Inventory optimization strategies for retail are the habits that keep best sellers on the shelf while holding as little cash in stock as possible. A strategy works only if you measure it, so each one below comes with its number, its formula and the point at which to act. Most shops need five or six of them, not all fourteen.
What inventory optimization is, and how to score what each category's stock earns, is covered in inventory optimization. This page is the toolkit, ending with one shop's monthly scorecard in Ghana cedis.
| Strategy | The measure | Act when (a starting rule you can change) |
|---|---|---|
| Set goals | Targets hit at each review | A target has no number or date |
| Classify stock (ABC) | A items' share of gross profit | An A item runs out |
| Forecast from past sales | Forecast error | Error above 20% |
| Track key metrics | Turnover, stockout rate, holding cost | One gets worse two months running |
| Min-max levels | Emergency orders a month | More than one on A items |
| Supplier performance | On time in full (OTIF) | Below 90% |
| Central and local stock | Days of stock by branch | One branch holds several times another's days |
| Inventory linked to the POS | Negative stock lines | Any line below zero |
| Regular audits | Record accuracy | Below 95% of lines counted |
| Clear dead stock | Dead stock share | Above 10% of stock at cost |
| Just-in-time | Supplier OTIF, days of stock | Only with near-perfect suppliers |
| Safety stock | A-item stockouts a month | Any |
| Staff training | Count corrections as a share of stock | Above 0.5% a month |
| Regular review | The scorecard's trend | Every quarter |
Set smart inventory goals
Write three targets, each with a number and a date, before you change anything else. "Better stock control" is not a target; "dead stock under 8% of stock at cost by March" is.
Measure: how many targets you hit at each quarterly review. Raising inventory turnover from 9 to 10 times a year is a target too, if it comes with a date.
Classify stock by value and demand
Rank products by gross profit, then manage the top group tightly and the long tail loosely. ABC analysis splits them into A, the few that earn most, then B and C; the 80/20 rule in inventory works the ranking.
Measure: A items' share of gross profit = gross profit from A items ÷ all gross profit. Hold enough A stock that it never runs out, without overstocking the B and C lines.
Forecast demand using past sales
Base each order on the same weeks last year and the last eight weeks, adjusted for school terms and festive weeks. Then check how far off you were.
Measure: forecast error = (actual − forecast) ÷ actual, ignoring the sign. Forecast 400 bags of sugar, sell 340, and the error is 60 ÷ 340 = 17.6%. A spreadsheet of till sales is enough to start, and inventory forecasting methods shows better formulas.
Track key inventory metrics
Three numbers each month show whether your stock is working.
- Inventory turnover = cost of goods sold ÷ average stock at cost (Corporate Finance Institute). GHS 90,000 sold at cost on GHS 120,000 of stock is 0.75 a month, or 40 days of stock.
- Stockout rate: how often products run out; stockout rate examples give the formula.
- Holding cost = holding rate × average stock value, as worked in inventory holding cost.
Act when any of the three gets worse two months running.
Reorder with min-max levels
Give each product a minimum, the stock level at which you order, and a maximum you order up to. When stock reaches the minimum, order the maximum minus what you hold.
The minimum is the reorder point. Let a low-stock alert say when it is reached, and let a person place the order.
Measure: emergency orders a month, placed after a product had already hit zero. More than one a month on A items means the minimums are too low.
Monitor supplier performance
Score every delivery: did it arrive on the agreed day, and was it complete? Unreliable suppliers force you to carry more stock than you need.
Measure: on time in full (OTIF) = deliveries on time and complete ÷ all deliveries. A supplier with 9 of 12 deliveries right scores 75%. Below 90%, raise the buffer on that supplier's lines or find a second source.
Balance central and local stock levels
Move stock to the branch that sells it before you buy more. One branch's slow line is often another's shortage, as multi-location retail management explains.
Measure: days of stock by branch = stock ÷ average daily sales. For example, a branch in Adum holds 110 bags of rice and sells 2 a day: 55 days. The Suame branch holds 12 and sells 2 a day: 6 days.
Moving 40 bags leaves 35 days in Adum and 26 in Suame, with nothing bought.
Connect inventory with sales and POS systems
Sell everything through the system that holds your stock, at the counter and online, so each sale takes stock off. Sales written in a book leave the records holding stock you already sold.
Measure: negative stock lines, products whose recorded stock fell below zero. Each one means a delivery went unrecorded or a count was wrong, so act on every one.
Run regular inventory audits
Count a section every week and compare it with the records. Inventory audits by cycle count keep the shop open while you check.
Measure: record accuracy = counted lines that match the records ÷ lines counted. 46 matches out of 50 lines is 92%. Below 95%, find the cause before you correct: theft, damage or receiving errors.
Clear dead stock quickly
List everything with no sale in 90 days each month, and decide what to do with each line. Dead stock covers how to find it and the ways out.
Measure: dead stock share = stock with no sale in 90 days at cost ÷ all stock at cost. GHS 14,400 of GHS 120,000 is 12%. Above 10%, stop reordering those lines and clear them.
Apply just-in-time (JIT) only where deliveries are reliable
Just-in-time buying keeps little stock and orders often. It cuts holding cost, but leaves no buffer when a delivery is late.
Measure: the supplier's OTIF, and your days of stock on that line. Use JIT only with suppliers who deliver on time and complete almost every time, such as a distributor across town.
Use safety stock as a buffer
Keep extra stock on the lines you cannot be without, sized by how late deliveries run and how much sales swing. Slow lines need little or none.
Measure: A-item stockouts a month. Any stockout on an A item means its buffer is too small; stockout cost shows what each one loses.
Train staff and follow standard processes
Write down how your shop receives, counts, records damage and handles returns, and train every new person on it. Well-trained staff make fewer receiving and counting errors.
Measure: count corrections as a share of stock. GHS 1,800 removed by corrections on GHS 120,000 of stock is 1.5% a month. Falling corrections mean the process is working.
Review and adjust strategies regularly
Once a quarter, put the measures side by side and change the settings behind the worst one. Reorder points, buffers and suppliers all drift as seasons and prices change.
Measure: the scorecard's trend, quarter on quarter.
For example: a Kumasi shop's monthly scorecard
For example, a provisions shop in Kumasi with branches in Adum and Suame scored one month. These are illustrative figures, not a real shop's.
| Measure | Formula | This month | The shop's target |
|---|---|---|---|
| Days of stock | 30 × average stock ÷ cost of goods sold | 30 × 120,000 ÷ 90,000 = 40 days | 35 days |
| Dead stock share | Dead stock at cost ÷ all stock at cost | 14,400 ÷ 120,000 = 12% | 8% |
| Record accuracy | Matching lines ÷ lines counted | 46 ÷ 50 = 92% | 97% |
| Supplier OTIF | Right deliveries ÷ all deliveries | 9 ÷ 12 = 75% | 90% |
| Forecast error | (actual − forecast) ÷ actual | 60 ÷ 340 = 17.6% | 15% |
| A-item stockouts | Count in the month | 3 | 0 |
| Negative stock lines | Count at month end | 7 | 0 |
| Count corrections | Value removed ÷ stock at cost | 1,800 ÷ 120,000 = 1.5% | 0.5% |
Two numbers need action first. Dead stock at 12% holds GHS 14,400 that could buy the A items that ran out three times.
The supplier at 75% is the likelier cause of those stockouts than the forecast, so its lines need a bigger buffer or a second source. See what ArmPOS includes for shops in Ghana.
Where each measure comes from in ArmPOS
ArmPOS holds most of the scorecard in its reports, and most of them print or export to PDF and Excel.
- Days of stock: Inventory Valuation gives stock at cost, and Stock Inventory Valuation adds the units sold in a period (Stock reports).
- Dead stock share: the Dead Stock Report lists products with no movement in the last 30, 60 or 90 days.
- Negative stock lines: the Negative Stock Report lists products whose stock went below zero.
- Record accuracy: Products › Count Stock makes a count sheet by warehouse and category, and adjustments correct the differences (Count and adjust stock).
- Count corrections: the Stock Adjustment Report totals what adjustments added and removed, with each one's cost.
- Min-max and A-item stockouts: set a Stock Alert per product; Product Quantity Alerts lists what is at or below it (Reports to check every day). Alerts can also reach your phone (Get alerts on your phone).
- ABC ranking: Top Selling Products and Sales by Category rank what sells (Sales reports).
- Branch balance: a transfer moves stock between warehouses once it is approved (Move stock between warehouses).
- Supplier OTIF: ArmPOS has no delivery score. Note the promised and actual dates in each purchase's Note, or in a spreadsheet (Record a purchase from a supplier).
Frequently asked questions
What are the main inventory optimization strategies?
ABC ranking, forecasting from past sales, reorder points with safety stock, clearing dead stock, supplier scoring, balancing branches and regular counts. Pair each with the measure in the table at the top.
Which strategy should a small shop start with?
Clear dead stock first, because it frees cash this month. Then rank products with ABC, so your reorder points and buffers go where the profit is.
What is OTIF?
On time in full: the share of deliveries that arrived on the agreed day with everything ordered. Nine right out of twelve is 75%.
How do I measure inventory accuracy?
Count a set of lines and divide the number that match the records by the number counted. 46 matches out of 50 is 92%.
Is just-in-time good for small retailers?
Only with a supplier who almost never misses a delivery. Where deliveries run late or roads close in the rains, a reorder point with a buffer is safer.
Pick the three measures where your shop is weakest, set a target for each, and check them again in a month. To pull those figures from your till's own records, create a free ArmPOS account.
