We built a site from scratch, published aggressively, and tracked everything in Search Console. Here's what the data says about what actually moves rankings.
Everyone asks the same question about SEO: "How long does it take?"
The generic answer is 3-6 months. Which is technically true and practically useless. It's like asking how long it takes to get fit and being told "it depends." You already knew that. You wanted specifics.
So here are the specifics. We registered a brand new domain in mid-February 2026. We published aggressively. We tracked everything in Google Search Console from day one. And after 30 days, we have real numbers — not theory, not best practices from someone else's blog, not projections. Actual data from an actual new website.
Some of it's encouraging. Most of it's humbling. All of it's useful if you're about to start the same journey.
This is the companion piece to our 45 posts in 30 days breakdown. That post covers the production side — how we published that much content that fast. This one's about what happened next. The strategic lessons. The stuff we'd change.
Domain registered: mid-February 2026. Brand new. No existing authority. No backlinks. No history.
Within 3 weeks, we had roughly 100 URLs live on the site. That included:
Every page was submitted to Google Search Console manually. Sitemap submitted. Internal linking built from day one — every new post linked to 2-3 related posts, and we went back to update older posts with links to newer ones.
We followed the AI SEO workflow for every piece of content. Keyword research, intent analysis, structured drafts, on-page optimization. Nothing was published without hitting the checklist.
In other words: we did everything "right." Let's see what happened.
Here's the raw Search Console data, week by week:
Week 1: 184 impressions. 4 clicks. Average position: 26.
Week 2: Impressions climbing. A few pages showing at positions 7-13. Clicks trickling in. Things looking promising.
Week 3: 758 impressions. 0 clicks. Average position: 70.
Read that again. Impressions went up 4x. Clicks went to zero. Average position nearly tripled — in the wrong direction.
If you'd only looked at impressions, you'd think we were winning. More pages indexed, more queries showing up. But the clicks told the real story. We were being shown to more people in positions where nobody scrolls.
What happened between week 1 and week 3 is the most important thing any new site owner needs to understand.
In weeks 1-2, several of our pages appeared at positions 7-13. Right there on page 1, mixed in with established sites. We hadn't earned those positions. Google was testing us.
This is what SEOs call the "Google sandbox" or the "honeymoon period." Google takes new pages, slots them into decent positions temporarily, and watches what happens. Do people click? Do they stay? Do they bounce immediately?
By week 4, most of our pages had dropped to positions 37-74. Some disappeared from the results entirely. The honeymoon was over.
Here's what we think happened: Google showed our pages at positions 7-8. Searchers had the choice between us — a brand nobody's heard of — and HubSpot, Indeed, or Neil Patel at positions 1-6. They clicked the known brands. Google measured that behaviour. And Google concluded: these new pages aren't what searchers want.
That's not a failure of our content quality. It's a failure of trust. A new domain doesn't have it. You can't shortcut it. And understanding that changes everything about how you should approach SEO in the first 6 months.
Not everything dropped. Two things held their positions across all four weeks, and the pattern is instructive.
Our branded queries held. The homepage sat at position 4 for "Syxo" the entire time. This makes sense — nobody else is competing for our brand name. If someone searches for you specifically, Google will show you.
One specific keyword held: "keyword research in 90 minutes." Average position 4.8 across all four weeks. Didn't budge. That page from our AI keyword research workflow just sat there, quietly ranking.
Why did that one survive when everything else dropped? Three reasons:
That third point matters more than people realise. Google doesn't just measure whether people click. It measures what they do after they click. If someone lands on your page and hits back within 5 seconds, that's a signal. If they stay for 4 minutes and scroll to the bottom, that's a different signal entirely.
Some of our pages had a brief moment of glory and then vanished completely. The pattern here is just as clear.
"AI skills for marketers" — appeared at position 7 in week 1. Gone by week 3. This keyword is dominated by Indeed (job listings), HubSpot (career content), and LinkedIn Learning. We were a new site trying to compete against platforms with millions of backlinks and decades of authority. Google gave us a shot. The data said no.
"AI prompts vs AI systems" — position 8 in week 1. Gone by week 3. Same story. The concept was good — it's actually core to our positioning — but the keyword space is owned by enterprise content sites with 10+ years of domain authority.
Most of our trade pages — "AI marketing for dentists," "AI marketing for personal trainers," etc. These showed brief impressions and then disappeared. The problem wasn't the content. The problem was the competitive set. We were an education platform competing against marketing agencies who've spent years building authority in those specific verticals.
The lesson is uncomfortable but clear: it doesn't matter how good your content is if your domain can't compete in that keyword space yet. Good content is necessary. It's not sufficient.
Rankings and authority are a chicken-and-egg problem, and nobody talks about it honestly enough.
You need authority to rank. You need rankings to get traffic. You need traffic to earn natural backlinks. You need backlinks to build authority. The loop is closed.
For a new site, there's no way to skip this. There's no AI tool that builds domain authority. There's no publishing cadence that replaces backlinks. The only way to break the loop is to create content that earns links without needing rankings first.
That means:
None of this is glamorous. None of it is fast. But it's the actual work. Anyone telling you they grew a new site's authority without doing some version of this is either lying or sitting on a domain they bought with existing backlinks.
SEO is one of five marketing systems. Find out which one is holding you back.
Take the Free QuizKnowing what we know now, here's what we'd change if we were starting over tomorrow.
We thought volume would win. It didn't. Most of our posts were 1,200-1,500 words hitting keywords that required 3,000+ word comprehensive guides to compete. We spread ourselves thin. Ten posts at 3,000 words each, with original examples, data, and screenshots, would've outperformed 45 posts that were good but not remarkable.
The maths is simple: one page ranking at position 3 for a 1,000-search-volume keyword gets more traffic than fifty pages ranking at position 47.
We targeted a lot of "ChatGPT marketing workflow" and "AI content system" type keywords. These are tool-focused queries. The problem? People searching these already know the tools. They're comparing options. And they're comparing us against established review sites and tool directories.
Better approach: target the problem upstream. People search "how to get more traffic to my website" before they search "ChatGPT SEO workflow." Problem keywords have higher intent, less competition from tool-comparison sites, and more room for a new brand to provide a genuinely different answer.
We did this backwards. We wrote 45 keyword-targeted posts first, then started thinking about backlinks. Should've been the other way around.
If we'd started with 5 data-driven, build-in-public posts — the kind journalists and bloggers actually link to — we'd have had some domain authority before publishing the keyword-targeted content. Those keyword posts would've started from a stronger position instead of competing with zero backlinks.
"AI marketing for dentists" is a fine topic. But we're an education platform, not a dental marketing agency. The sites ranking for those keywords are agencies with 50+ dental client case studies, testimonials, and years of niche authority. We brought a blog post to a gunfight.
For a new site, the competitive set matters as much as the keyword difficulty score. A keyword might show difficulty 25 in Ubersuggest, but if every result on page 1 is a niche specialist, you're not competing against difficulty 25 — you're competing against relevance and authority you don't have yet.
Our one ranking success — "keyword research in 90 minutes" — worked because of the constraint. The "90 minutes" part. HubSpot doesn't promise 90 minutes. Ahrefs doesn't promise 90 minutes. The constraint created a niche within the niche.
Every post title should have something like this. A specific time frame. A specific budget. A specific audience. A specific limitation. Something that makes the big sites' generic content not quite match the query.
"AI SEO workflow" — too broad. "AI SEO workflow for solopreneurs under $0/month" — that's a lane.
We've adjusted the strategy based on everything above. Here's the current playbook:
Moat content first. Posts like this one — real data, original analysis, build-in-public transparency. Content that earns links because nobody else has the data. We're aiming for 2-3 of these per month.
Problem keyword posts with Syxo-specific angles. Instead of "AI marketing tools" (competing against every review site on the internet), we're writing about specific problems we've solved with specific approaches. The angle is always something a big site can't easily replicate.
Active backlink building. HARO responses every week. Relevant directory submissions. Reaching out to bloggers who write about AI marketing. None of this scales beautifully, but it compounds. And compounding is the only game that works in SEO.
Fewer posts, more depth. We're down from 45 posts/month to about 8-10. Each one is longer, more detailed, and includes original examples or data. The AI SEO System we built helps with the research and drafting, but the depth and originality still come from actual experience.
Updating existing content. Half of our 45 original posts are worth saving. We're going back, adding depth, improving examples, strengthening internal links. Google notices when you update content. It often triggers a re-crawl and can improve positions.
There is no SEO shortcut. Not even with AI.
AI makes content production fast. We proved that — 45 posts in 30 days, every one following a structured SEO workflow. The production problem is solved. You can write and publish a fully optimised blog post in 90 minutes.
But AI doesn't make domain authority fast. It doesn't make backlinks fast. It doesn't make Google trust you faster. Those things take months of consistent, quality publishing plus active authority building.
Here's what that means practically: the solopreneurs who succeed at SEO aren't the ones who publish the most. They're the ones who publish the right things and are patient enough to wait 6 months for the compound effect to kick in.
Month 1 will be frustrating. Your impressions will climb but your clicks won't. You'll watch your pages appear and disappear from the results. You'll see competitors with worse content outranking you because they've been around longer.
Month 3, if you've been publishing the right content and building authority, you'll start seeing positions stabilise. A few pages will climb to page 2. Maybe one or two crack page 1 for long-tail keywords.
Month 6 is where it gets interesting. The compound effect of 30-50 well-targeted posts, each with proper internal linking, starts to build. Your domain authority ticks up. New posts start ranking faster because Google already trusts your site on those topics.
Month 12, you have an organic traffic engine. But only if you spent months 1-6 doing the boring, unglamorous work that nobody wants to talk about.
We're on month 1. We've made the mistakes. We've seen the data. And we're adjusting — not by publishing faster, but by publishing smarter.
We'll keep sharing the numbers as they come in. That's the point of building in public. The wins are easy to share. The failures are where the actual lessons are.
Most new websites start seeing impressions in Google within 1-2 weeks of submitting to Search Console. But impressions aren't rankings. Realistic timelines: weeks 1-2 you'll appear at positions 40-100, months 1-3 you may climb to positions 15-40 for low-competition keywords, and months 3-6 is when page 1 rankings become possible for well-targeted content. High-competition keywords can take 12+ months regardless of content quality. The biggest factor isn't time — it's whether you're targeting keywords your site can actually win.
Google has never officially confirmed a "sandbox" for new websites. But the pattern is well-documented: new pages often appear at decent positions (page 1-2) for a few days, then drop to page 4-8 before slowly climbing back. Our data showed exactly this — pages ranking at positions 7-13 in week 1, then dropping to positions 37-74 by week 4. Whether you call it a sandbox, a testing period, or an evaluation phase, the effect is real and new site owners should expect it.
For very low-competition keywords with few competing pages, yes — you can rank without backlinks. But for anything moderately competitive, backlinks remain one of the strongest ranking signals. Our experience showed that content quality alone wasn't enough to hold positions against sites with established domain authority. The practical answer: you can start ranking for long-tail keywords without backlinks, but you'll need them to compete for anything with real search volume.
Click-through rates depend almost entirely on your average position. At positions 1-3, expect 3-10% CTR. At positions 4-10, expect 1-3%. Below position 10 (page 2+), CTR drops below 1%. In our first 30 days, we had 758 impressions in week 3 but zero clicks — because our average position had dropped to 70. Impressions without clicks usually means your content is being shown but you're too far down the results page for anyone to see it.
Neither — publish based on quality thresholds, not schedules. Our biggest mistake was publishing 45 posts in 30 days. Most were surface-level and couldn't compete with established content on the same topics. A better approach: publish 2-4 high-quality posts per week, each targeting a specific low-competition keyword with genuine depth. One post that ranks on page 1 is worth more than twenty posts sitting on page 8.
SEO is one piece of the puzzle. Take the free quiz to find out which of your five marketing systems needs attention first.
Take the Free Quiz