Tuesday, November 1, 2016

Turnover Tuesdays - Past Performance is not a Guarantee of Future Results

For those who are not familiar, I started a series a while back called Turnover Tuesdays. Every Tuesday I like to highlight one item that I have resold. This will include profitable and non profitable sales. I hope that there is always something to learn.

The previous posts in the series can be found at the bottom of this post.

Using Data from Keepa to Your Advantage

The Keepa extension is probably the most important (and free!) extension available to Amazon sellers.  I use it to evaluate every single deal.  Rank is almost meaningless to me at this point.  Rank is a snapshot currently but sales significantly affect the rank.  If you used rank to evaluate purchases on Prime Day and Amazon put it on sale, every item would have been #1 rank in its category so rank can be very misleading.  You need to evaluate rank over time.  You need to see the number of sellers over time.

There is a ton to be gleaned from Keepa, some of it obvious and some of it not so obvious but it is absolutely essential.

One thing you can use Keepa for is to check out prices last year during busy seasons previously.  This can be a guide for you of what items have huge demand.  As I will show, this does not necessarily correlate to a big payout.

Halloween Lights from Keepa

I was looking at the Halloween lights section at a local store and stumbled on something that was close to profitable already and it has was still well before Halloween.  I bought a few and then looked at the Keepa chart which looked like this.

You can see that last year the price hovered around $40-$50 (Blue lines on the top chart) around Halloween time and the full price cost was about $20.  I got really excited and went big.  I bought 40 from one store and 40 from another store.

Well, here is what the current Keepa chart looks like:

Price went down to about $20 in August and never recovered much.  They went up to almost $25ish at the high but not higher.  

What Happened?

Did demand go down?  Most definitely not.  If you look at the rank (green line on the top chart,) rank went down like crazy as Halloween approached.  I can attest to the fact that it was selling extremely fast.

2 things happened which are related.  One is obvious and one is not.

The first problem was the number of sellers.   If you look at the number of sellers (bottom chart on Keepa), there was a maximum of 24 sellers at any given point last year.  This year that number was 71.  That's almost 3x as many sellers.  Demand may have even increased significantly but the supply was more than able to keep up with the demand.

Why did that happen?  Why did people buy even though it wasn't profitable?  I think the answer is Keepa and people like me.  People saw the chart from last year and got excited hoping for the same increase which never came.  Another possibility is that there are just more sellers now than their used to be even last year but there aren't 3x the number of sellers.

The second problem was inventory which you won't see from Keepa.  Keepa will tell you how many sellers but it won't tell you how many total units they have.  Hopefully that is coming.

When I checked inventory levels there were 3 sellers with more than 100 units.  One had over 800 FBA fro about $24 and another had a few hundred units FBM at around $20.  These sellers kept the market down just by themselves.  Eventually they sold out but it was so late in the game that there wasn't enough time to recover.  

How Did I Do?  Price Matters, a lot

I mentioned that I bought 40 from one store and 40 from another store.   I didn't buy them for the same price.  At the first store I paid $20, full price.  At the other store, there was a Halloween special going on so I got them for $14.94.

I pretty much sold all of them in the range of $24.74-$24.88.  My profit or lack there of was directly correlated to my purchase price.  When I paid $20, I lost about 15%.  When I paid close to $15 I gained about 15%.  Obviously that is a far cry from the 100%+ I had been hoping for.

According to Inventory Lab I pretty much broke even but that doesn't take into account shipping fees so I lost a little bit.  I guess that was my Manufactured Spending for the month (at lest until the post Halloween return come).

The takeaway for me is that Keepa is a great guide to predict demand, but because it is available for everyone you have to treat previous price increases very cautiously.  

Week 64 - The Buy Box
Week 65 - Amazon Restrictions and the Future of Selling on Amazon
Week 66 - Fun with Inventory Reimbursements
Week 67 - Q4 Storage Fees
Week 68 - Start Your Own Listings
Week 69 - A Long Tail Sale and Calculating Storage Fees
Week 70  - Prices Always Come Back Except When They Don't