Restock with context.
Estimate days of inventory cover from recent sales, and compare it against supplier lead time.
Fuji Apple3.0 days coverLead time: 4 days · Review reorder
INTERACTIVE PROTOTYPE
Know your stock.
Keep it fresh.
A clearer view of fruit inventory. Try a guided, interactive demo to spot restock needs, review freshness, and make better decisions from a simple spreadsheet.
✓ No account · Sample data included · No payment needed · Prototype only
Sample inventory overview DEMO
OVERVIEW
Freshness at a glanceProducts08In sample file
Restock signals03Needs attention
Freshness reviews053 need review soon
Fresh produce snapshotIllustrative weekly sales
↗
Freshness-first thinking
↗Spot inventory risks before they become waste.
◈ Insight-driven
✓ Built for fresh produce
FROM SPREADSHEETS TO SIGNALS
01 / THE PROBLEM
Fresh inventory moves fast.
Your insights should too.
Fruit and fresh produce businesses deal with short shelf life, changing demand, and fragmented spreadsheets. StokLensa makes those signals easier to see.
Catch spoilage risk sooner.
Flag products that may stay in inventory longer than their remaining estimated shelf life.
Keep the spreadsheet.
Import a simple CSV. Get a readable snapshot and suggested next actions without a complex setup.
02 / HOW IT WORKS
Less manual work.
More useful answers.
Our first prototype uses transparent calculations so you can inspect the assumptions behind each signal.
Give the demo a try01
Bring your data
Use a CSV with product name, stock, recent sales, shelf life, days stored, and supplier lead time.
02
Review the numbers
StokLensa estimates daily sales, days of coverage, and freshness exposure.
03
Decide what matters
See restock and spoilage signals. Use them as a starting point, not a guaranteed forecast.
NEW · INVENTORY MANAGER BETA
Go beyond charts. Manage your produce.
Try product records, batch receipts, stock-out tracking and freshness monitoring using fictional sample data. No account needed.
03 / LIVE DEMO
Try it with your data.
Start with sample fruit data and follow the guided tour. Want to try your own spreadsheet? Export it as CSV UTF-8 first. This demo processes files in your browser only.
NEW · 3-MINUTE PRODUCT TOUR
Not sure where to start?
Follow four simple steps using example fruit inventory. Free sample data: no registration or file upload needed.
Inventory workspaceSample produce dataset
Start with the right format.
One row per product or batch. Keep stock and sales in the same unit within each row.
CSV format & how the signals work
product,stock,sold_7d,shelf_life_days,days_stored,lead_time_days
- product
- Unique product or batch name, 1–100 characters. Add a batch ID for repeated products.
- stock / sold_7d
- Current stock / total sales during the last 7 days. Use the same unit, such as kg or pieces, within each row.
- shelf_life_days / days_stored
- Estimated total shelf life / time already stored, in days. Shelf life must be greater than zero.
- lead_time_days
- Expected supplier delivery time, in days.
Use exactly these six columns, in any order. Numbers must be non-negative, with a dot for decimals. Comma or semicolon separators are supported. Quote names containing commas. CSV UTF-8 only; up to 1 MB and 1,000 products.
Daily sales = sold_7d ÷ 7. Cover = stock ÷ daily sales. Remaining shelf life = max(0, shelf_life_days − days_stored). With zero sales, cover is unknown.
Restock review: stock is zero, or cover ≤ supplier lead time + a 1-day buffer. Freshness review: stock remains and cover exceeds remaining shelf life, or remaining shelf life is at most 1 day. A freshness review with at most 4 days remaining is marked “Freshness · soon”.
Multiple signals can apply. With no recorded sales, check the sales record and demand; an unknown cover does not prove spoilage. These are rules based on your estimates, not AI forecasts or a physical quality check.
Summary cards use the full dataset. The table and recommendations follow your search and filter. Export includes all inventory rows in the six-column input format.
Inventory signals
Estimates from the last 7 days of sales
| Product | Stock | Daily sales | Cover | Shelf life left | Signals |
|---|
04 / FRESHNESS LAB
Every batch has
a different deadline.
Try a practical FEFO (*first-expiring, first-out*) sales allocation simulation. Adjust demand and the planning window to compare potential expiry exposure. These are examples, not actual spoilage predictions.
INTERACTIVE SCENARIO Local · no server upload
Columns: product, batch, stock, days_left, daily_demand, unit_cost. The daily demand must be the same for every batch of one product. Each product is simulated independently.
FEFO in practice
Sell batches with the least remaining shelf life first, while checking actual quality. Incoming deliveries, temperature effects, returns, and demand variability are not modeled.
SIMULATED STOCK ROTATION
Rule-based · no AIMango Harum Manis
FEFO expiry exposure—units within horizon
Newest-batch-first exposure—comparator only
Illustrative difference—inventory cost, not guaranteed savings
| Batch | Life left | Stock | FEFO sold | Expiry exposure |
|---|
Illustration only. Shelf-life estimates are user inputs; inspect produce quality before selling.
Sold Expires within horizon Remains
05 / BUSINESS IMPACT
Know the cost of waste.
Then test what could improve.
Explore a transparent scenario using your own assumptions. Nothing here predicts real savings or represents measured results from customers.
YOUR SCENARIOInteractive · estimates only
Based on inventory purchase value, not revenue. Adjust the two assumptions yourself. Actual spoilage depends on temperature, handling, quality, and demand.
SCENARIO RESULTS
Hypothetically preventable waste value per month
Rp300.000Estimated present wasteRp1.500.000
Annual potential scenarioRp3.600.000
Compare scenarios against the proposed Pro plan, not a current subscription offer.
07 / PRICING CONCEPT
Start small.
Grow when it pays off.
These are proposed future prices for research, not live subscriptions. The prototype is free to explore. Features listed below are a roadmap unless already available in the demo.
Free prototype
Available todayTest the product without sending your CSV to a server.
Rp0 /month
- Browser-only CSV analysis
- Stock and freshness signals
- Search, filters and export
- Sample inventory included
Starter
ConceptFor one small fruit store that wants saved reports.
Rp79.000 /month
- Everything in Free
- Saved inventory history · planned
- One store and one user · planned
- Scheduled report exports · planned
Pro
Target planFor stores that need deeper decisions about freshness.
Rp179.000 /month
- Everything in Starter
- Batch and expiry tracking · planned
- Waste and margin analytics · planned
- Limited Claude-powered insights · planned
Business
ConceptFor produce suppliers and multi-outlet teams.
Let's talk
- Multi-outlet dashboards · planned
- Roles and permissions · planned
- Supplier performance · planned
- Custom report workflows · planned
Pricing is preliminary and excludes applicable taxes. There is no checkout, payment collection, account system, or Claude API integration in this prototype.
08 / COMMON QUESTIONS
Clear answers.
No hidden promises.
We prefer realistic calculations and understandable limitations over opaque AI claims.
What is the difference between the Live Demo and the Workspace?
The Live Demo is usable now with sample inventory or a local CSV file. The Customer Workspace is a concept preview showing a possible future logged-in experience. It does not currently store data or provide accounts.
Is StokLensa available to buy?
No. This is an early prototype. Starter, Pro, and Business are proposed plans; there is no checkout or payment collection.
Does the current prototype actually use Claude AI?
No. Current restock and freshness signals use documented formulas in local JavaScript. Claude AI integration is an optional future development and will be labeled only when it actually works.
What happens to the CSV I upload?
The current demo processes CSV content in this browser tab and does not send it to a StokLensa server. A future cloud service would need its own privacy policy, storage rules and security controls.
Can the tool guarantee produce won't spoil?
No. Shelf-life estimates are rough decision aids; actual quality depends on handling, temperature, batch condition and inspection. Always verify the physical product before acting.
Does the Freshness Lab work with my own batch data?
Yes, using the separate six-column batch CSV template. The lab uses a simplified day-by-day FEFO simulation and a newer-batch-first comparator. It does not predict true shelf life or guarantee savings.
Can I upload Excel files directly?
Not yet. Export your spreadsheet as CSV UTF-8 first. Direct XLSX import is on the development roadmap.
09 / ABOUT THE PROJECT
Built for the realities
of fresh produce.
StokLensa is an early-stage project exploring better stock visibility for fruit and perishable goods. We're starting small: a useful, explainable inventory prototype, developed openly and improved through honest feedback. StokLensa AI is the project name; this demo uses transparent rules. AI integration, accounts, server-side storage, and store synchronization are future work. A separate Inventory Manager beta stores fictional test records locally in the browser.
Public changelog ↗How demo data is handled ↗Customer workspace preview ↗