
Optimize gross margin by analyzing inventory levels, costs, and variances in real-time to reduce waste and improve pricing.
.png)


Inventory typically represents 20-40% of total assets for manufacturers and retailers, yet most companies only analyze it monthly or quarterly – far too late to prevent costly issues. This use case controls costs related to inventory and production, ensuring cost of goods sold (COGS) is accurately tracked and optimized. It gathers data on inventory levels, movements, and valuation (standard cost vs actual cost, purchase price variances, production yields) to help controllers and supply chain managers maintain optimal inventory levels and understand COGS components. It highlights issues like excess stock tying up millions in working capital, slow-moving items heading toward write-offs, or variances between expected and actual production costs that erode margins. In manufacturing and retail where inventory dominates the balance sheet and COGS represents 60-80% of expenses, this visibility is vital for profitability. The result: one retailer reduced inventory by 15% while improving in-stock rates by identifying misaligned stock levels, freeing millions in working capital.
Without proper analysis, companies carry too much inventory "just in case," tying up cash and incurring holding costs (warehousing, insurance, obsolescence risk). They don't realize some items are slow-moving or obsolete until forced write-downs hit the P&L.
In production environments, differences between standard (planned) cost and actual cost (due to waste, scrap, overtime labor, material price changes) aren't obvious until much later or only at aggregate level.
When inventory costs aren't updated or variances not allocated properly, product costing becomes inaccurate, meaning pricing decisions are wrong. One product could be overpriced (hurting competitiveness) while another is underpriced (selling at a loss).
Companies get hit with inventory write-offs (for obsolescence, damage, theft) that weren't planned or anticipated. Without continuous monitoring (comparing inventory records vs actual counts, checking stock aging, analyzing turnover rates), these accumulate and then suddenly hit the P&L.
Keboola blends data from ERP (inventory counts, movements), sales forecasts, and warehouse systems to give a complete picture. Controllers see days of inventory on hand by product category, identifying which items are overstocked (relative to forecast) or understocked.
The platform automates calculation of production variances: material usage variance, material price variance, labor efficiency, overhead absorption – if standard costing is in place. It analyzes per product or per batch order and rolls up to show where the biggest impacts are.
By combining cost data and sales data, Keboola shows actual gross margin per product or category each month compared to expected. If certain products' margins are eroding (due to cost increases or excessive discounting), the tool makes it immediately visible.
Beyond current data, when trends indicate issues, Keboola issues proactive alerts: "Inventory for Product X will exceed 3 months of sales within next 2 weeks based on current trajectory" or "Scrap rate in Department Y has increased 5% above norm for two months running – investigate."
[stakeholder] Supply Chain Manager
[stakeholder] Cost Accountant/Plant Controller
[stakeholder] CFO/COO
ERP reports give raw data (stock levels, valuations) but not analytical insight across multiple dimensions or over time. Keboola consolidates across modules (inventory, production, sales) to deliver contextual analysis like combining sales forecasts with inventory, or linking variances to financial outcomes. ERP reports are often rigid – customizing them for "flag items with over 90 days supply" or "split cost variance by cause" isn't easy or requires expensive modules. Keboola lets you tailor exactly what you want and include external data (commodity price indices to explain cost changes, competitor stock levels from market data). Plus, if you have multiple ERPs or data sources (common in large firms or after acquisitions), Keboola unifies them into one view. ERP is the transaction source, but Keboola is the brain making sense of it all flexibly. Many companies use ERP for processing but rely on BI platforms for controlling insights – that's the gap we fill.
Ideally, inventory data updates daily (especially for fast-moving businesses). Variance analysis typically happens monthly when books close (when standard vs actual gets computed formally). But you can track operational metrics like yield or scrap daily/weekly to catch issues early – those won't be in final dollars until month-end, but trend-wise you'll see problems developing. So: inventory levels – daily; production metrics – daily or weekly; full costing analysis – after period close (with mid-period estimates if needed). Keboola handles different cadences, pushing updated data to the same dashboard. Early in the month you see operational previews, after closing the financial version. This is far better than waiting weeks after month-end to realize you had major variances. In volatile cost environments, some companies even do mid-month cost roll-ups to foresee margin impacts – which the system supports by mid-period actualization of some costs. More frequent data refresh equals quicker reaction.
Yes, we can implement those calculations. LIFO reserve calculation (difference between FIFO cost and LIFO cost) can be automated by layering inventory by purchase dates/costs and valuing accordingly – if your ERP doesn't do it or if you want separate verification. Lower of cost or market (net realizable value) testing can be done by bringing in selling price info: compare cost vs current selling price for items, flag if cost > price (impaired inventory), then produce suggested write-downs. These tasks are often done with painful spreadsheets annually; we can do them periodically with up-to-date data, so you aren't surprised by big write-downs at year-end. If multiple accounting standards require different treatment (maybe FIFO for internal, LIFO for tax), we can compute both. Transfer pricing for inventory (if goods move between countries at certain prices) can be monitored to ensure no unintended profit buildup in one entity's stock. Essentially, any accounting logic related to inventory can be programmed in, ensuring compliance and foresight. It's like having continuous audit on inventory values so you remain compliant and optimize outcomes (like timing LIFO layer liquidations). This goes beyond basic controlling into strategic inventory accounting, which Keboola's flexibility supports.
AI tools fail when they don’t connect to your real data or respect production workflows.