> For the complete documentation index, see [llms.txt](https://docs.flowlity.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.flowlity.com/concepts/metrics-glossary.md). # Metrics glossary This page defines every metric you will encounter in Flowlity — in KPI columns, the Dashboard, and product detail views — and explains how to read them. ### Forecast accuracy metrics These metrics measure how close a forecast was to actual demand. They are available for three forecast sources (Flowlity forecast, Final forecast, External forecast) and several lookback periods (last 30 days, 90 days, 180 days, 12 months). #### MAPE (Mean Absolute Percentage Error) MAPE measures the average percentage difference between the forecast and actual demand: > **Formula:** ABS(demand − forecast) / demand, expressed as a percentage. A MAPE of 0 % means the forecast was perfect. The lower the value, the better the forecast. Flowlity calculates three variants of MAPE: | Variant | How it works | | -------------- | -------------------------------------------------------------------------------------------------------------------------------------------------- | | **Aggregated** | Compares the cumulated forecast against the cumulated actual demand across all products. Products with larger volumes naturally carry more weight. | | **Average** | Computes MAPE per product and then averages across products, giving every product equal weight regardless of volume. | | **Weighted** | Computes MAPE per product but weights each product by its actual demand, so high-demand products have more influence on the result. | You can see MAPE variants on the Dashboard's **Forecast accuracy** chart, and as KPI columns in the Demand and Planning product lists. {% hint style="info" %} There is no universal "good" MAPE — it depends on your industry and product mix. As a rough guide, under 20 % is generally considered good for consumer goods and under 10 % is excellent. Focus on comparing products against each other and tracking trends over time rather than chasing an absolute target. {% endhint %} #### MAE (Mean Absolute Error) MAE measures the average absolute difference between the forecast and actual demand in quantity units: > **Formula:** ABS(demand − forecast), in quantity (pieces, cases, kg, etc.). Unlike MAPE, MAE gives you an error in real units, which makes it easier to understand the business impact. For example, an MAE of 50 units tells you the forecast is off by 50 units on average. MAE is available as a KPI column with the same sources and periods as MAPE. {% hint style="info" %} Use MAPE to compare accuracy across products of different scales. Use MAE when you need to know the actual volume of the error — for instance, when sizing safety stock or estimating the cost of forecast errors. {% endhint %} #### FVA (Forecast Value Added) FVA tells you whether the forecast is better than a simple statistical baseline — a rolling average of the previous three months: > When FVA is **positive**, the forecast outperforms the baseline by that percentage. When FVA is **negative**, the baseline would have been more accurate. A negative FVA signals that the forecast may need attention — for example, the product might benefit from forecast events, manual adjustments, or a review of its demand history. ### Demand predictability Demand predictability is a per-product score that reflects how much usable demand history is available for forecasting. It appears as a column in the Demand product list and can be used to filter products in automations. | Level | What it means | | ---------- | ------------------------------------------------------------------------------------------------------------------------------------- | | **High** | Flowlity provides the best forecast — demand history is rich and stable. | | **Medium** | Flowlity provides a good forecast, but demand has shown erratic behavior. Consider reviewing forecast events or adjustments. | | **Low** | Past demand data is limited. The forecast may be less reliable — manual review is recommended. | | **No** | No past demand data is available. Flowlity cannot generate a forecast. Consider linking similar products to bootstrap demand history. | {% hint style="info" %} Use Demand predictability to prioritize your review effort. Let High-predictability products run on autopilot and spend your time on Medium and Low products where manual input can make the biggest difference. {% endhint %} ### Stock and inventory metrics #### Stock coverage Stock coverage tells you how many days your current stock will last based on the forecasted demand. Flowlity calculates it by accumulating the forecasted demand day by day and counting how many days pass before the stock is depleted. It appears on the Dashboard, in the Planning product list, and in supply order detail views. {% hint style="info" %} Compare stock coverage to your supplier lead time. If coverage is shorter than the lead time, you risk a stockout before the next delivery arrives. If coverage is much longer, you may be holding excess inventory. {% endhint %} #### Zero stock days Zero stock days is the average percentage of days without any stock, calculated on MTS (Make-to-Stock) products only. It is available for past periods (last 30 days, 3 months, 6 months, 12 months) and as a projection for the next year. A decreasing trend in zero stock days means your planning and replenishment are improving. This metric appears on the Dashboard. #### Inventory level Inventory level is the stock level measured at the end of each month, available in both value (currency) and quantity. It is displayed on the Dashboard. {% hint style="info" %} Track inventory level alongside demand trends. Rising inventory with flat or declining demand may signal an overstock situation that needs attention. {% endhint %} ### Dashboard summary metrics #### Demand plan The total final forecast for the selected scope (sites, tags) and period, available in value and quantity. It represents the demand your supply chain needs to fulfill. #### Supply plan The sum of firm and projected supply orders for the selected scope and period, available in value and quantity. It represents the supply side of your plan and can be compared against the demand plan to check balance. ### Service level Service level is the percentage of demand that is fulfilled over a defined horizon. For example, if demand is 100 units and 95 units are delivered, the service level is 95 %. In Flowlity, service level is a **target you set** — it is configured per product or as a site-wide default in [Buffer policy settings](/settings/buffer-policy.md). Flowlity uses this target to calculate how much safety stock to hold: a higher service level target means a larger buffer. {% hint style="info" %} Service level drives your safety stock. A jump from 95 % to 99 % can significantly increase the required buffer. Choose a service level that balances customer satisfaction with inventory cost — and consider setting different targets for different product categories using buffer policies. {% endhint %} ### Related pages * [KPIs](/concepts/kpis.md) — Add metric columns to the product list to sort and filter. * [Demand forecasting explained](/concepts/demand-forecasting.md) — How the probabilistic forecast is generated. * [Dashboard](/modules/dashboard.md) — High-level KPI summary across your portfolio. * [Buffer policies and safety stock](/concepts/buffer-policies.md) — How service level drives safety stock. * [Planning](/modules/planning.md) — Where stock coverage and inventory levels appear in context. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.flowlity.com/concepts/metrics-glossary.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.