The greatest tools for identifying food cost problems include:
Weekly inventories instead of monthly
Weekly ideal cost calculations
Properly segmented inventory
Weekly inventories instead of monthly inventories help you isolate potential cost issues to a particular week. If you are only counting monthly, you may be allowing a problem to persist for weeks before identifying it, possibly costing thousands of dollars in increased costs. Weekly inventories sound daunting, but the extra labor dollars needed are more than made up by lowered food costs. In restaurants, switching from monthly to weekly inventories shaves 4-6% points off the food cost, on average, in my experience. There are many "shortcuts" that can drastically cut the time necessary for counting inventory, such as having cooks prep every station to an exact count every Sunday before the inventory is taken, or taking a standard costs for your spices or FOH condiment station.
Are you calculating Ideal Costs? If you aren't, you can only guess at what your problem is. If you do calculate ideal costs, you will be able to pinpoint or rule out theft and waste as the source of your problem.
Calculating ideal costs is the practice of multiplying your sales by item counts by the recipes costs of each of the items you sell. This only requires a simple spreadsheet that you input up to date recipe costs into, then have FOH staff or managers input the weekly sales by item counts. Then, you multiply the number of each item sold by its recipe cost to calculate how much the food that you sold should have cost to sell. If you don't know what your food should have cost to sell, you can't know if there truly is a food cost problem or if you simply have a change in your sales mix. Some times, higher food costs can mean more profit if it is caused by selling an increased number of high cost, high profit items. When you are focused only on your actual food cost without comparing it to your ideal food cost, you can get too hung up on a percentage and make decisions that can adversely affect your profit in the quest for a lower percentage. If you are comparing the two and there is no discrepancy, but the actual costs are high, you know it was caused by a change in the sales mix. If there is a big discrepancy, and the actual costs are lower, you likely have a calculation error. If there is a big discrepancy, and the actual costs are higher, you may have waste, theft, calculation errors, or unrecorded sales. You'll still have some work to do running it down, but at least you won't be chasing your tail if its all just due to a change in your sales mix.
Missing sales are easy to identify by looking at ticket averages. If customer counts are steady, but ticket averages down, you likely have some FOH employee theft, such as employees voiding tickets or items from tickets, collecting the same dollars then pocketing the difference. Most point of sale systems have a void report. If servers have the ability to void without manager approval, there is likely some of that going on.
Calculation errors can be spotted by having the inventory proofread by someone other than the person who put it together. I suggest training hourly employees, shift managers and other staff to count inventory. The chef is responsible for making purchase cost adjustments (monthly) and proofreading the counts. You will be more likely to identify errors if you are looking at the inventory with fresh eyes, and not after you've just spent 4 hours counting and typing the count into a spreadsheet.
One of the most effective techniques for pinpointing cost problems is properly segmenting your inventory. Most chefs make the mistake of segmenting the inventory into storage area sections. This causes many items to be listed more than once and makes the sheet, much, much longer than it needs to be, which greatly increases the opportunity for errors inputting, pricing and calculating.
A food inventory should be segmented into food type sections. My spreadsheets use the food categories: meat, poultry, seafood, dairy, produce, bakery, grocery, kitchen alcohol and soft beverages. You could also segment your inventory into the same warehouse groups your food purveyor uses to segment your invoices. I prefer the first method because it makes the inventory easier to reference when counting products and writing recipes. The second method saves some time when you are calculating purchases by category though.
When you segment your inventory by food type, you eliminate all the extra instances of the same inventory item appearing on your count sheet which means less time updating item prices from your invoices, and lessening your opportunity to forget to update all the occurrences of the same item in separate areas. If you also track your purchases by the same food types, you can then have a cost of goods for each separate segment of your inventory. This makes it very easy to pinpoint the source of a cost issue by comparing week to week fluctuations in the cost of a particular food type.
To use this type of organization on your inventory spreadsheet, you have to name your inventory items in a way that when they are alphabetized, items that are stored together appear together on your spreadsheet. For example, two different beef items might be named, "beef, ground patties 4/1" and "beef, ground bulk". When the sheet is alphabetized, its easy to locate these items on the sheet. From there, you organize your shelves to match your sheet instead of making the classic mistake of organizing your sheet to match your shelves. Either way, its easier to count, but the first way, you gain a lot of other advantages.
Your inventory counting method can also greatly increase/reduce the occurrence of counting errors. Inventory should always be a two-man project. One person (the writer) should be focused only on recording numbers and checking the count unit to make sure the other person is counting the right size. The other person (the counter) is 100% focused on the shelf, painting the walls with his eyes and counting everything on the wall from top to bottom, left to right. They should not be counting based on what appears next on the sheet, but by what appears next on the shelf. This ensures nothing is missed because the counter's eyes never leave the wall. To keep inventory a short process, use more than one team of counters to speed things up. Each week, rotate the teams between storage areas. This eliminates the opportunity for those employees to cover up theft by manipulating counts.
There's a little more to it than what is above, but that covers the "gist" of it.