Accurate forecasting

for better business results

Amazon Forecast is a fully managed, machine learning service by AWS, designed to help users produce highly accurate forecasts from time-series data. Amazon has utilized machine learning to solve hard forecasting problems since 2000, improving 15X in accuracy over the last two decades. Built on the same technology as is used at, Amazon Forecast can be utilized for a variety of business use cases, from financial and resource planning to predicting future performance and product demand across a wide spectrum of industries from retail to healthcare.

A Premier AWS Consulting Partner, Cloud Kinetics specializes in helping customers implement cloud-bases Analytics solutions, to enhance opportunity identification, decision-making and developing business strategies.

Forecasting is the Science of Predicting the Future

Why Amazon Forecast?

50% More Accurate Forecast using ML

Reduce Forecasting Time from Months to Hours

Create Any Time-Series Forecast

Secure Your Business Data

CK Customer Success Story

About the Company

Customer is an financial services company offering personal banking services in the areas of Deposit Services, Cards, Bill Payments, and ATM services applications with a large user base across Thailand.

Business Challenge

  • Customer uses large IT infrastructure to support their banking operations solutions. In a phase of rapid expansion, customer needed to identify future costs, to be able to make better investment decisions impacting business profitability. Accurate forecast would be key   to be able to make financial and operational decisions based on economic conditions.
  • Efficient way to collect & handle data irregularities and comparative analysis across various algorithms so that patterns/trends can be found with high accuracy.
  • Building a powerful BI solution to present the trends & insights enabling decision-making

Solution Highlights

  • Using P50 output value to create Quicksight visuals
  • Setting up data repositories, Forecast and visuaIization platforms for the project
  • Automate importing of Customer’s Monthly costs data across all environments
  • Pushing Targeted Time-Series data into Amazon Forecast using Lambda functions
  • Predictor Training using Auto ML mode over a forecast horizon of ensuing quarter
  • Validation of winning metrics result (Arima, in this case) by comparing key value metrics like wQL, MAPE,RMSE
  • Generating and exporting forecast to S3 bucket

Success Metrics Achieved

  • Providing a number of new key indicators, ML-based insights, and forecasting services with high accuracy usually higher than 90%.
  • Simplify and accelerate investment decisions
  • Providing visibility into various factors causing expenditure trends
  • Automated handling of data irregularities
  • Scalable Solution with Optimized TCO
  • Solution integration with existing architecture

Read more Success stories.