How to Utilize AI for Amazon Forecasting Amid COVID-19 and Beyond


If just a year ago, someone told you that “toilet paper” would become the leading search term on Amazon in March 2020, you would hardly believe them. The coronavirus crisis has truly shaken up consumer behavior, but also brought novel opportunities to many.

In Q1, Amazon has seen its sales up 26% in comparison to the previous year. To take advantage of this growth, vendors must position themselves to avoid a lack of inventory, supply chain bottlenecks, slow stock replenishment and other obstacles to driving sales.

Navigating this precarious landscape requires a solid strategy — and forecasting should be its foundation. With AI-powered predictions, vendors can utilize historical Amazon data to draw statistical patterns and model multiple situations.

How can vendors leverage AI to make effective, data-driven decisions?

Price optimization to boost sales

With the ever-growing e-commerce competition, vendors should differentiate their products — and price is one of the focus points. 9 out of 10 consumers price-check a product on Amazon, and with more time on their hands due to social distancing, they will definitely continue to do so. 

By leveraging AI, vendors can get insights into their pricing options and tailor the most effective discounts. For example, if you see that the average price of a specific ASIN is $100, you can project how sales would be affected were it decreased by $10 or $20. In the same way, vendors can experiment with coupons and gain intelligence on how to optimize their performance.

Agile inventory management

Unfortunately, today’s lack of inventory and the subsequent delays in shipments have caused a decrease in sales for many vendors. With AI, vendors can predict the impact of increased “glance views” (Detail Page views) and better plan ahead.

Vendors can also benefit from proactive and localized inventory management. When shipping to Amazon, they know which warehouses it sells to, and Amazon also provides end-user ZIP code information. Such granular geographical data empowers vendors to be strategic in observing shipments to regional fulfillment centers.

For example, if you sell air purifiers and your sales spike during forest fires in California, you can leverage ZIP code reporting to anticipate the amount of inventory needed in that specific area.

Insight into ad success

With the slight increase in Amazon Ad conversions, vendors should resist the temptation to cut back advertising budgets. Amazon Advertising (AMS) can be leveraged to ensure that with more customers browsing, Vendor listings dominate the top of search and optimize captured share of impressions, clicks and resulting sales.

Not only can AI be used to predict the performance of ads, it can also analyze the impact of increased spend and bidding. This brings us to the true core of running successful ads: incrementality. It’s not just about pushing promotions, but also about understanding where they yield the highest effectiveness and conversion rates. By looking at keywords and bidding to determine how to truly gain incremental sales, vendors can unlock the full potential of AMS.

Now more than ever it’s crucial for vendors to maintain profitability in advertising, excellence in operations, and effectiveness in promotions. By making full use of data, they can outline trends to forecast, plan and justify their investment in Amazon both during this crisis and beyond.

Rohan Thambrahalli is founder and president of UpstartWorks.

This ad will auto-close in 10 seconds