Worldwide Prices & Trends
Mandi Price Outlook looks forward rather than at today. Market Prices tells you what a crop is fetching now; this one is about whether holding it for a few weeks is likely to be worth the storage, the risk and the interest.
It takes the crop and, optionally, the district, and returns a directional view along with the change against three months ago — the comparison that tells a farmer whether what looks like a good price is actually a recovery from a bad one.
It is explicit that it is reasoning, not reporting. The screen says "AI-reasoned qualitative estimate — not live/sourced data", in those words, next to the answer. That is a deliberate and slightly commercially inconvenient choice: a price outlook that presents a guess with the confidence of a quotation is how a farmer ends up holding a crop into a falling market.
It carries a confidence level with every outlook, and where a current price could not be sourced it says that too rather than substituting an estimate and staying quiet about it.
It is for the decision a farmer actually faces at harvest — sell now at a known number, or store and hope — which is usually made on rumour from whoever was at the mandi last.
It is one of the paid feeds, and it pairs with Post-Harvest Loss Prevention, because the answer to "should I hold this" depends entirely on whether it will still be worth anything when you do.
What it is. Where a crop's price is heading, rather than where it is: a trend reading — strengthening, stable, weakening or mixed — with the change against three months ago, the seasonal factors behind it, and the reasoning stated so you can disagree with it.
When to open it. When you have a choice about timing: whether to sell now or hold, whether to book storage, whether to sow the crop at all.
Have ready.
Step by step.
What it will not do. It is a qualitative, AI-reasoned reading over sourced price data, disclosed as such on the screen. It does not forecast a price, does not know your storage costs, and cannot see what a policy announcement will do next week. Where it is uncertain, it says uncertain instead of choosing.
You are holding onion and deciding whether to sell this week.
What goes in:
What comes out:
'Mixed / uncertain' is a real output, not a failure. It appears when the evidence genuinely points both ways, and it is more useful than a confident direction produced by rounding away the disagreement.
Nobody can tell you that, and a tool that gives you a number is inventing one. What this does give is a defensible direction — strengthening, stable, weakening or genuinely mixed — with the change over the past three months as hard arithmetic and the reasoning stated so you can weigh it against what you know about your own district.
Take the outlook, then do the sum it cannot do for you: the price change you would need against the cost of storage, the weight and quality you will lose, and the interest on money you are not holding. A strengthening trend that gains less than your storage costs is a loss, and that arithmetic depends on your godown, not on the market.
That the evidence genuinely points both ways. It is shown rather than resolved into a direction, because a confident answer manufactured out of conflicting evidence is worse than an honest shrug — you would act on it with a confidence the data does not support.
For India, AGMARKNET via data.gov.in. For worldwide context, World Bank commodity markets. The screen names the source and says whether the response was fetched just now or served from cache.
Market Prices tells you what buyers are paying today, in which markets, and how far away they are. This tells you which way that has been moving and why. One informs where to take the crop; the other informs when.
An AI operating system for agriculture: free guides, manuals and tools written for every country.