Matthew Ferguson explains how following “the data brick road” and embracing change can prevent your Amazon sales from flatlining
Five years ago, we started selling kitchen products on Amazon.de. We started with about 500 SKUs and doubled this to 1,000. We then got rid of 500 SKUs that didn’t sell well, and listed new products. At the moment we are at about 800 SKUs, and always sell between 14,000 and 15,000 units a month.
We also have a top seller that sells 2,000 units every month. We had two months where we weren’t able to stock this product so didn’t sell a single unit, yet we still sold between 14,000 and 15,000 units overall those months.
If we have some weeks above average then the next few weeks are below average. We do not change prices much and we do not change our sponsored ads. This is really frustrating as we have the resources to grow, but no matter what we do, we always have the same amount of sales every month.
Do you know anything about Amazon controlling sales, to keep sellers on a certain level?
— Bernd M., Austria
Sounds like an interesting topic to explore. It’s worth mentioning, I don’t have Jeff on speed dial and sadly don’t have any manner of inside knowledge on Amazon’s complex search and ranking algorithms. Besides, if I did have him on speed dial, I’d probably ask about those interesting Alexa eavesdropping stories.
One thing I can say with certainty – it’s easy to read too much into the tea leaves. We have an innate need to find patterns in the randomness. But to take advantage of patterns we may or may not see, we need to be constantly trying new ideas. That means an ongoing cycle of changes.
A quick tally
Lets first summarize the key starting points and try to keep these in mind throughout:
- You had 500 unique SKUs.
- You added 500 more SKUs.
- You cut some under-performing SKUs, resulting in a net 800 left over.
- Despite the SKU cut and 300 new additional items, you continue to sell the same units per day.
- You have weekly sales peaks and troughs, but seem to have a predictable average each month
- You keep advertising and pricing fixed.
- Your top seller consistently sells well, except for a 2 month hiatus.
In conclusion, you don’t get why sales remained virtually the same during this journey. You figure, given the steps taken, you should have seen something: changes. Things will never be the same, oh yeah? Wait, sorry, that’s a Tupac song.
The Top Seller Song
Let’s first pluck out the top seller topic. This is the abnormality. Everything else should change, yet doesn’t. You couldn’t get stock, so your overall sales should have been down around 2,000 units per month. But instead they were flat.
I have had this kind of conversation with a lot of sellers. A belief there is something “going on” with Amazon “at times”. A gut feeling something is wrong, or working against them in cycles. I don’t blame anyone for thinking like this. It’s a human desire to want to make sense of things. Two sales less than yesterday? It must be because of the thunderstorm. Eight sales more this week? Must be that error I fixed on the image. No sales yesterday? Amazon must have downgraded me somehow! We by our very nature want to explain things, and believe every event has a legitimate, traceable reason.
Amazon’s algorithm won’t always be perfect. But its multi-threaded and dependent on lots of overlapping variables. I don’t know this because Jeff told me, I know this because it’s what search algorithms do. They use lots of different inputs, usually with some randomized elements involved, to produce an outcome.
Most likely, your sales staying flat even without your top seller was the result of an external force. External forces can be easy to spot when they positively or negatively affect a single item. They can be really difficult to nail down when the net effect is zero: boosting sales of your other products at the same time as your top seller going out of stock.
It is mightily coincidental that your other products exactly filled the gap left by your top seller going out of stock for two months. It’s seriously spooky. We may never know what happened. Sun spots?
Sometimes the reason changes occur may have nothing to do with you. And that’s why change is a very very important routine, because ongoing trials of new ideas can allow you to find opportunities.
No pricing or PPC changes?
You mentioned you don’t really change your prices or PPC campaigns. My question to you is unapologetically abrupt: Why?
I mean… why? Why are you not creating sale events, or dropping the price for special weekends?
Why aren’t you adjusting bids to be more aggressive at times, or adjusting them down on terms that aren’t returning positive gains?
Why aren’t you trying new campaign keyword combinations with different variations of the product, or wrapping campaigns around targeted search phrases with ASIN pricing adjustments?
Why aren’t you using repricing tools and PPC auto-adjustment tools, or having someone in your team do some analysis on past trends and trying out new ideas?
“The definition of insanity is doing the same thing over and over again and expecting a different result.” Attributed to Albert Einstein perhaps incorrectly, but wise words nonetheless.
Changes can cause you to break things which are running well, so I am not suggesting you go change everything and expect to win the lottery the next day in sales volumes. But changes are important if you want to have change.
If you want to see growing sales and a graph changing for the better, you cannot expect changes with the same consistent routine you’ve adopted for the past few years.
Follow the data brick road
It sounds like some number crunching might be too simplistic presently. You need to drill into all numbers and segment them out by theme, batch, type, etc., always striving to drill down to the final denominator. By denominator, I mean the final comparison of values which give you an idea, or insight.
You added 300 new SKUs net, nearly doubling your catalogue, but did you add 300 new items into a new category? Did you add 300 new SKUs like your top seller product? Did you add these 300 as new variants on existing items, or were they a completely new range? Were these 300 new SKUs new ASINs, or on existing listings? Were existing listings well optimized? Were you competitive?
And for the 500 weak sellers, were they always weak sellers, or did they become so on Amazon? Were you competitive? Did they have high traffic and low conversion? Or low traffic and low conversion?
I can understand the logic you would associate with volume. If you have more items to sell, surely you should sell more, right? Well… not necessarily. You need to add SKUs based on what your data says. Data will paint pretty paths to pivot along, and sunken dark roads riddled with quagmires.
For example, if you removed a few hundred SKUs which weren’t selling, perhaps “Kim Kardashian lip hair cream remover”, and added a few hundred SKUs of a similar type, “Nicki Minaj mustache anti-grow cream”, you might have replaced one bad item for another. Maybe having hairy lips is fashionable, and there simply isn’t a strong market for that item. Or maybe, Taylor Swift has her own “anti-bearded lady lip cream” selling for an absurdly low price that you can’t compete with.
You should cut out ranges which aren’t selling and replace them with items which are. Take your best sellers and wrap new lines around them. Picking items which will sell well on Amazon is becoming increasingly competitive. Take more time with your sourcing analysis. We did an article on this a few months back actually if it helps.
If your top seller is a pair of socks, sell some sock racks, sock hangers and tracking devices for lost socks. Sell items which complement your item, not just because the residual sales might piggyback from the top seller, but because it’s where the data is leading you.
You mentioned selling a consistent amount of units per day. Which SKUs are these? Are they always the same SKUs? Or the same product type? Data is going to help you with any decision. Extract the last 12 months of sales and start cross-comparing the data. How many sold each day? At what time? To where? With what? Which were they? Who? Where? What? When? How?
Get every ounce of data outlined and see if there are patterns to it that you hadn’t noticed. It’s hard to see patterns when you are only seeing a small batch of 10-15 sales a day. But add all these up over time and I’m sure you’ll see a few patterns that weren’t clear beforehand.
Amazon was an inside job…
Amazon, in my humble opinion, is not the beast you need to tame right now. I do not believe Amazon is purposefully holding you back. I do not believe they are tracking sales and when you hit a certain amount, a light goes off on a dashboard and then a new algorithm pushes you down in search because you have sold “enough” for this month.
Amazon rewards the best items and the best sellers with the best traffic because it’s the best buyer experience. That is all Amazon cares about. They don’t care about you. Sorry, but it’s true. Amazon doesn’t care if you fail or succeed. Amazon only cares if buyers are happy and come back. If you sell a product buyers like, Amazon is making money every time you convert a buyer. Why would they hold your sales back when it’s making them money and keeping buyers happy?
But Amazon does balance the equation. All marketplaces will try to keep giving sellers chances to sell new product. Marketplaces cannot funnel all traffic only to items that sell, otherwise no one would ever find new product and new sellers would never have a chance. If you make mistakes or your items don’t sell well, chances are you won’t get much traffic. But new sellers and new listings are given chances to be found. It’s illogical for Amazon to take any action in the other direction, but I’m sure the algorithm will oscillate.
Based on the limited information you outlined, I may be talking nonsense with the below suggestions. Or, maybe I am talking nonsense always. Anyway, I think you need to focus on four things:
- Make some calculated changes, and keep doing it. Amazon PCC campaigns should be undergoing constant small changes, and you should be trying new bigger ideas monthly. This topic should have rolling changes, ideas and fresh perspectives.
- Analyze your sales, not just the top sellers. Track the mid-range performers. Find the ones which have consistency on a weekly, monthly or yearly basis. Capture all the details around these sales. Look for patterns. Those patterns might breed clear cut directions you need to take, or might require testing ideas.
- Check competitors and keep a pulse on the industries you are selling into. Keep tabs on the demographics, trends and changes which are happening externally. External changes can be critical to track, so you stay ahead of the curve. Or, they may highlight waves of opportunity that you want to ride, or risk being stuck at shore. Use this when considering new lines for your catalog.
- And here is a big one – diversify! Sell on other channels besides Amazon. Sell internationally. Develop multiple streams of sales and tap into any market where your item has an audience.
I am confident that if you examine your data in more detail, you’ll find some insights to use. If you start making calculated changes, and taking small ideas and making them live, you will gradually find opportunities.
Growth is about adaptation. Amazon has grown by constantly reinvesting in itself and trying new ideas. Grow with Amazon by growing like Amazon.
Have a question for us? Send it to firstname.lastname@example.org.