Why I Created Good, Cheap and Fast
I built Good, Cheap and Fast to save people time.
It took a personal crisis for me to realize how much time I was spending on things that I didn't care about (and how little time I was spending on things that I did). Each year, I'd spend one month commuting on the subway; two months in meetings; four months asleep. Time is precious! And shopping began to feel like a time-sucking chore.
Fake Reviews Are Everywhere
More than 50% of products on Amazon are sold by third-parties. Sites like AliExpress make it easy for people to order cheap products from China, slap their logo on them, and sell these products at a huge markup. (I've seen $1.25 TV antennas being resold for $50!) The Amazon Marketplace has become a dog-eat-dog world where sellers are willing to do anything to get their product to the top because the rewards are huge.
% of Amazon Sales by Third-Parties
Today, somewhere between 30-50% of product reviews on Amazon and Walmart are unreliable. Fakespot and ReviewMeta are two wonderful tools that detect suspicious reviews and I use them as a first line of defense. But, the war on fake reviews is evolving; relying solely on algorithms can lead to collateral damage.
I Value Human Learning Over Machine Learning
Artificial intelligence makes mistakes, too. And because so many algorithms are a "black box," important knowledge is kept secret and proprietary (or critical mistakes aren't explained).
Would you buy a jump starter that isn't capable of jumping a car? Thousands of Amazon shoppers have, and worse yet, many of those folks left 5-star reviews for the product. Why? Because they've haven't needed a jump start yet. They use the jumper as a portable battery to charge their phones.
Real people. Real reviews. Bad product (that algorithms won't detect).
Don't get me wrong, technology is great -- but, I don't want to live in a world where machines make every decision in secret for us.
Product Testing Isn't Always the Solution
Good, Cheap and Fast isn't meant to be a substitute for lab-tested product reviews, but I hope it's a complement. I'm a fan of Cook's Illustrated and Consumer Reports. These publications buy products, test the heck out of them and cover their costs by charging a subscription. This is the best business model for protecting readers.
Plenty of free websites test products, but readers "pay" in other ways:
Ad Bloat - The websites have great content, but it's hard to enjoy it because of excessive advertisements, autoplay videos (with sound) and third-party networks that hijack your browser and redirect you to pop-up ads claiming that you’ve won a free gift card.
Price Creep - The websites are great at reviewing products, but in order to cover costs, the tested products are more expensive than necessary. Many publishers do this (unwittingly) because the affiliate commissions that they receive are based on a percentage of a product's sale price.
These phenomena are common on legitimate websites. Nevermind the sites that: rent studio space to stage their "lab testing;" don't know how to use lab equipment properly; fail to disclose pay-to-play relationships; give inflated reviews to maintain relationships with PR people; or, fail to investigate the issues that products have after extended use.
I've worked in publishing for most of my career and I’ve seen these trespasses: Hucksters are everywhere.
Data Science Is Great, But One Rule Does Not Fit All
I've spotted useful trends, patterns and outliers that are true in one dataset, but not another.
For example, many eyebrow razors sold on Amazon are counterfeit or of questionable quality. These suspicious products have one thing in common: They are not returnable. But, tempting as it might be to say that all "non returnable" products are bogus, it just isn't true. 3M's Ultrathon insect repellent is one of the best bug sprays available, but you can't return it because it contains materials that are not allowed on airplanes.
Exceptions like these can be integrated into an algorithm, but I'm not sure they should be. I once found a promising way to filter reviews, but upon reflection, I saw how an innocent input might have a potentially-racist output. Dangerous.
A Simple Q&A (So I Don't Prattle On Forever)
What does this website do?
I flag products with suspicious reviews, cull the ones that seem to have quality control issues, and toss out products with above average prices. Whichever products are left over have above-average reviews and below-average prices.
How does this website make money?
This site is part of Amazon's affiliate program. I might earn a referral fee (at no expense to you) when you click a link on this website that takes you to Amazon.com.
Does this site collect personally identifiable information?
Why don't you list products on websites other than Amazon?
I don't have the resources to do so. This website is a passion project. I hope that the site can develop into a business that will allow me to improve and expand it, but until then, I need to keep it simple.
What's the deal with all the pictures? Isn't this supposed to be a text-only website?
This website originally had no images. I decided to add images for two reasons:
1. This site received press coverage for being text-only, but not many users.
2. I found a workaround that allows me to display large images that are only around 40kb in size. That's small enough to not intrude on your mobile data plan. (I hate when websites load huge multimedia files over a mobile connection.)