When I started selling on Amazon in 2014, it was almost laughably easy to rank a listing for a keyword...any keyword, really.
Back then I was targeting a massive keyword (over the year 2019 it averaged roughly 100k searches per month, and demand for a baby product like this was not much lower in 2014).
I did a single deep discount promotion, selling my baby carrier for $5 with a coupon (regular price was $67) and managed to sell around 40 units in a single day.
There was an additional 6 units that spilled over into the next day from that promotion. Before the discount campaign, I was lucky to get one sale every two days.
Two days after the promotion my listing was ranked number four on page one and I made almost 20 sales. I averaged 15 per day from that point forward.
That’s an average of a little over $1,000 per day revenue! All from one promotion that cost me a sacrifice of around $2,852 in potential revenue.
And I wasn’t alone. There were scores of people using this method to rank on the top of page one for their primary keywords and then watching the organic sales roll in. Ahhh….those were definitely the days.
Shortly after my baby product experience I started using the same method to help clients, and even began working for a promotion service that offered the same for all Amazon sellers. I became the “king” of ranking as I oversaw and facilitated “blasts” for hundreds, and eventually thousands, of sellers.
Learn all about Amazon's A9 algorithm, the "honeymoon period," and how to leverage these things for greater keyword rank in our comprehensive 70 page white paper. Just subscribe to download NOW! ⬇️⬇️⬇️
HISTORY AND PRINCIPLES OF AMAZON'S A9 ALGORITHM
Over time, these easy results diminished. I did my best to stay on top of how to adapt, though. Testing different URL structures, different discount amounts, different volume levels and promotion timing; I was determined to stay ahead of the Amazon algorithm curve.
With each small tweak in tactic, renewed vigor injected itself into the results. Where lower volume blasts stopped working as well, we increased the number of units available.
Where one URL stopped getting us quick results, we shifted to a different URL. Where single day promotions stopped rapidly ranking, we changed it to multi-day or staggered promotions.
In the end, however, ranking through deep discount promotions remained the most effective and business-changing strategy. That is, until the big Amazon TOS change of 2015.
I will never forget the day it happened.
The day Amazon changed the Seller Prohibited Activities and Actions section of their Terms of Service.
The prohibited activities largely focused on obvious infringements back in 2015. Things like not buying your own products or leaving yourself or your competitors reviews, not having multiple seller accounts without permission, etc.
However, the big 2015 update injected stipulations about “manipulating sales rank” that were vague and confusing at best. Essentially the new terms stated that it was now prohibited to give away “excessive” coupon codes for the purpose of “sales rank manipulation.”
While there were dozens of ways to interpret this, and still many questions left unanswered, a lot of people were convinced they knew what this meant and who it was targeting.
The first assumption was that this was specifically targeting “blast services” or otherwise services that offered keyword ranking for Amazon listings.
Many were convinced this was a blatant attack on the entire business model, saying that things like super URLs were what Amazon was referring to with the “sales rank manipulation” and that the deep discounts violated the new “excessive coupon” stipulation.
As a result, the entire ground under our industry shook. Many sellers stopped promotions entirely and the service I worked for (as well as all the others in the industry) took a huge hit.
I had to get to the bottom of this. Not only were the terms extremely vague but this went entirely against Amazon and their customers’ best interests.
Deal sites were a fantastic way to drive people to Amazon.com and they also got the benefit of awesome discounts. This represented a win-win for everyone so it didn’t make sense why terms would put an end to this.
After several back and forth emails and phone calls over the span of a couple of months, I finally began to get consistent answers from Amazon. Not just me either, but other players in the space (service providers) as well as colleagues I had instructed to reach out with specific questions.
What we were told didn’t necessarily define every aspect of the terms of service, but it did fill in the mysterious gaps from the newest addition.
Basically, Amazon was attempting to shut down the practices of sellers facilitating fake orders or buyers with several (read hundreds) accounts placing orders to drive down BSR.
Excessive coupons referred specifically to giving out discounts in bulk to a singular actor in an effort to mitigate the costs. And Amazon defined sales rank manipulation as artificially driving down BSR with orders not placed by legitimate accounts.
This revealed the truth that the strategy of deep discount promotions to rank had not been targeted and was not going away, despite the fervent efforts of competitors to convince the public otherwise (that was a whole can of other drama we won’t get into here).
However, this didn’t mean that Amazon wasn’t paying attention to promotions as a keyword rank strategy, or that they didn’t care. There were other moves being made.
Shortly after the big TOS shakeup Amazon also removed a key promotion feature from its catalog; Dollar-off coupons.
So, one of the key components of stimulating ridiculous sales volume is reducing as much friction for the buyer as possible. The easiest way to do that is to make the product entirely free and that meant, for a while, many were using dollar-off coupons to remove the entire dollar amount of the purchase.
When Amazon removed this feature, it meant percent-off was the only option and a number of complications followed.
Once everyone adjusted to the new way to do promotions, this action was forgotten. Everyone assumed Amazon had their reasons for removing the feature but no one really knew why.
That is until a former Amazon employee revealed on the (now, unfortunately no longer running) Private Label Podcast that the removal of the dollar-off coupon feature was a concerted effort by Amazon to do away with free-product ranking.
This was literally the first time we had ever seen Amazon address any tactic that directly correlated with keyword rank and even us experienced veterans, at that point, had been too new to the game to truly grasp what was going on.
But clear as day it was explained on that podcast that Amazon did not like products that had no dollars exchanged for it to rank above products people were genuinely purchasing.
This implied that not only was Amazon paying attention to keyword rank and promotion activities, but also they were concerned with revenue to some degree.
The plot thickens…
The other thing we could get from this action, as well as many that have since followed, was that Amazon appeared to prefer to handle controlling the search and buy experience programmatically.
See, they could have easily just started suspending sellers who were using deep discount promotions specifically to rank.
They would have been justified in doing so too, since technically their vague TOS did state that the use of “excessive coupons” was prohibited. They could have defined that any way they like.
That would have been punishing people for using Amazon’s promotion tools.
Sellers would have been punished and buyers would have been punished as well (by not getting as many deals). That was a lose-lose situation.
It made MUCH more sense for Amazon to appreciate the fact that sellers use their promotion tools and allow them (and even encourage them, as they do) to keep using them to draw buyers to the marketplace.
It also made sense that Amazon would want deal sites and other promotion services to exist to help facilitate that at scale.
See, promo services put coupons into the hands of thousands of buyers in order to help sellers rank (win). Those buyers get a great deal (win). They also are on the Amazon platform buying other products, so Amazon makes even more money (win).
Also, the use, even over-use, of these promotion tools was not technically against any terms (even to this day). So it was a lot easier for Amazon to simply police the ranking algorithm programmatically.
While we cannot be completely sure when it happened, it seems likely that around the time these TOS changes occurred, Amazon implemented a “buyer trust score.”
Essentially, this is a score that adjusts while Amazon tracks the activities of a buyer. They do this for a number of reasons:
A buyer earns a lower score through a few activities, but the ones that most pertain to third-party private label sellers are when they ONLY purchase items using deep discount coupon codes and when they leave reviews for those items.
Amazon does like when sellers utilize their promotion channels, so it would seem counter-intuitive for them to “punish” a buyer for using coupon codes. However, that doesn’t mean they like them to be “abused.”
And when a buyer never spends money using the platform in the way it was intended; to search, find and purchase products at full price, then Amazon certainly does not want that person’s shopping behavior to impact the search and browse experience for other customers.
Aside from the purchases and behaviors of individuals who only use the platform for deep discounts propping up keyword rank for certain listings, the reviews they leave tend to be skewed much more heavily to the positive.
These two factors then take away from the organic buying experience of other customers.
If Amazon cares about this “search and browse” experience, and doesn’t want buyers OR sellers to manipulate it, why don’t they just suspend sellers who are blatantly attempting to use tactics to do so?
So, here’s the thing; the overuse of an Amazon promotion channel isn’t necessarily against any TOS.
Giving out deep discount coupons aren’t (after all Amazon allows you to create them).
Doing buy-one-get-one deals aren’t (even though it makes your unit-session percentage awesome).
Sending people to purchase through a storefront URL isn’t (hell, Amazon provided that one to us).
Not only that, but launch services send buyers to the Amazon platform, where they spend money, and they give those buyers a great deal. What could be better for Amazon?
They get free traffic. They get sales. The buyer gets a great deal. And most of the time those buyers spend money on other things as well. That’s what we call in the industry a win-win.
If Amazon suspended sellers for using tools they make available to them and basically said “hell no” to a ranking strategy that involves the purchase of product from real buyers with real accounts, launch services would go out of business.
This wouldn’t be the best way Amazon could serve its customers. It makes more sense for Amazon’s ranking algorithm to be a self-regulating system. So that is why they choose to handle most issues with regard to keyword ranking programmatically.
The next big major TOS update happened in October of 2016 when Amazon banned incentivized reviews.
See, Amazon may be able to handle keyword rank programmatically, but they were getting TONS of negative press for outright fake, or unevenly skewed reviews. That meant they had to address this problem head on.
There had been a few articles here and there, but the whole issue of fake reviews on Amazon came to a head with an article that went viral by a group of data scientists working for a program called Review-Meta.
In the article, they showed plainly and for all to see, that the majority of Amazon’s reviews were skewed positive due to incentivization. It was so blatant and undeniable that it caused a surge of media to cover the topic and Amazon was under the microscope of public scrutiny in a big way.
Shortly after this happened, they amended their prohibited seller activities and actions terms yet again, this time to state explicitly that free or discounted product, even with a coupon code, were no longer allowed to be exchanged for a review.
This was a big jump from the previous terms that merely stated you couldn’t refund people for a review or compensate them.
And this caused yet another hailstorm of sky-is-falling panic from the sellers.
While many people used promotions as a mechanism for ranking their products for keywords, it was taken entirely for granted that a by-product of that would be reviews.
It was also an unspoken understanding that reviews were necessary for continued ranking and results.
Everyone’s strategy was tossed onto its head yet again and many sellers injected chaos needlessly into their marketing plans for fear of retribution from the great life/business-giver Amazon.
Amazon slowly rolled out more, smaller changes from there. Many of which were never noticed, or they were overlooked. Most of which were connected in some way to grander plans of making the platform the most readily able to keep shoppers entirely dependent.
Some of these subtle changes included:
Search term limitations would play a role later in relevance (what would become the most important factor for keyword ranking).
The addition of language addressing search and browse experience, while vague and not necessarily threatening, was the first real indication that this was something that Amazon cared about (and the only link in the TOS to keyword rank).
The changes to reviewer community guidelines finally held buyers accountable for actions that hurt Amazon’s organic experience too. As such, this later would lead to Amazon deleting and even banning entire reviewer accounts.
This last addition disrupted the couponer and freebie communities that had thrived on Amazon’s lax treatment of promotions.
There were whole communities and forums dedicated to spreading these discount-for-review deals that dried up almost overnight as buyers lost both their review history and review privileges.
All in all, while some sellers got unfortunately swept up in the whirlwind of it all, the changes that Amazon has made have been good. They’ve led, by and large, to a safer and more organic platform that can potentially allow any brand, no matter their size, a chance at visibility and big profits.
And thus is the sordid history of Amazon’s A9 algorithm.