Skip to main content

Metric Troubleshooting

Why metrics seem wrong and how to work through them.

The golden rule​

If a metric seems wrong, check the data before assuming a bug.

Every metric in Synplex is a formula applied to input data. If the inputs are correct, the metric is correct — even if the result surprises you. The most common cause of a metric that appears wrong is incorrect or stale input data, not a calculation error.

The standard verification sequence for any metric:

  1. Confirm data is fresh — trigger a manual sync via Settings → Sync if the Last Synced timestamp is more than 24 hours old
  2. Confirm on-hand inventory matches Shopify
  3. Confirm the settings the metric depends on (lead time, safety stock, service level) are accurate
  4. Read the metric definition in the Data Metrics Reference to understand exactly what is being calculated
  5. If all inputs are correct and the metric still seems wrong, contact support

Problem 1: Reorder quantity seems too high​

Symptom: The To Buy quantity is much larger than you expected.

How the calculation works: To Buy = (Daily Sales × Lead Time) + Safety Stock − On-Hand Inventory

For example: 10 units/day × 14-day lead time = 140 units to cover the lead time window, plus 50 units safety stock, minus 20 units currently on hand = 170 units to order. If your lead time setting is wrong — say it is set to 30 days instead of 14 — the result will be nearly double what it should be.

What to check:

  • Lead time — this is the single biggest driver of reorder quantity. Confirm the lead time set in Synplex matches your actual supplier delivery time. Even a few days' difference compounds significantly at high sales rates.
  • Safety stock — check whether your safety stock setting is appropriate for your demand variability. A high service level target produces a high safety stock figure.
  • On-hand inventory — confirm Synplex matches Shopify. If on-hand is understated, To Buy will be overstated.
  • Daily sales rate — check whether the sales rate Synplex is using reflects your current reality. A spike in recent sales will increase the rate.

If all inputs are correct and To Buy is still higher than expected, the metric is telling you something real — your lead time and safety buffer together require more stock than your intuition suggested.

Problem 2: Days to Order doesn't match my intuition​

Symptom: Days to Order shows 5 days but you have 20 days of stock on hand.

How the calculation works: Days to Order = (On-Hand − Safety Stock) ÷ Daily Sales − Lead Time

Your intuition is probably calculating: on-hand ÷ daily sales = days of supply. That tells you when you will run out, not when you need to order. The metric is calculating something different — it is working backwards from when stock needs to arrive (before safety stock is breached) and subtracting lead time to tell you the latest safe date to place the order.

Using the example above: (100 units on hand − 50 units safety stock) ÷ 5 units per day − 5 days lead time = 5 days to order. If you wait 20 days you will run out at day 10, well before new stock could arrive.

What to check:

  • Safety stock — your usable inventory is on-hand minus safety stock, not total on-hand. If safety stock is 50 units and you have 100, you only have 50 to work with.
  • Lead time — every extra day of lead time brings the order date forward by one day.
  • Daily sales rate — a higher rate depletes usable stock faster.

If all inputs are correct, trust the metric. It is accounting for factors that mental arithmetic typically misses.

Problem 3: Grade changed unexpectedly​

Symptom: A product's grade dropped from A to B or C without any changes you are aware of.

How grades work:

Grades recalculate automatically every day based on a combination of sales volume, profit margin, inventory position, and stockout risk. A grade change means one or more of those underlying signals shifted — it is not a bug and it is not random.

What to check:

  • Sales — did sales drop or slow in the past 7–14 days? Even a modest decline in a rolling average can shift a grade.
  • Inventory position — is the product overstocked or understocked relative to its current sales rate? Both extremes push the grade down.
  • Margin — did the cost or price change recently? A tighter margin affects the grade.
  • Stockout risk — is Days to Stockout getting close? Rising stockout risk lowers the grade regardless of sales performance.

The grade reflects your actual business situation at that moment. If the grade dropped, something in the business changed — the metric is doing its job by surfacing it.

To improve a grade: resolve the underlying issue the grade is reflecting — reorder if understocked, run a clearance if overstocked, review margins if profit has thinned.

Problem 4: Sales at Risk seems very large​

Symptom: Sales at Risk shows a figure — say $50,000 — that seems implausibly high.

How the calculation works: Sales at Risk = Daily Sales Revenue × Days to Stockout

This is not a projection of what you might lose eventually — it is the revenue at risk if you do nothing and stock runs out on schedule. At $10,000 per day in revenue on this product and 5 days until stockout, $50,000 is exactly right.

What to check:

  • Daily sales revenue — confirm the daily revenue figure Synplex is using matches your actual average. If revenue is genuinely $10,000/day on this product, $50,000 at risk over 5 days is real.
  • Days to Stockout — confirm on-hand inventory is correct. If on-hand is understated, Days to Stockout will be shorter than reality and Sales at Risk will be overstated.

If both inputs are correct, the figure is real and the metric is doing exactly what it should — creating urgency. Place the order.

Problem 5: Safety stock seems arbitrary​

Symptom: Safety stock shows 50 units and you cannot see why.

How the calculation works: Safety Stock = Service Level Factor × Standard Deviation of Daily Demand

It is not a round number chosen at random. It is calculated from your historical demand variability and your configured service level target. A 95% service level uses a factor of 1.65. If daily sales vary by 30 units on average, safety stock = 1.65 × 30 = ~50 units.

What to check:

  • Service level setting — a higher service level produces more safety stock. If 50 units seems too high, consider whether your service level target is appropriate for this product.
  • Demand variability — if demand is genuinely volatile, a higher safety stock is correct. If the product has very stable demand and safety stock still seems high, check whether there is a historical data anomaly inflating the variability calculation.

Adjust the service level in Settings if you want to lower safety stock across the board. Do not lower it to hide a stockout risk — the metric exists to prevent lost sales.

Problem 6: Multiple metrics all look wrong at once​

Symptom: Reorder quantity, Days to Order, Grade, and other metrics all seem off simultaneously.

Why this happens: When multiple metrics are wrong at the same time, the cause is almost always stale or incorrect underlying data rather than a calculation error. All metrics share the same inputs — if on-hand inventory is wrong or the sales rate is stale, every metric built on those inputs will be off.

Fix:

  1. Go to Settings → Sync and trigger a manual sync — wait for it to complete
  2. After the sync, confirm on-hand inventory matches Shopify for a sample of products
  3. Check whether daily sales rates look reasonable against your recent actuals
  4. Confirm lead times are still accurate — supplier lead times change and are easy to forget to update
  5. If data is now correct but metrics still look wrong, go to Data Metrics Reference and trace through the formula for the specific metric with your actual numbers

Verification checklist​

Before concluding a metric is broken, confirm each of these:

  • Last Synced timestamp is less than 24 hours old — if not, trigger a manual sync
  • On-hand inventory in Synplex matches Shopify
  • Lead time setting reflects actual current supplier delivery time
  • Safety stock setting and service level are intentional
  • Daily sales rate reflects current trading patterns
  • You have read the metric definition and understand what it is calculating

If every item on this list checks out, the metric is correct.

When to contact support​

Contact support if:

  • All inputs are verified correct and a metric is still clearly producing an impossible result
  • A metric changed dramatically overnight with no corresponding change in underlying data
  • You traced through the formula manually and the numbers do not add up

See When & How to Contact Support. Include the metric name, its current value, the inputs you have verified, and what you believe the correct value should be.