[weisbaum]

Taking a look at your site, it seems quite clear where all your traffic went. You have a pretty great backlink profile compared to other affiliate sites ive seen but the content is incredibly thin. Based on this, me or anyone else who visits the site just sees an advertising site. There are no personal experiences with the products, the products are not related and the UX does not make the user want to continue exploring the page they landed on. The content here seems more like a statistics site for products than it is a reviews site.

The other comments here mentioning cheap as a flag word are somewhat accurate but not entirely. Its possible to rank a site using the word cheap assuming you're supporting it correctly with related content about a specific product and proper explanation as to what makes it cheap.

Honestly, I am very sorry that you lost all your traffic and your revenue but this seems like a VERY fixable problem considering your backlinks. Take your best performing posts, and dramatically increase the word count, add more photos, add overviews of each product instead of just a single line or two pulled from reviews.

Happy to help with additional pointers if you want to shoot me a PM.

[alvah]

This. Google has had it in for affiliate sites for as long as I can remember, and they specifically dislike thin content. While the site has ~250 words of value editorial at the top of the pages I've looked at, the descriptions of the products themselves are extremely thin (~30-50 words), and would likely fall foul of this Google policy: https://developers.google.com/search/docs/advanced/guideline...

"Pages of product affiliation where the majority of the site is made for affiliation and contains a limited amount of original content or added value for users."

Also, the site may not comply with this policy: https://developers.google.com/search/docs/advanced/guideline...

"Sites that copy content from other sites, modify it slightly (for example, by substituting synonyms or using automated techniques), and republish it"

While I have seen much poorer quality affiliate sites than this ranking in Google, IMO this model is on borrowed time. I think the way to survive in this market is by increasing the value:affiliate ratio, and diversifying income sources, so affiliate revenue isn't the sole focus.

[Syzygies]

Agreed. I used to loathe the experience of trying to Google an Amazon alternative, and the first fifty links would be Amazon affiliates. I'd pay Google a subscription to simply block all affiliate links. I'd call these changes in Google's algorithms progress.

[visarga]

This could be handled locally, like uBlock. I'd like to have such a browser extension to re-filter and re-rank the search results, where I could simply ban/boost a keyword or a topic. The same extension could be applied to FB and Twitter as well - they too generate feeds that could be re-filtered and re-ranked.

[johnwdefeo]

Fair points. I don't scrape/spin any content, and I would argue that it's easier to write 500 words than 50, but I can see how my content might run afoul of an automated system. Plenty of the comments on this thread make similar points.

Ironically, I know of a scraper site that mirrors Amazon's bestsellers list, reformatted as blog posts: https://gistgear.com - and it's growing like gangbusters.

[g051051]

> it's growing like gangbusters

For now...

[germanjoey]

Is this the reason why top recipe results always contain the poster's entire life story before they ever show a list of ingredients?

[graeme]

That plus you can’t copyright a recipe but can copyright a story. So it thwarts scrapers.

[ihaveajob]

This. I keep telling my wife how much I hate recipe websites. There has to be a better way! And yet, you're probably right that this is the reason. It's insane how much they have favored this anti-pattern.

[tunesmith]

I agree - I hadn't heard of the site, but after reading the OP's blog post I was excited to check it out. However, after finding a bunch of categories I was interested in and opening them in a series of tabs, I started to review them and was disappointed. It just felt like a long list of products, and I felt like it was putting more work on my shoulders, not less.

The site needs more focus. Make some actual recommendations and tell us why we should trust them. I think wirecutter has a good formula for this. I've bought several products after reading wirecutter articles.

I don't know if this has anything to do with google's algorithm, but it did actually feel to me that the site was coming close to the line between "here is some helpful information for you" versus "here are a lot of extra links that will help me get a little bit more revenue".

[santoshalper]

I love and use wirecutter a lot, but I really enjoyed this site. He does not try to go the route of having an expert personally review every category (which is a great, high credibility approach). Instead he seems to just do the math finding the best rated items/average price. I find this more useful than another inferior wirecutter clone because it adds a data point.

It's definitely something I'll keep in my toolbox.

[XCSme]

How is the list better than going to a shop and filtering by reviews/price?

[johnwdefeo]

Thanks! My intention was to complement longform content rather than going head on with it.

[johnwdefeo]

Interesting. I've struggled with the balance of explaining my process vs. getting out of the way of the recommendations. And for that matter, how many products is the right amount to feature. The earliest iteration of the site only listed 3-5 products per category, which may have been an inadvertent sweet spot. I had hoped that my awards program would help people to learn more about my selection process: https://www.goodcheapandfast.com/articles/value-award-winner...

[s17n]

I actually really like the concept, as I understand it - a vetted shortlist of products worth buying in each category, with customer reviews distilled down to a brief summary (instead of a game-able star rating). Unfortunately I can definitely see how Google wouldn't be able to distinguish this from low quality affiliate spam. Also, I agree with "the UX does not make the user want to continue exploring the page they landed on" - if I were you I'd invest in improving the design.

As for "dramatically increase the word count, add more photos, add overviews of each product instead of just a single line or two pulled from reviews"... don't do this! If I wanted all that junk I'd just read the Amazon page. I agree that there is probably room to add a bit more information about the products without ruining what makes your site cool but don't abandon your vision.

[creato]

Do you really think this is vetted? There's no way this guy has actually tested all of these things. He's just reading Amazon reviews instead of you.

[s17n]

Exactly, he's reading Amazon reviews instead of me, that's the whole point. (If you don't want to call that "vetting", fine by me I guess)

[santoshalper]

No, specifically I think he does not do any manual vetting or reviewing. I think he essentially gets a ton of products in a given category, then takes the average review and price and runs some kind of formula to find a value sweet spot. I've done this same thing dozens of times myself in excel and it's kind of cool to see someone do it systemically.

Combine this with something like wirecutter that does vetted reviews (but probably looks at fewer products as a result) and it seems like a good complement.

[johnwdefeo]

The manual vetting that I do revolves around discarding real-yet-unhelpful reviews (which aren't easy to spot algorithmically). E.g.

Off-Label Usage - Customers rate a portable jump starter 5-stars, even though they have only used the product to charge their smartphones, not to jump start a vehicle.

Self Validation - Customers rate a carbon monoxide detector 5-stars because they feel a sense of relief and validation that their purchase will protect their families.

Customer Service Uprating - A 1-star rating is later updated to 4- or 5-stars because the manufacturer offers the customer a replacement product (and suggests altering the review).

Misunderstanding - A customer leaves a negative review because he or she didn't read the product description carefully and is consequently disappointed with the product.

Ideology or Spite - A positive review is paired with a negative rating because the customer disagrees with the business practices of the manufacturer (e.g. It's a great product, but Widget Corp. is a POLLUTER!).

Wrong Model - A review for one variation of a product is lumped in with reviews of another version of the product. (Hard drive failure rates can differ by 900% depending on the size of the drive.)

Wrong Product - A product page is repurposed by a seller, thereby mixing the reviews of one product with a completely different one. E.g. A page about a protective phone case contains reviews about a wireless charger.

Shipping Issues - Customers leave negative reviews because their packages arrived late or damaged in a way that reflects negatively on the shipping carrier, not the manufacturer.

Joke Reviews - A customer uses his or her review as a platform for comedy. Sexual wellness products, or those that are gender-based, seem to be disproportionately affected.

Empathy or Pity - A customer has a bad experience with a product, yet he or she leaves a positive rating (typically, 4-stars) because "someone" might like the product.

[s17n]

I think the key difference between this and the Wirecutter is less that the Wirecutter looks at fewer products (although that might be true) and more that the Wirecutter model is for an expert to form their own judgements about what is important whereas this guy focuses purely on deriving a consensus from the user reviews without using any priors or first party research.

[johnwdefeo]

That's exactly right.

[johnwdefeo]

I've never claimed to test products, but I do consider my selections to be highly-vetted: https://www.goodcheapandfast.com/about - There is around 2,000 hours of research behind them. The trick (from a UX standpoint) is deciding how much work to show when the spirit of the site is to be dead simple.

[hcurtiss]

And then actively qualifying any recommendation by adding “cheap.”

[weisbaum]

You guys are missing the point. Sure, he has a vision and he doesn't want to abandon it but in order to compete in ranking for super high competitive terms his site has to be on par with the other top ranking sites. Google doesn't know what his intent is and as the site is right now, the algo compares what he has to what others have. He has great backlinks, but really sub par content when you look at who he is competing against. He does not seem like an authority on any of this stuff.

If the concept is to attract the 'satisficers' which by the way, i love that word and that book also, there needs to be some other kind of optimization on the pages. Going after 'best of' for everything looks spammy. And that spammyness is amplified by the lack of content when compared to the other sites ranking for the terms he's optimizing for.

There are other ways to attack this problem. Instead of having it be best of everything - it really should be a statistics site for popular products. Add in some kind of ranking system or some kind of kayak-like search and BOOM - new unique concept that provides real value to people. As the site is right now, its gimmicky, thin on content and is NOT going to rank in today's SERPs the way it is now. Sorry.

[johnwdefeo]

Thank you for the thoughtful reply.

I agree that the spirit of my site is closer to "a statistics site for products" than a typical review site. Adding more content, photos, etc. may help to remedy my situation, but I would feel like I was abandoning my original purpose. I designed the site with "satisficers" in mind, i.e. allow people to make a decent decision as quickly as possible. No frills.

Until recently, I worked for product review sites like Anandtech and Tom's Hardware, and I'm not equipped to compete with them head on. My hope was to create a new niche by using data analysis, which is more of my skillset: https://www.johnwdefeo.com/articles/amazon-review-analysis

[weisbaum]

If that is the case than perhaps organic traffic is not the best vector by which your site will grow.

I could make the case that targeted paid search for low comp, buy ready keywords would work well. Your content is what would push someone over the edge to buy. If you think about it, your site is much further down the funnel than other affil sites because you've vetted reviews, instead of just copy/pasting/spinning from other sites. You should also retarget the people who come to those pages with ads specific to the products they're looking for urging them to "not make a decision without all the facts" which is what you've done by mining all of the review data.

I do very much like the approach and the original idea but sadly google likes what google likes. If deeper authoritative content is not what you want to be doing, that's fine, but you shouldn't have the expectation you are going to be able to compete with the big dogs who operate in that way.

Im happy to take this convo offline as you're getting a lot of weird advice in this thread that are just random people's opinions. Shoot me an email at [e-mail] if you want to keep chatting.

[captainmuon]

I wouldn't say the content is incredibly slim. If you check a few categories, he writes a introductory post and a short summary "review" for each product:

https://www.goodcheapandfast.com/articles/best-noise-cancell...

https://www.goodcheapandfast.com/articles/best-gaming-keyboa...

I know Google says they are all about the algorithm, but I'm convinced they are relying heavy on manual review farms. Of course admitting it would damage their brand, and disincentivize people from coding "to their algorithm". But I'm pretty sure a manual reviewer thought "yep, spam site" and penalized it.

One other thing, the landing page is basically a search engine / portal. In my experience, google also hits everything in this category. (I built a small search engine for a specific vertical in the past - no longer online - and it too showed strange search behavior after a while.) Maybe replace the landing page with a blog / magazine like feed?

[ryan29]

I think this is right. I couldn't quite put my finger on it, but the site felt a bit "meh" to me even though I like the concept. I like the simple design, but if there were ever a case to be made for a floating header that functions like a menu bar, those categories is it, right?

I skimmed a couple product categories that I've looked at recently, mainly gaming keyboards. The information felt a bit mediocre, but I also thought it did a pretty good job of picking out decent value products. I can see the value in it as a short list of products to look into, but there's not enough info for me to make a choice without leaving the site.

[Abishek_Muthian]

>add more photos

How much does image actually weigh in the site's score? What about the text only websites e.g. PG's essays, not including forums like HN. I would presume PG's blog has great backlink score but would a similar site with same number of backlinks but with several images rank more?(hoping for clarification from an anonymous Googler).

I'm asking this because my text only blog content titled 'Startup ideas vs Problems'[1] was the featured snippet when the search term is 'Startup ideas vs Problems'[2] until recently as the current featured snippet features another website with a featured image within the blog post.

Although my blog is still the first result, but google now suggests 'Did you mean: Startup ideas and Problems' which buries my blog post effectively killing my traffic.

[1]https://hitstartup.com/startup-ideas-vs-problems/

[2]https://www.google.com/search?hl=en&q=Startup%20ideas%20vs%2...

[londons_explore]

All modern search engines use neural networks for final ranking. Even the Google employees who work in it wouldn't be able to tell you exactly how much images help or hinder your rankings.

[Abishek_Muthian]

Yes, I just wanted to know whether article with more images for SEO is a common knowledge now as the parent comment suggests.

[weisbaum]

More images isnt directly correlated with rankings but it can help increase time on site, improve user experience and lower bounce rate - all which are ranking factors.

There is no one silver bullet for SEO. Its many things combined.

[praestigiare]

Yup: Make the site worse, fill it with fluff that Google wants, hide the actual useful thing you are doing.

[johnofthesea]

>increase the word count, add more photos

It seems that thanks to this policy of Google most cooking/recipe sites are full of fluff. Not always 'more content' means the content.

[johnwdefeo]

Yes! It would be easier to write 500 word blurbs than a meaningful 50 words. I despise recipe sites for this reason. So much "content" gets in the way of the actual content. The recipes in my favorite cookbook are around 100 words on average, yet if they were published online, I doubt they would rank.

[jfengel]

If they published the whole cookbook, or at least chapters, it would avoid looking like the minimum effort applied by content farmers.

I wonder if the same thing would work for you: organize your site with more content per page. You could add value with lists organized by some principle, or just borrow Amazon's catalog tree.

For that matter, would the same notion work for recipe sites? In addition to the main recipe per page, add several other related dishes -- compatible side dishes, similar main dishes, cocktails, etc. Reduce the blurb to please readers, but still produce content.

I'm far from an expert in SEO so I apologize if I'm telling you something you already know (or already know doesn't work). I'm just spitballing.

[love_anddd]

Exactly my thoughts looking at the site. It screams affiliate link farm where the only purpose is just to make money from product affiliate links.

While this isn't necessarily bad to make money, it definitely doesn't do anything for a visitor.

After reading the blog post, I was expecting something like consumer reports or some intricate posts, but it's obviously little tid bits of information that looks like it was posted by some virtual assistant he was paying $2 per post or something.

[jsdalton]

The person who wrote the article is named John DeFeo, while the username of the HN submitter is josephjrobinson. While it is possible they are the same person, it seems unlikely.

This seems to be a common mistake on HN, and I point it out because you and others in this thread seem to have invested a fair amount of time in crafting a personal response to the author that may never actually be read by them.

[doe88]

Exactly. Objectively this is bad. It makes me think of sometimes Youtube review videos I encounter, where the author instead of reviewing themselves the products display images or other bits of videos about the products from elsewhere. I don't deny it must take time to make these pages and videos but objectively it has little to no real information value at all.

[INTPenis]

Are the things you know taught anywhere? Maybe people who run a business that is dependent on Google traffic should learn what you know through some SEO class?

[lbj]

This sort of thing is why I keep coming back here. Kudos!