Amazon.com receives 200 million unique visitors a day. Each of those potential visitors conducts a search, maybe more than one. That’s tens of millions of searches an hour that could lead to a sale for your product. The only trick is getting that product in front of the right buyer, at the right price.
To win in a market this vibrant and dynamic you need performance that rivals a top-of-the-line sports car. The platform you use to advertise needs to be highly responsive to the slightest change in your strategy, turn on a dime when the market shifts underneath it, and have a blistering top speed that lets it keep up with the flow of data in a real-time market.
I’d like to introduce you to the Flywheel Hourly Bidder, the performance sports-car of algorithmic bidding on Amazon.
What It Does
The Flywheel Hourly Bidder examines each ad target you have (keyword, PAT, etc.) and evaluates whether it has the best bid, multiple times a day. Each evaluation uses more computing power than the Apollo Mission and nearly as much math, to find the optimal bid for that target by taking into account your true margins for a sale and the likelihood of the buyer purchasing if they click.
Each evaluation is informed by Teikametrics’ understanding of the Amazon advertising ecosystem, an understanding that comes from managing the performance of over $150 MM of ad spend annually and millions of bids per day.
A Smooth Ride
The Bidder evaluates often, but only changes bids when needed. Constantly and unnecessarily shifting bids can lead to a lot of ups-and-downs in ACoS and a really bumpy ride.
Most ad targets require bid changes only a couple of times a day, or less because they don’t see a large volume of impressions, clicks, or sales. But by evaluating the bid multiple times a day, the Flywheel Hourly Bidder makes those changes the moment there is enough data to justify them, and cuts down on valuable hours of opportunity lost to a bad bid.
Blistering Top Speed
The Flywheel Hourly Bidder has the ability to pull data from Amazon on every ad target you have in play and calculate a statistically informed bid for that target, every hour.
A seller doing $1 MM in annual sales is going to require, on average, 40,000 bid changes a month to keep their ad performance optimized. For each of those changes, the bid will need to be evaluated a dozen times at least. That’s 500+ Million bid evaluations a month, each of which requires a graduate degree in statistics and a background in econometrics to make correctly.
No human can do that. More to the point, no sane human would want to try.
Flash sales. New multi-product bundles. Price changes. Shipment delays. There are dozens of reasons a seller may need to change their advertising strategy and ACoS targets for a product in a hurry. Lost time in making these changes can lead to lost profits. When you tell your advertising platform to shift gears, it needs to make that shift gracefully, no matter how fast it’s currently going.
Since the Flywheel Hourly Bidder is evaluating targets multiple times a day, it picks up on the changes to your ACoS targets quickly and responds fast by raising or lowering your bids. It also protects against ‘sales whiplash,’ the result of a big target change leading to a sudden drop or spike in sales. The algorithm approaches new targets with appropriate caution and gets bolder as it gains data.
The market on Amazon is one of the fastest moving in the world. Traffic patterns can change. A Tweet somewhere can create a rush on a particular keyword. A competitor can show up with cut-throat pricing and spendy ad-budgets, and leave just as fast.
The Flywheel Hourly Bidder is ingesting and reporting data every hour to keep on top of what’s happening in this dynamic market. The volume of data it handles is immense and one of our major investments when building the Flywheel Hourly Bidder was creating the data pipeline that feeds it.
There is no possible way to know your exact performance for a given hour on Amazon.com. Amazon’s own reporting systems are daily. They also have a reporting lag of up to 72-hours for many kinds of data. However, it is possible, every hour, to know a little more than you knew before and to respond to it. That’s what the Flywheel Hourly Bidder does. It senses market shifts before they solidify and begins heading your bids in the right direction, early.
The Bottom Line
If you’ve never driven a Lamborghini (or one of its equally exalted cousins), you should try it, at least once in your life. And if you’re selling on Amazon and have never experienced what it’s like to have an hourly algorithmic bidder handle your ad spend, you should give the Teikametrics Flywheel platform a test-drive.
Having a high-performance platform will allow you to take bigger bets, take more risks and be a sharper, edgier seller. You wouldn’t want to take a mid-range sedan up to 100+ mph, and then try and change lanes in traffic, even if it meant you might get ahead. But if you had a high-performance machine driven by algorithms, one that could compensate for the slowness of human reflexes, you just might take that chance, and beat the other guy to the finish line.
Ready to start moving quicker while maximizing ad spend on Amazon? Try the Flywheel Hourly Bidder today!
CTO | Teikametrics