$200k and still no recovery from Google's algorithm

This story comes three years of rebuilding a news website in the shadow of relentless Google algorithm updates, from organic search to Google Discover and Google News. We tested everything. We spent hundreds of thousands of dollars and countless hours trying to crack what now feels like an unsolvable problem.

But what if the truth is harder to swallow? What if there were never a way to recover from Google’s Helpful Content (HCU) updates? What if the game was unwinnable from the start, and we’ve all just been flogging a dead horse?

Why write this? Few publishers are willing to share the real stuff because the SEO industry is notoriously secretive (and dodgy at times). TBH I’ve given all the f*cks I have to give, so consider this free advice you didn't have to pay $200,000 for.

In late 2023, one of Australia’s longest-running men’s lifestyle publications (my business — DMARGE) was essentially thrown out of Google. It was not penalised. It just gradually lost visibility. Traffic dropped from over eight million monthly uniques to three hundred thousand. There were no manual actions, no warnings, just the result of algorithm evolution.

We weren’t publishing spam or gaming the system. We were doing what we had always done: creating original content with a 10-person editorial team focused on relevance and tone, watches, cars, food, travel, and more.

We were setting the pace, no question. Our local competitors routinely lifted stories, angles, and ideas from us, reworked them, and claimed them as original. But we were always ahead. First never follows. Ever.

Like many others, we turned to the SEO world to make sense of it all. We followed every update from John Mueller, waited for clarification from Danny Sullivan, listened to Barry Adams telling publishers to cut harder, and absorbed Glenn Gabe’s detailed breakdowns of visibility drops.

BTW, big love to Barry Adams, he jumped on a call with us for free to chat it through. Top bloke. He said your site looks damn good to me, no reason for it to cop a penalty.

Everyone had theories and solutions. Most commentary only confirmed what we suspected. We were quietly pushed out, and technical SEO changes alone wouldn’t save us.

We also approached the Google News Initiative, a program supposedly built to support journalism. We found surface-level marketing and no meaningful help, support, or answers. Just more empty promises while independent publishers like us sank quietly into irrelevance. ‘Have you read our documentation?’ was the usual response from Google’s News representatives here in Australia. To their credit, they did say they were ‘flagging it with engineering’, which could mean anything, but we were appreciative as it gave us a glimmer of false hope.

The catalyst? Twofold. First, with the November 2021 Core Update, things started slipping. Then came Google’s so-called Helpful Content Update, which promised to reward content made by people, for people. In reality, it turned into a vague, punitive shake-up that hit small to mid-sized publishers the hardest.

So we tried to fix it. Not with tricks or shortcuts. But by going line by line through our twelve thousand article archive. We noindexed thin content. Deleted dead categories. Removed tags. Hired real experts. Rebuilt the editorial structure from scratch. And we spent thousands.

Over two and a half years (longer actually) and countless hours, we did everything we were supposed to do. It didn’t work. In fact, we lost even more traffic and continue to lose it today.

This is the reality no one talks about. Here is the full breakdown of what we did, and why following Google’s rules no longer guarantees survival.

Out of sheer desperation, we brought in SEO consultant Lily Ray, one of the few people consistently vocal about Google’s erratic treatment of publishers. We paid six hundred dollars (USD) for an hour. She was sharp, pragmatic, and cautious about drawing conclusions without seeing all the data. Here is what she told us:

Lily Ray’s Recommendations:

1. Do not delete categories. Demote them in navigation or move them to the footer or sitemap
2. Make categories more granular, not broader
3. Audit every URL using GA, GSC, backlinks, and traffic source data
4. Strengthen internal linking using Link Whisperer or InLinks
5. Add actual text to video-heavy pages
6. Submit each Discover-style section to Google Publisher Center separately
7. Remove or isolate NSFW content, which could be dragging down the entire domain
8. Consider testing a new subdomain just for Discover
9. If Discover shows signs of life on any topic, double down and publish two or three related posts immediately
10. You may be covering too many topics.

That last point somewhat contradicted her first recommendation, but it was still worth investigating. If GQ, Men’s Health or Esquire can cover everything, why can’t we?

She confirmed what we feared. We were not just caught in an update. We were probably soft-banned from Discover. No warning. No confirmation. But zero impressions for twelve months says enough. This also extended to Google News and organic search.

1. Purged what we assumed was thin or low quality, but probably wasn’t

We started with word count. Articles under two hundred words are not always low quality, but they often lack depth. Thousands were noindexed, converted to draft, or deleted. It was not about hitting a specific number. We just wanted to avoid anything Google might label unhelpful.

Remember that over 15 years, we have interviewed some awe-inspiring people—doctors, CEOs, athletes, you name it. It’s not like our publication was a content farm on the outskirts of Mumbai. We even had a print magazine, 100,000 email subscribers, and a few hundred thousand social followers.

2. Stripped embed heavy content

Next we tackled stories built around embedded media. TikToks, YouTube videos, tweets. About thirteen hundred of them. These stories often had one or two paragraphs of text and then someone else’s content. We removed the embeds, rewrote the copy, and rebuilt them as original articles.

3. Cut quote padded news or interviews

We looked at stories padded with quotes. Common practice in newsrooms, but risky when there is little original value. Articles based on Reddit threads, press releases, or celebrity statements were either rewritten or deleted. It didn’t matter that other publishers do it. We are not other publishers.

4. Fixed the basic editorial structure

Every surviving article was reviewed:

1. Internal links were added to strong-performing articles
2. External links were added to brands, research, or original sources
3. More than one image was included
4. Inline related reads were added to help signal topical relevance

It was manual. It was obsessive. It was slow. And ultimately? No visible impact.

5. Deleted every tag page

We deleted all tag pages across the site. Not noindexed. Deleted. They weren’t ranking. They weren’t being crawled. And they weren’t being used. The impact on traffic? Zero. This confirmed our long standing belief that tag pages were just leftover clutter from WordPress.

6. Tested E-E-A-T theories

We brought in real subject matter experts. Fashion stylists. Car journalists like Mike Sinclair. Watch experts like Jamie Weiss and Felix Scholz. Grooming specialists, hairdressers, GPs and plastic surgeons. We created bios. Cross-linked their profiles. Gave them proper credit. Interviewed them regularly and ensured their stories were high quality. Based on the guidelines, this should have helped. But it didn’t. The content performed the same.

7. Pruned dormant categories

Then we went further, too far, some would say, but desperate times call for desperate measures. And let’s be honest, it can’t get any worse at this point.

Despite Lily’s original advice and our own instincts, we completely deleted all content verticals, such as style, sport, grooming, food, and entertainment. Later, she suggested that we may be covering too many topics. All lifestyle websites are broad. That is the nature of the format.

Our thinking was based on the Google Search documentation leak that mentioned SiteFocus. We assumed being too broad was hurting us, so we narrowed our scope. The result? We lost even more traffic due to the disappearance of long-tail content. No recovery followed.

Personally, I was certain this would do the trick. Just slowly remove entire categories in hope the ghost in the machine was in one of those. It appears not.

8. Google Discover and News continued to reward garbage and not ‘quality’.

The most frustrating part? While we were cutting carefully written journalism, Google Discover was filled with spam. AI generated content. Clickbait. Content farms with fake author profiles. Image heavy junk with no editorial value. It completely undermined the promise of helpful content. It proved we were not even playing the right game.

9. Y M Y L & NSFW

We had a large health section focused on fitness and mental health for men. We were proud of it. Trainers and doctors contributed. But we were unsure if this was holding us back. So we deindexed and removed the entire health section. Two thousand articles. A real loss. Especially since men need more guidance in this area, not less.

NSFW is a pretty obvious one. However, we were clutching at straws and desperate, so we systematically removed every instance of swearing (like sh*t or cr*ap). Even words that could be slightly flagged as ‘inappropriate’ got rewritten. Thousands of articles were edited. Manually. We then wondered if fitness articles with shirtless guys working out could be an issue, so the entire fitness category was edited.

You can see just how far we were willing to take this. I was willing to kill years of work just to find the smoking gun.

10. Too many ads!

Let’s be clear, this is complete nonsense. Those vague video chats John Mueller does about whether ads impact rankings are meaningless. Glenn Gabe also mentioned this too.

If your site is loaded with pop-ups, autoplay videos, and a terrible user experience, you can still rank just fine. It doesn’t matter whether you have five or fifty per page. Just look at the Daily Mail or Daily Express. Those chaotic, ad-choked messes rank everywhere and print money doing it. We became so cautious that we all but removed ads from the site. Like wtf.

We were willing to throw away revenue to try and find a solution to the problem. So much so that today we earn about 3-%5% of what we once did on programmatic ads. Them’s the brakes when you build a house on quicksand.

11. Changed subdirectory structure in permalinks

We read and heard from Lily, John Mueller, and others that Google couldn’t properly identify our site’s content themes because we didn’t include categories in our URL structure.

i.e. www.dmarge.com/watches/rolex-daytona-buyers-guide rather than the old one parent structure www.dmarge.com/rolex-daytona-buyers-guide

This was a completely plausible theory, by the way, and one that was worth testing.

So we rebuilt the entire permalink setup to include subdirectories for every article — watches, cars, travel, business, and so on — to give Google a clearer content hierarchy. It was a big change that created thousands of redirects across the site. The result? Absolutely zero impact. This was 18 months ago too. No dice.

12. Competitive analysis

It won't surprise you, but my analysis of other websites was exhaustive. What was most interesting was the inconsistency between seemingly similar sites to ours and the difference in visibility across all Google channels.

This led me down the backlink path; however, nothing stood out. We had a 15-year backlink profile with Bloomberg, CNN, The Guardian, Business Insider, and Wikipedia linking to many of our stories. Sure, there was a fair share of sh*t links in there, but everyone has those. We had even tried the disavow path, but that proved to be useless.

My only conclusion here is maybe too much affiliate-focused content, which is why we removed every ‘Best…’ story we had (there was a lot btw — most products we purchased and reviewed in the studio. Again, trial and error. Mostly errors, and things that just didn’t make sense.

13. Cadence

One theory Barry Adams had, and we did, too, was the cadence of publishing news. The owners of Vice once said to me, "The more shit you throw at the wall, the sooner something will stick."

We tried holding back content and published eight stories daily, three days a week. Other weeks, we would only publish one per day, and all stories were original and long-form.

Generally speaking, it is a numbers game. Some things pop off, others do not. That said, when you’re making two fitths of fuck all from display ads, you can’t keep investing in quality journalism. You will eventually start cutting corners and eventually editorial altogether.

14. Lastly, technical SEO.

We genuinely believed technical SEO would be a game-changer. Every year, we invested around $60,000 to $70,000 into development, testing, and benchmarking against sites that were thriving. We scrutinised everything from permalink structures, server configurations, Core Web Vitals, whether using www made a difference, and dozens of other technical factors. We double-checked, triple-checked, ran audits, optimised crawlability, and made the site as fast and clean as possible. It should have been a textbook example of technical best practice. But once again, it made no difference.

SEO consultants (more than one, 3 or 4) — $25,000
Web Development (changing site, tech SEO) — $120,000
Content creation (Editing, rewriting, testing, etc) — $75,000 (a very conservative figure)

My time. Owner of a company’s hourly rate — Say $100 p/h? Over 3 years. 250 Hours over 12 months x 3 years.

Paying for advice was a drop in the ocean. What cost the most was my time, my staff’s time, and the investment in Web dev to make all the changes. My ex used to say, dogs like to dream about chasing cars, I like to dream about chasing traffic. Was true.

After two and a half years, thousands of hours, and more than two hundred thousand dollars, we reached a difficult but honest conclusion. Google does not operate with a single set of rules. And that is fine. There is no point crying foul. Hate the player…

We tried to fix things the right way. We did not take shortcuts. We followed the rules. Not just the public ones, but the ones implied through leaks and updates. We treated the site like a real publication and tried to regain Google’s trust.

But when we compare ourselves to others in the same space, including newer sites with weaker content published at scale, it is clear that the playing field is not even. Some of them dominate Discover and News. Some run headlines that no proper editor would approve. And yet, they continue to grow. We continue to shrink. We were not an authority even for categories like watches, which we had covered extensively since day one. The first DMARGE post was a watch story ffs. We visit Watches & Wonders every year and it has zero impact on our authority in this space.

The part that hurts the most is that it's taken the joy out of finding stories, the thrill of beating the news cycle and spotting something big the media missed—that lightning bolt moment that drives real journalists and publishers. We no longer chase it. We no longer have morning WIPs to brainstorm because what is the point if no one sees it? It’s better to publish on Instagram or in our newsletter now rather than as an article.

As of today, we have gone from a twelve-thousand-article site with fifteen years of authority to three thousand articles focused only on watches, cars, and business travel. I do not understand how all this effort and precision can lead to zero gain. It tells me the issue is not with us. It is with them. And it took three years and a lot of money to figure that out. Along the way, we also lost many full-time journalists.

My gut says one of Google’s many signals is wrong. Not for everyone. I think that because other competitors are in the same boat. The difference is that we did everything to try and fix it. Also, I think generally Google’s focus is now e-commerce and AI, as that makes money. Just look at GSC and GA4, they’ve evolved for shopping, not news.

If there is any value left in this process, it is in being honest. Maybe this post will save someone else thousands of hours trying to decode a rulebook that keeps changing.

I have spent fifteen years building a digital publishing company that people actually enjoy reading. I have never seen an industry move the goalposts so often and punish the people who try to play fair. Honestly, I don’t know how much longer I will keep doing this. TBH journalism is generally fucked, and will soon be thing of the past.

But if you are still reading, at least now you know. You are not alone. And if you ever find the golden ticket, please share it. Your peers deserve success too.

Reach out to me on LinkedIn if you want to know more. (If you're going to sell SEO services or ask me to join your ad network in exchange for the solution, then f*ck off.)