The Accuracy of Amazon Product Research Tools

Richard Morgan
March 18, 2019

[et_bloom_inline optin_id=optin_3]

Greg Mercer of Jungle Scout is with us to talk through the accuracy of Amazon Product Research Tools and his approach to assessing this using real data in as transparent a manner as he can find so people can make their own judgements.

The accuracy of Amazon Product Research tools is the focus
Case study with spreadsheets and analysis to assess tools

The Case Study

Jungle Scout estimates the sales of any product on Amazon
First one to come out with it and the algorithms for estimating have changed over the years
The sales estimates are a key metric for users
Have invested in the data science around this with our internal data nerd team
How to prove the quality and accuracy of the algorithm
How to make a bullet-proof case study

Real Data against Tools

Got a bunch of Amazon Sellers to donate their data to JungleScout
Letting JS log into their Seller Central accounts of 50 Sellers with real data
Posted in FB groups and ads for people to join the sae study
Purchased all of the competitor tools and went into each person’s account and ran the tools against the data
Pretty hard to poke holes in this 80hrs of tools testing against the real sales data
Hired VAs to go through and run all the different tools
With the data we’re then able to show it to our customers
Had 50 Sellers in total share their data to assess the accuracy of amazon product research tools
Full breakdown of very detailed analysis
100% transparent as we’re 100% confident
An overall error percentage with Jungle Scout coming out on top

Data Results

Unicorn Smasher error percentage was 92% so about as useful as flipping a coin
Estimating 1,000 units a month would really be somewhere between 80 and 2,920
So a free tool with data that is so far out is hardly worth it
Jungle Scout (25%)
Viral Launch is 2nd (34.34%)
Amaze Owl 3rd (44.29%)
Helium 10 4th (46.44%)
A re-cut of the data will be coming out again in Spring 2019

Tools Don’t equal success

Having the tool does not guarantee success but gives better insights into what products to select
Helps understand the demand in the market and niche
Using Unicorn Smasher with such a high margin of error impacts your ability to be successful in that niche

The Spreadsheets that assess the accuracy of Amazon Product Research Tools

You are welcome to use the data in any way you like to make your own assessment of the accuracy of amazon product research tools
Started with 20 different graphs and analysis of the categories
Decided the overall errors percentage was the best metric to help identify this
Presented with PHD level analysis in the spreadsheets and also presented more for the layman as well
You can do your own analysis of this data
The median overall error percentage is the best metric we’ve come up with

Jungle Scout Improves Accuracy

Going from 25% to around 12% on Jungle Scout
So plus or minus 12% on the data
Good to see Viral Launch in 2nd as a new entrant
Nice to see competition and raising people’s games
Jungle Scout has 3 x Full Time Data Scientists and one part time with 3 x PHDs
About £1m per year just to run these systems for sales estimates
The bigger players can invest more in this

Jungle Scout Supplier Database released

Releasing the Supplier Database feature in Jungle Scout this week
Import data into the US released through the Freedom to Data info
To let you know who the best factories are for them
Helps to find new factories and sources
Anyone with a Jungle Scout subscription will get access
Find out which factories your competitor is using
Look for best rated product on Amazon
Figure out their legal entity
Search in the Supplier Database for the factories as you know they are high quality already