, just ask. Stockcount asks for the batch size and what goes into it. (The New Recipe form on the iPhone app builds dish recipes only.)
Doing a lot at once? In chat, paste a list of recipes, or upload a spreadsheet or a photo of them. Say something like "Add these Sunday brunch recipes." Stockcount goes through each one, matches the ingredients to your catalog, and checks with you before creating anything new.
Plate cost: what you actually see
Every recipe shows:
Plate cost: what every ingredient in the recipe costs, added up, at today's prices. This is the whole recipe as you entered it, not a per-serving number.
Cost per serving (iPhone app): plate cost ÷ the number of servings. It shows up once you tell Stockcount the recipe makes more than one serving. This is your true cost for a single plate.
Margin %: the share of the menu price left after ingredient cost — basically, how much of the price isn't food. It shows up once the recipe is linked to a priced menu item, and it's color-coded so you can read it at a glance (see the table below).
Per-ingredient bars (iPhone app): a simple bar for each ingredient, so you can see at a glance which one is costing you the most.
The iPhone app and the web color-code margin on slightly different scales, so the same dish can show green on your phone and a neutral gray on the web. The web holds the higher bar:
Margin %
iPhone app
Web recipe list
70% and up
green
green
60–69%
green
gray
40–59%
amber
amber
Below 40%
red
red
Don't let the gap rattle you. A 65% dish is the same 65% dish either way — the web just waits for a higher margin before it calls one "good."
Costs refresh every time you open a recipe. New invoice → new ingredient cost → updated plate cost. No retyping, no consultant.
Yield: the setting people skip
Two separate settings decide whether your numbers come out per-plate or per-batch.
Recipe yield: how many servings the recipe makes. Starts at 1. Plate cost always shows the whole recipe. Once you set the yield to the real number of servings, the iPhone app adds a separate cost-per-serving figure underneath.
Ingredient yield %: how much of an ingredient you actually use after trimming and waste. Starts at 100%. A whole lobster might give you only 25% usable meat; heavily trimmed romaine maybe 80%. You set this on the ingredient itself, not on each recipe, so it applies everywhere that ingredient shows up.
Practical rule:
A recipe that makes one serving → leave yield at 1.
A recipe that makes many servings → set yield to the real count, so you get cost per serving.
Heavily trimmed ingredients → set their yield % when you create the ingredient.
Link recipes to menu items
A recipe only shows margin once it's tied to a menu item with a price. You link from the menu-item side, not from the recipe:
iPhone app: open the menu, tap Link recipe on the item, pick the recipe.
Web: on the Menus → Items tab, use Smart linker to accept the recipes Stockcount matched up by name. Or from Recipe Coverage, hit Create recipe on a menu item and link it as you build.
Once linked, the recipe shows the menu price and the margin %. On the web you also see food cost % — ingredient cost as a share of the price, which is just the flip side of margin.
Not sure which recipes to build first? Open Recipe Coverage (under Analytics on stockcount.io). It lists the menu items that don't have a recipe yet, so the gaps are easy to see — build your biggest sellers first.
Don't do this
What good looks like
A 60-seat restaurant with 24 menu items. They've built recipes for the 12 best-selling dishes, plus 6 prep recipes those dishes rely on. The recipe list shows each dish's margin at a glance — they run from 62% to 78%. Two dishes show amber (chef's-choice cuts that come and go). Recipe Coverage says 81% of revenue now has a recipe behind it.
That's enough to see where the money is. The other 12 menu items can wait for the next quarterly review. They don't move the needle.
Monthly check-in (60 minutes)
Open Recipe Coverage. Build recipes for the top five best-sellers that still don't have one.
Open any recipe showing amber or red on margin. Ask Stockcount in chat: "Why is the burger plate cost so high?" It'll go through the ingredients with you.
Check your par suggestions (open Par suggestions from the Items screen or the Stock List tab on the iPhone app). Par is the amount of each item you want to keep on hand. Stockcount applies the suggestions it's confident about on its own each week; the rest wait in a queue for you to approve or skip.
Spot-check a prep recipe yield against reality. Are you actually getting 5 quarts of marinara per batch, or closer to 4.2?
That's it. Recipes are build-once, refine-quarterly. Not a daily chore.
What's not here yet
No what-if simulator. You can't model "olive oil +10%, how does margin shift?" yet.
No menu-engineering grid — the classic chart that sorts every dish by how much it sells and how profitable it is. Recipe Coverage is the closest view today.
No per-recipe yield % override. An ingredient's yield % is the same in every recipe that uses it.
No live cost preview while editing on the web. That's on the iPhone app only for now.
Plate-costing the whole menu sounds like a consultant's project. It's an afternoon — if you build it in the right order.
Our take
Plate cost is what one finished dish costs you in ingredients. The slow way to work it out across a whole menu is dish by dish, top to bottom, each one from scratch. The fast way starts with your prep recipes — the marinara, the vinaigrette, the brine, the spice blends — because every dish on the menu reuses them. Build those few first and the dishes that depend on them go quickly, each one pointing at work you've already done. And you don't need the whole menu. You need the dishes that move money. An afternoon on prep recipes plus your top sellers covers most of your revenue. The long tail can wait.
How to plate-cost your menu in an afternoon
List your prep recipes and build them first. Anything you make in a batch and store — sauces, dressings, braises, doughs. These are your building blocks. Build each one as a prep recipe (open the ingredient on the web, or ask Stockcount in chat) and set its real batch yield, not 1.
Build dish recipes for your top sellers. Start with the ten or fifteen dishes that carry the menu. Add ingredients from your catalog — and add the prep recipes you just built, which now show up as items in their own right, instead of re-entering raw oil and tomatoes.
Link each recipe to its menu item. A recipe only shows its margin — the share of the menu price left after ingredient cost — once it's tied to a priced menu item. On the web, the Menus → Items tab has a Smart linker that matches recipes to items by name. On the iPhone, tap Link recipe on the menu item.
Open Recipe Coverage to see the gaps. Under Analytics on the web, Recipe Coverage lists the menu items that still have no recipe. Work down it, biggest sellers first.
Stop at about 80% of revenue covered. Once your costed dishes account for roughly four-fifths of sales, you have what you need to make decisions. The bottom 20% of the menu can be costed another day.
Why this works
Prep recipes are the trick. A marinara built once and saved is a marinara that costs five pasta dishes at the same time. Without that, you'd enter the same tomatoes, oil, and garlic into five separate dishes — five times the work, and five places to fix when the tomato price moves. Build the shared layer first and the menu stops being 40 recipes. It becomes a handful of prep recipes plus the dishes that lean on them.
And it stays costed on its own. Every dish is a live calculation, not a snapshot. When the next invoice changes an ingredient price, every recipe that uses it re-costs — the prep recipe, and every dish built on that prep recipe. You do the building once. The pricing keeps itself current after that.
The math
Don't do this
What good looks like
A 60-seat restaurant sets aside a Tuesday afternoon. The chef lists 9 prep recipes and builds them first, each with its real batch yield. Then 15 dish recipes for the best sellers, most of them pointing at those 9 preps. Each recipe gets linked to its menu item with the Smart linker. By 5pm, Recipe Coverage shows 84% of revenue has a recipe behind it, and every one of those dishes shows a live margin. Nobody opened a spreadsheet. Nobody called a consultant. The rest of the menu is a half-hour job for a slow week.
Where this connects
Cost a recipe: how a single recipe works in depth — dish versus prep, yield, margin.
Every menu has dishes that pay the rent and dishes that just take up space. The hard part isn't fixing them. It's being honest about which is which.
Our take
You don't need a menu-engineering matrix with four quadrants and a color key. You need two facts about each dish and four moves. The two facts: does it sell, and does it make money. The first you already know — you watch the tickets every night. The second is where most operators are guessing, and it's exactly what Stockcount gives you: a real margin per dish — the share of each sale left after the dish's ingredients — kept current as costs move. Put the two facts together and every dish on your menu falls into one of four moves. The mistake is treating all dishes the same. A handful carry the whole menu.
How to decide what to push, reprice, or 86
First, make the margins real: every dish you're judging needs a costed recipe behind it. Then open the recipes list and sort by margin. Now, dish by dish:
Sells well, strong margin → push it. This is your money-maker. Put it where eyes land on the menu, have servers recommend it, build specials around it. You want more of these orders, not fewer.
Sells well, thin margin → reprice or redesign. A popular dish losing you margin is the most expensive problem on the menu. Nudge the price up a dollar, or redesign it: re-portion, or swap a costly ingredient for one that performs as well. Small fixes on high-volume dishes add up fast.
Sells slowly, strong margin → reposition it. Good money, not enough takers. Move it up the menu, rename it, let servers talk it up. If it stays slow, that's fine — a profitable dish that sells a little is not a problem.
Sells slowly, thin margin → 86 it. To "86" a dish is to take it off the menu. A dish that neither sells nor earns is costing you prep time, fridge space, and a line on the menu a better dish could use. Cut it.
Why this works
Menu engineering fails when operators chase the percentage. A dish at 78% margin feels better than one at 66%, so it gets pushed. But percentage isn't money. What lands in the bank is contribution margin — the actual dollars a dish keeps after its ingredients. A lower-percentage dish that sells in volume can out-earn a high-percentage dish nobody orders, every week, by a wide gap.
That's also why you don't spread your attention evenly. Most menus earn the bulk of their money from ten or twelve dishes. Those are where a reprice or a redesign moves real dollars. The slow, cheap corner of the menu can be wrong for a year and barely cost you. Fix the dishes that carry the load. Leave the rest.
The math
Don't do this
What good looks like
A bistro owner sits down on a Monday with the recipes list, sorted by margin. Twenty-two dishes. The two best sellers also run strong margins — those get a server push and better menu placement. The brunch hash sells hard but runs thin; the price goes up $1.50 and the potato spec gets re-portioned. A high-margin lamb dish that sells eight a week gets moved to the top of the entrées with a fresh description. Two appetizers that neither sell nor earn get 86'd. Five dishes, four moves, thirty minutes. The other seventeen are left exactly as they are, because they aren't where the money is.
Where this connects
Cost a recipe: how to get the margin number you're judging dishes by.
The owner doesn't need to count. The counter doesn't need to see your costs. Get roles right and the right person sees the right number — and the wrong person doesn't.
Our take
Most operators give everyone owner access on day one because the role picker is unfamiliar. Two weeks later, a counter is staring at vendor prices they shouldn't see, and the dishwasher knows what the general manager makes. Roles are how Stockcount keeps things fast for the counter and private for the owner. Set them once, set them right.
The three roles
Role
Can do
Can't do
Owner
Everything. Billing, settings, team, all data.
Can't be invited or transferred away.
Manager
Counts, items, recipes, invites, role changes for counters and other managers.
No billing access. Can't promote anyone to owner.
Counter
Count inventory and view stock levels.
Can't invite, change roles, or edit items.
Roles aren't just about hiding buttons. Even if a counter went hunting for a manager-only screen, Stockcount would still stop them — the limits are real, not just visual.
How to invite
Owners and managers can invite:
Open the Team page. (On the phone, tap Team in the side menu. On the web, Team in the sidebar.)
Tap Invite.
Pick the role (Manager or Counter — owners can't be invited).
Optionally add an email. If you do, Stockcount emails the invite code to that address.
Stockcount generates an 8-character invite code.
Send the code to your teammate.
Your teammate enters that code to join — on the web, or in the iPhone app under Join a team. If you sent it by email, the link in the email does this in one tap. Codes expire after 7 days.
What each role actually sees
The counter's app is deliberately small. They open the iPhone app, tap the Chat tab, and count. Stock List shows current quantities; Today shows what to count next. No recipes, no cost history, no dashboard — counting is fast because there's so little on screen.
Managers and owners get the full picture. Both see the catalog, recipes, vendors, the Team page, and the web analytics dashboard, and both can scan invoices, build recipes, and review variance. Owners additionally get billing and the credit balance. Most owners live in chat and the web dashboard; the rest is there when they want it.
Don't do this
Notifications and alerts
Notifications are set per person, not per role — you can't set them for someone else. Each team member opens Settings → Notifications (web or phone) and picks their own:
Low stock alerts: tells you when an item drops below its par — the level you want to keep on hand.
Overdue count alerts: tells you when an item is past due for a count.
Daily digest: one notification each morning, covering what to count and what changed yesterday.
Weekly food cost report: a Monday-morning recap of where last week's food cost landed. Off by default; the other three are on.
These all arrive as notifications on the iPhone app. There's no web or desktop notification yet.
What good looks like
A 3-location operator with 14 team members. The owner is the only owner at each location. Two managers per location (the general manager and the sous chef). Eight counters across the three locations — line cooks who count their own stations. The counter role hides vendor costs but shows stock levels, so they catch out-of-stock items right where they happen. Managers handle invoices and recipes. The owner reads the weekly COGS notification on Monday and decides what to act on.
What's not here yet
No ownership transfer. If you're selling a location, contact us. The owner role can't be reassigned in the app yet.
No close-this-location flow. Same deal.
No combined dashboard across locations. Each location's roles and analytics stand on their own.
No live two-person counting. If two people count the same item, the most recent count wins and replaces the earlier one (both are still kept in the count history). Split the walk-in into zones so two people never count the same shelf at the same time.
Your best counter just gave two weeks' notice. The count habit shouldn't walk out the door with them — and teaching the next one shouldn't cost you a day.
Our take
A new line cook should be counting your walk-in cooler, by voice, correctly, inside fifteen minutes. Not fluent and not fast — that comes within a week — but doing it right the first day. This works because the Counter role is deliberately tiny: three tabs, no costs, no recipes, no settings, nothing to break. So don't train them on "the app." There's barely an app to train. Teach the two things that actually matter — the route through your walk-in and how to talk to the phone — and let everything else stay invisible. Over-training a counter is the most common mistake, and it's worse than under-training, because it teaches them the job is complicated. It isn't.
How to train a new line cook
Block fifteen minutes at the start of a shift. Have your phone and theirs. Here's the session.
Before they arrive, invite them as a Counter. Open the Team page, tap Invite, pick the Counter role, and send the 8-character code by text. The Counter role lets them count and see stock levels — and nothing else. They cannot open your costs or your margins. That's the point.
Minutes 0–2: get them signed in. They install the app and enter the code under "Join a team." Their app opens to three tabs: Today, Chat, Stock List. Let them see how small it is. Most of the worry about "learning the system" dies right here.
Minutes 2–7: walk your count route with them — in the walk-in, not at a screen. Start at the door, top shelf, left. Move shelf by shelf, the way you always do. Don't mention the app once. You're teaching a path through a cold room, and that path is the actual job.
Minutes 7–12: hand them the phone and have them count one section. They tap Speak and say the item and the count — "romaine, two cases." Stockcount turns "two cases" into the unit you track. The first time it hears a pack size — how many units are in a case — it asks; show them how to answer it once, and it remembers for good.
Minutes 12–15: show them the Today tab and the sanity check. The Today tab tells them what's due to be counted. And if Stockcount ever pauses to question a number — a count ten times higher than expected, or a well-stocked item suddenly at zero — tell them plainly: that's a question, not an accusation. Look at the shelf, answer it, keep going.
Why this works
Three things make a fifteen-minute handoff realistic.
The Counter role removes the curriculum. A counter's app has no recipes, no costs, no analytics, no settings. There is almost nothing to explain because there is almost nothing on screen. A new cook can't get lost in a room with three doors.
Voice isn't a skill you train. Tapping numbers into a grid is a small skill, and small skills need practice. Talking is not. "Romaine, two cases" is a sentence your new cook already knows how to say. The app meets them where they already are, on day one.
The route is the real lesson, so you teach it in the room. Inventory accuracy comes from counting the same things in the same order every time. That consistency is a physical habit — a path — and you can't teach a path from a chair. Five minutes walking the walk-in with a new cook beats an hour of screens. It also means the route survives the turnover: the next cook learns the same path, not a new one.
The math
Don't do this
What good looks like
A new grill cook, second day on. Monday morning the GM opens the Team page and invites them as a Counter; the code goes out by text before the cook has tied an apron. At the start of shift the cook installs the app and joins in under a minute. Three tabs. Nothing to be nervous about.
The GM walks the walk-in route with them once — door, top-left, shelf by shelf, five minutes, no phone. Then the cook takes the phone and counts the produce shelf by voice while the GM stands there. Stockcount doesn't recognize a case of a new baby lettuce and asks how many heads are in it; the GM says "twelve," and it's remembered for good. By minute fifteen the cook has counted a section, start to finish, on their own.
The following Sunday, that cook runs the entire walk-in solo in twelve minutes. The count habit didn't leave with the last hire. It belongs to the restaurant now, not to any one cook.
You bought Stockcount. You believe in it. And you're still the one counting the walk-in at 11pm on Sunday, because the GM "hasn't gotten to it." A tool the owner runs alone isn't adopted. It's a hobby.
Our take
A tool the owner loves and the GM tolerates will be dead in a quarter. Adoption isn't something the software does — it's a handoff you have to actually make. And the hardest part of the handoff isn't teaching your GM the app. It's getting yourself to stop doing the count. As long as you keep covering it "just this week," your GM correctly reads that it's still your job. Handing over the loop — the weekly count-and-close routine this whole site is built around — takes three honest conversations and, if those don't land, one clear line in the sand. It also takes you sitting on your hands while the first few counts come back rough. That last part is the price, and most owners won't pay it. Pay it.
How to hand the loop to your GM
This runs over about a month. Don't compress it — and don't let it drift past it, either.
Conversation one: give them the number, not the tool. Don't ask your GM to "use Stockcount." Ask them to own food cost percentage — what your food and drink cost as a share of sales. Tell them the target, say 30%, and that it's now theirs to hit and theirs to explain. The tool stops being your pet project the moment the GM has a number they answer for. Stockcount becomes the instrument they reach for, not the thing you're making them do.
Conversation two: kill the surveillance fear before it grows. Your GM will assume the analytics exist to watch them. Say out loud that they don't. The variance breakdown — the screen that shows where food cost moved and why — is the GM's evidence, not your weapon. A GM with it can answer "why was last week high." A GM without it can only shrug. Frame it as the thing that protects them in the conversation you're both going to have anyway.
Conversation three: actually hand it over. Give the GM the Manager role — full operating access, no billing. Put the Sunday close-out on their schedule as a recurring block. Switch on their weekly food cost report so Monday morning the number finds them on its own. Then do the thing that makes or breaks this: stop doing the count yourself. Next Sunday, don't go in.
If it hasn't taken in three or four weeks, draw the line. Some handoffs need a fourth conversation, and it isn't really a conversation. The Sunday close-out is part of running this restaurant — the same way the nightly deposit is part of it. Not a favor, not "when you get to it." Said once, plainly, without heat. Optional things do not get done. So this stops being optional.
Why this works
People don't run software. They run numbers they're accountable for. As long as "Stockcount" is the thing the owner wants used, it's a chore with your name on it. The moment food cost is the GM's number — their target, their explanation, their win when it drops — the tool becomes the obvious way to chase it. You're not selling adoption. You're assigning ownership, and the adoption follows it.
The surveillance conversation matters more than it sounds. Every back-of-house manager has been handed a number as an accusation at some point. If the first time your GM really sees the variance breakdown is the day you point at it angrily, the tool is now the enemy, and they will quietly route around it forever. Get there first. Make the numbers theirs before they're ever used against them.
And the reason you have to stop doing the count: a GM cannot own something the owner still does. Every week you "just cover it" is a week of proof that the real owner of the loop is still you. The rough counts in weeks one and two aren't the handoff failing. They are the handoff. Let them happen.
The math
Don't do this
What good looks like
An owner-operator who's done the Sunday count for two years decides to hand it off. Over four weeks: one conversation handing the GM the food cost target, one making clear the numbers are the GM's armor and not the owner's whip, one handing over the Manager role and a recurring Sunday block. The weekly food cost report gets switched on for the GM's phone.
Then the owner does the hard thing and stays home on Sunday. Week one's count is rough — a couple of items missed, the close-out runs long. The owner says nothing. Week three the count is clean and the close-out takes the promised twenty minutes. By week six the GM walks up on Monday with last week's food cost number before being asked — and when a week goes sideways, the owner hears it from the GM on Monday, not from a surprise at month-end.
That's the whole goal. Not that a GM uses an app. That an owner hears things from their GM instead of from their accountant.
Adding a second location is the test of whether your first location runs on a system or runs on you. The system has to scale. You have to stop being the bottleneck.
Our take
Multi-location in Stockcount today is honest, not magic. Each location is its own world — its own catalog, its own vendors, its own counts. Billing is shared across your whole account. There's no combined dashboard yet. The way it works today is good enough for most operators running 2–4 locations. If you're at 5 or more, talk to us before you scale further.
How to add a location
On stockcount.io, open Account → Locations and add a location. This is web-only — the iPhone app can't create locations.
Pick a name. The actual restaurant name works well; you'll see it in the location switcher.
Start fresh. Catalog, vendors, and recipes all start empty on the new location.
A location switcher sits in the side menu on the phone and in the sidebar on the web. Tap it to switch between locations.
What's per-location vs shared
Per-location (separate per restaurant)
Shared across your account
Catalog (items)
Subscription and billing
Vendors
Credit pool
Counts and count history
Your login
Recipes
Team members (people can have roles at multiple locations)
Receipts and invoices
Analytics dashboard
Par levels and count schedules
Practical workflow
What this means day to day:
Counting: Each location's team counts that location. The counter at Main St. never sees Park Ave.'s walk-in. Voice mode works on one location at a time — switch locations and everything the agent can do switches with it.
Receipts: Scanned at the location they arrive at. The Sysco invoice for Main St. updates Main St.'s costs only.
Recipes: Built per location. If you run the same menu at two locations, you build the recipes twice today. (We know. It's on the list.)
Vendors: "Sysco" at Main St. is a separate vendor from "Sysco" at Park Ave. Their cost histories don't cross. If Sysco raises prices at one location, that update only touches that location.
Reporting: Each location has its own analytics page; there's no combined view yet. Operators pull the locations together by hand (see below).
The role split across locations
The same person can have different roles at different locations. The owner of Main St. can be a counter at Park Ave. A role belongs to a person at a location, not to the person overall.
In practice: every invite is for one specific location. You can invite the same person to other locations later, with a different role each time.
Don't do this
What good looks like
A two-location independent operator. Each location runs its weekly routine on its own, and both close out by Sunday night. The combined view lives in a Google Sheet the owner updates Monday morning — two food-cost percentages, a labor number, and a few lines on what moved and why. Total time: 10 minutes. Stockcount does the per-location heavy lifting; the owner does the 30-second summary.
That's the honest picture today. It works. It's not yet what it should be.
What's not here yet
No combined dashboard across locations. Pulling food cost, variance, and sales together is all manual today.
No shared catalog across locations. The same item gets built twice.
No location ownership transfer and no close-this-location flow. The owner role is fixed at the location level; contact us if you sell or close one.
No location cloning. You can't say "copy Main St.'s catalog to Park Ave." today.
If multi-location is the reason you're picking Stockcount, talk to us first. Several of the gaps above have prototypes; we prioritize based on real operator demand.
Your data is yours. Not "yours, with conditions." You can take a full copy out of Stockcount whenever you want — whether you're leaving or just want a backup in a drawer.
Our take
Plenty of software makes leaving quietly difficult — the export button is buried, or it's a support ticket, or it's a partial dump missing the half you need. Stockcount's position is the opposite: a complete copy of your data, one download, any time, no one to ask. One thing to know up front, because it's easy to miss: the full one-file export lives in the iPhone app today, not on the web. The web gives you per-page exports instead. Both are covered below.
How to export everything
The full bundle comes from the iPhone app:
Open the iPhone app and go to Account → Security.
Under "Your data," tap "Export my data." This is available to owners and managers.
Stockcount builds one compressed file — a .json.gz bundle holding your items, count history, recipes, expenses, and cost history, plus your chat conversations. Vendor names ride along inside the expense and cost records.
Save or send it through the iPhone's share sheet — to Files, to email, to wherever you keep records.
Exporting from the web
There's no single full-account export on the web yet. What the web gives you instead is export page by page, as CSV files — plain spreadsheet files that open in Excel or Google Sheets:
Catalog page — an "Export CSV" button for your full item list.
Analytics pages — Variance, COGS, Plate costs, and Revenue each export to CSV.
If you want everything in one file, use the iPhone bundle. If you just need one slice — say, your item list for a spreadsheet — the web CSV is quicker.
Why this works
Data you can't get out isn't really yours. An export you can run any time, without asking, is what makes the rest of Stockcount trustworthy: you're never locked in, so you stay because it works, not because leaving is hard. It's also plain good practice. Pulling an export before any big change — a catalog cleanup, an ownership handover, a switch of tools — costs two minutes and saves you the one bad afternoon where you wish you had.
Don't do this
What good looks like
An owner about to hand the books to a new bookkeeper opens the iPhone app, goes to Account → Security, and taps "Export my data." A few seconds later there's one .json.gz file in their Files app — items, counts, recipes, expenses, cost history, conversations, all of it. They email it to the bookkeeper and keep a copy. Total time: under a minute. No ticket, no waiting, no one to convince.
Where this connects
Cancel your subscription: what happens to your data if you leave, and why you can always come back to it.
Most software hides the cancel button three menus deep. Here's exactly where ours is, exactly what happens when you press it, and exactly how to come back.
Our take
Canceling should be as plain as subscribing. So here's the whole truth in one paragraph: you cancel from the web, you keep access until the end of the period you've already paid for, and your data is never touched. Canceling is not deleting — your items, counts, recipes, and history all stay exactly where they are, and if you come back, they're still there. We'd rather you leave clean and return later than feel boxed in. A tool people stay with because leaving is painful is not a tool worth running.
How to cancel
Open stockcount.io and sign in as the account owner. Only the owner can cancel — billing lives with the owner role. Go to the Billing page.
On the Plan card, select Cancel. A confirmation dialog explains what happens; confirm it there.
Keep working until the period ends. Cancellation is scheduled, not instant. You keep full access through the end of the period you've paid for — the Plan card shows the exact date.
Changed your mind? Resume. Any time before that date, the Plan card has a Resume option that calls the whole thing off. Nothing was lost.
There's no cancel button in the iPhone app — billing is handled on the web. The app will point you to stockcount.io if you look for it there.
What happens to your data, credits, and access
Your data stays. All of it. Canceling never deletes anything — items, counts, recipes, vendors, cost history. It sits there, intact, for whenever you want it.
Coming back is just resubscribing. Pick a plan again and everything is where you left it: same locations, same catalog, same history.
Purchased top-up credits stay and never expire. Any credits you bought outright remain on the account. Your monthly included credits simply stop renewing. There are no refunds for the current period — you've paid through it, so use it.
The app still opens after the period ends. You can still count and view your stock. Only the AI features pause, and only once any remaining credits run out.
Why this works
A clean exit is a feature, not a loss. When canceling is honest — no retention maze, no data held hostage — the decision to stay is a real one, made on whether the tool earns its keep. That's the only kind of customer worth having. It also means a slow season isn't a trap. There's no separate "pause" button, and you don't need one: cancel when you close for the winter, resubscribe when you reopen. Your data waits for you the whole time, at no cost and no risk.
Don't do this
What good looks like
A seasonal cafe closes for January and February every year. In late December the owner signs in at stockcount.io, opens Billing, and cancels — access runs out the first week of January, right on schedule. The café's whole history sits untouched for two months: catalog, par levels, recipes, last winter's counts. In March the owner resubscribes in two minutes, and the first count of the season picks up exactly where the last one left off. No data re-entered, no setup redone, nothing lost to the off-season.
Where this connects
Export everything: pull a full copy of your data before you go — good practice, even though canceling keeps it.