21 Day Forecast Challenge

Can Peak's Decision Intelligence improve your demand forecasting? Put us to the test.

Get Started Now

Want to hear what the 21 Day Forecast Challenge could do for your business? Enter your email below to get the lowdown.

We'd like to help you, free of charge!

At Peak, we’re passionate about demand forecasting and believe it’s one of the fundamental pillars needed for a successful business. That’s why we’ve teamed up with AWS, integrating its powerful Forecast algorithm into our Decision Intelligence platform.

This same algorithm sits at the core of Amazon’s retail and supply chain operations – and now you can discover how it can help you and your business.

Find out more

How can this help you?

The 21 Day Challenge is designed to rapidly identify opportunities for Decision Intelligence to improve your business processes, saving time and money. Some of the major benefits are…

Release Working Capital

Greater accuracy leads to improved inventory planning and puts an end to speculative ordering. This frees up working capital and ensures fewer write-offs of obsolete inventory.
Reduce costs

AI has the potential to reduce forecasting errors by 20-50%. It will also lead to improved workforce planning and reduced waste resulting in higher margins.
Save valuable Time

AI speeds up the forecasting process, enabling you to focus on Supply and Operations Planning, resulting in agile decision making across the supply chain.
Improve Efficiencies

AI can remove the guesswork from the process of demand forecasting, taking care of routine tasks and freeing your team to make the strategic decisions that will help your business grow

The challenge

We'll run your data through the Amazon Forecast algorithm and confirm, within 21 days, just how much we can improve the accuracy of your forecasting. This is a great first step in helping you to drive greater efficiency in any of the following areas:

Inventory optimization
Allocation and assortment
Demand and labor planning
Warehouse optimization
Logistics optimization
Distribution resource planning

Next steps

We’ve outlined each step of this process below. Once you have contacted us, we will:

01

Set up a quick call

We’ll quickly assess your needs and confirm whether we can help

02

Explain what's possible

Capture the scope of work, to get a better understanding of what you’re trying to achieve

03

Upload your data

We’ll send over a mutual NDA and a secure data link, so you can upload the data we require access to

04

Run Amazon Forecast on Peak

We’ll run your data through Amazon Forecast on our platform, maintaining regular communication with you

05

Playback the results within 21 days

You’ll receive a bespoke report containing our findings, and we’ll schedule a video call to walk you through the results and take your questions

06

Delete your data

Your business’ data will be removed from our AWS cloud infrastructure once you instruct us to do so

FAQs

We have pulled together some FAQs on our 21 Day challenge that will quickly answer your burning questions. If your question isn't below, just get in touch – we'll get an expert on the phone to you in no time.

How much data do I need to submit?

A minimum of 12 months of historical data at a product/SKU level will allow us to confidently run it through Amazon Forecast.

What sort of forecasting algorithms will be used on my data?

Amazon Forecast uses AutoML(Automated Machine Learning) which automates complex machine learning tasks, such as algorithm selection, hyperparameter tuning, iterative modelling, and model assessment. It evaluates and compares the forecast output of multiple algorithms, and chooses the best one based on the prediction accuracy per datasets.

Does my data need to be in a specific format before I can take part?

It’s not essential but in order for us to return our results within 21 days we’d need to receive a target time series data set. This data type defines the target field that you want to generate forecasts for. For example, if you want to forecast the sales for a set of products, then you must create a dataset of historical time-series data for each of the products that you want to forecast. We can provide the details of the mandatory data fields upon request or during our initial call.

Will someone be able to explain the results and be challenged on what I’m receiving?

Yes, we’ll set up a call with you to walk you through the results. A Peak data scientist will be available on the call to discuss the results in detail.

What will the report tell me?

The report will confirm how confident we are at predicting demand in your business. We will provide confidence levels and aim to evidence a couple of examples at a product level to further support our results.