Forecasting Indian Tea Auction Prices

The challenge

As the second-largest producer in the world, India is in close competition with China and Kenya in the centuries-old trade of Tea. While the system of auctions provides the upper hand to the planters to sell their produce to the highest bid, the buyers have to carefully calculate their costs and profits before purchasing, so as to remain ahead of their competition not just locally but internationally.   

Tea traders have to make calculated decisions on their purchases, not just in terms of quality and quantity, but also with regard to time and venue since the tea auctions are held each week simultaneously in different states of the country.   

The pandemic has only added to this complication by skewing much of the past trends and data. According to Tea Board India, tea produced in the North Indian region namely Assam, Darjeeling, Dooars & Terai fetched a price 50% more in July, August, and September than in the corresponding months of 2019. The local consumption has risen significantly, packet tea players no longer pay as much attention to their own particular blends and there is a sudden crash in prices in some of the auctions.  

This presents them with the challenge of foreseeing the price for a particular category of tea they are interested in purchasing.    Price variations largely depend on the availability of the specific tea type, the credit periods, and the total demand at the time of bidding. Therefore, predicting how much a quantity would cost to an acceptable degree of accuracy is difficult.    

This presents the tea buyers of India, large and small with the challenge of ‘when to buy and how much to buy’     


Why is making the decision to purchase complicated? 

Tea buyers (both local buyers and tea exporters) buy in large quantities to manage demand for the upcoming weeks. This can amount to thousands of rupees per auction. On top of the purchase costs, these companies also must manage their manufacturing and inventory costs for bulk purchases.   

Tea sold at the auctions is extremely price-sensitive, especially because one’s own purchasing decisions, as well as that of the competitor, contributes to the price of a particular type of tea; the higher the demand for a certain grade higher the price would be.     

One misjudged decision can let the competitors take over the market space and leave long term damage to the business. 

On top of all this is the complexity of the market dynamics. It requires years and years of experience to understand the potential market instabilities and make confident decisions on purchases. And now, tea traders must account for higher unforeseen uncertainties to the matrix. 

Tea auction linear squared

What industry-specific parameters are considered?

Most predictions in the trade are currently done by subjective calls from the industry experts. Indisputably, the number of variations and influences of external factors is overwhelming enough to be calculated by a novice hand; it needs the experience gained over the years. This is why usual trend-seasonality forecasting methodologies fail to produce acceptable results. When manually predicting price variations, industry experts look into the following;

  • Supply (of various grades from different regions) 
  • Next season’s yield
  • Foreign market performance  
  • Local market stability  
  • Natural Disasters  
  • Credit periods  

What does FORECAST Squared do?

FORECAST² will give a modern-day perspective for the long-held traditions of the Indian tea industry by analyzing historical data and building forecasting models.  These time series models can incorporate all considerations an expert makes in terms of variables and external factors that drive the forecasting variable. This allows users to generate customized models that produce highly accurate forecasts for a significantly long-term horizon.   

How does FORECAST Squared work?

Historical data; weekly prices, quantity supplied, and the total demand seen at each auction in each region. With the use of Machine Learning algorithms, FORECAST² builds models that capture the trends, seasonal patterns, and calibrates the effect of supply and demand on price. These models can accurately estimate the most likely price value for each week by tea category. It also recognizes that the price can still vary due to unforeseen factors. Therefore, it estimates not only the price but also the variability of the forecast. 

forecast squared

An added benefit of having the demand and supply is that users can generate different scenarios for comparison. While FORECAST² can produce accurate forecasts for demand and supply figures for the forecasting period, it can give visibility to how the price may react under varying conditions, by simply keying in expected values for the supply and demand for future auctions. Running such scenario analysis can help tea purchasers to make data-driven decisions on when to buy and how much to buy.  

The Outcome? 

  • Achieve over 85% accuracy in tea prices for any given type of tea
  • Predict the monthly supply of tea with 80% accuracy
  • Plan purchasing plans across weekly auctions to minimize the buying cost using the insights on auctions prices  
  • Minimize high inventory buffers and wastage  
  • Approximately 30% accuracy boost for COVID period forecasting


Sign up for a trial to try out FORECAST2!

Sri Lanka
No. 63/1,
Dhrmapala Mawatha,
Sri Lanka.
No. 9, Raffles Place,
Republic Plaza,
Singapore, 048619.
Menara Digitaraya,
Jl. Raden Saleh Raya,
No. 46A, Jakarta,