Plan · New websites
SEO content plan for a new website: what to do in the first 90 days
Short answer
For a new website, spend the first two weeks on foundations (Search Console, a clear structure, About and Contact pages), weeks three to six on one tight group of articles around a single topic, weeks seven to ten on supporting pages and internal links, and the last weeks on review. Aim for indexed pages, first impressions and a handful of real queries by day 90, not top rankings. Google gives no timeline: it says to wait a few weeks before judging, and that changes can take hours to months.
This is a documented plan, not a test. The mechanics come from Google's SEO Starter Guide and helpful-content guidance; the schedule and article counts are our suggestion, labelled as such. CheaperStack has not run this plan on a new site and publishes no ranking results for it. Where a figure is ours, it says so.
The 90-day plan at a glance
| Phase | Days | Goal | Output |
|---|---|---|---|
| 1. Foundations | 1 to 14 | Be findable and trustworthy | Search Console and Bing verified, sitemap, About, Contact, 3 to 5 cornerstone pages |
| 2. First cluster | 15 to 45 | Cover one topic properly | 1 pillar page and 6 to 10 supporting articles |
| 3. Expand and connect | 46 to 75 | Link, fill gaps, improve | 4 to 8 more articles, internal links, rewrites from real queries |
| 4. Review and decide | 76 to 90 | Judge, then choose | A written scorecard and a keep, change or stop decision |
The article counts are our suggestion for one person publishing roughly one to three pieces a week. They are not a Google threshold. Scale them to what you can review.
Phase 1 (days 1 to 14): foundations
- Verify the site in Google Search Console and Bing Webmaster Tools. Google recommends Search Console to monitor how your site performs, and its URL Inspection tool shows how Google sees a page. Setup is free; reports are in Search Console and Bing AI reports.
- Publish a sitemap and check robots.txt. Google calls a sitemap optional and says it mainly finds pages through links from pages it already knows, so also link your pages together and get a few real links in. Do not block your own site in robots.txt by accident.
- Create the trust pages first: About (who is behind the site and why they know the subject), Contact, and a disclosure page if you earn commission. These are what quality raters look at first; see the E-E-A-T guide.
- Choose a simple URL and category structure. Google recommends URLs with words that mean something to users. Pick 3 to 5 categories you can fill, not twenty.
- Set titles and descriptions properly. Unique per page, clear, accurate. A WordPress plugin helps; see WordPress SEO plugins.
- Add analytics so you can see engagement as well as search impressions.
How to choose topics when you have no data
A new site has no queries in Search Console yet, so start from what people ask, not from volume numbers.
- Write down 20 questions your best customer asks before they buy. Support emails and sales calls are better sources than tools.
- Search each one yourself. If the first page is full of large sites, that exact phrase is a long shot; a narrower version of the question may not be.
- Use free tools for direction, not precision. Google Trends shows direction of interest and Keyword Planner gives volume ranges; the options and their limits are in Best free SEO tools.
- One search intent per page. Two pages that answer the same question compete with each other. Our solo blogger guide covers research tools by budget.
- Prefer questions, comparisons and how-tos you can answer with first-hand experience. That is where a small site can add something others do not.
Phase 2 (days 15 to 45): one cluster, done properly
Pick one topic and cover it from several angles rather than touching ten topics lightly. Google's helpful-content questions reward original information, depth and clear sourcing, and warn against writing about many unrelated topics hoping some rank.
- Write the pillar first: a complete guide to the topic that links out to every supporting article you plan.
- Add 6 to 10 supporting articles, each answering one question the pillar raises, and link back to the pillar with descriptive anchor text.
- Add one thing per article that only you can add: a number from your work, a screenshot, a decision you made, a mistake. This is the non-commodity test from Google's AI guidance.
- Publish on a steady cadence you can keep. Ten reviewed articles beat thirty skimmed ones.
- Request indexing sparingly through URL Inspection for the most important pages, and otherwise let links and the sitemap work.
Phase 3 (days 46 to 75): expand, connect and improve
- Read your first real queries in Search Console. They tell you what people actually used to find you. Write a page for the ones you do not cover and sharpen the ones you do.
- Add internal links from every new page to two older ones, and from older pages to the new ones. Google uses links to find content and understand how pages relate.
- Rewrite the weakest page of the cluster before adding a new one. Do not change the date unless the page changed in substance; Google lists that as a freshness trick to avoid.
- Promote the pages that deserve it. Google says promotion leads to faster discovery. Share where your audience already is, and avoid anything that looks like manipulation.
Phase 4 (days 76 to 90): review and decide
At day 90, write the answers down. Judge it on leading indicators, because Google gives no promise on rankings.
| Measure | Where | A reasonable day-90 result |
|---|---|---|
| Pages indexed | Search Console, Pages report | Most of what you published, with the exceptions understood |
| Impressions | Search Console, Performance | Present and trending up on several pages |
| Real queries | Search Console, Performance | A handful of queries you did not expect |
| Engagement | Your analytics | Visitors reading, not bouncing on the first screen |
| Quality | Your own review | You would still publish every page |
These targets are our suggestion. Google publishes no benchmarks for a new site, and nothing here predicts rankings.
Then choose one of three: keep going if indexing and impressions are healthy, change something specific if pages are indexed but ignored, or stop a topic that produced nothing. Setting the stop rule before you start is covered in Autonomous blogging tools.
Planning for 2027
A site launched now will be judged in a search environment with AI answers on top. Google's own guide says generative AI features rely on its core ranking systems, so the plan above is the plan. Make pages crawlable and indexed, write non-commodity content, and measure in Search Console; the full list is in the AI search checklist. Nobody can promise what a 2027 update will change, so build on documented basics.
Disclosure: This post contains affiliate links. If you sign up through one and later make an eligible purchase, I may earn a commission at no extra cost to you.
Tools for each phase
If your site runs on WordPress, Rank Math's free plugin covers the foundations in phase 1: titles and descriptions, a sitemap, schema, redirects and a Search Console connection. Its MCP tools can run an audit and list weak pages in phases 3 and 4; the steps are in Rank Math MCP with Claude. Get Rank Math (affiliate link).
If writing is the bottleneck
Phase 2 is where most plans stall: a topic list exists and nobody has time to write it. GrowGanic is an autonomous SEO engine: add a domain and it researches the keywords, writes and scores the articles, publishes them to your site or a hosted blog, and tracks the rankings. On a paid plan it publishes every weekday, up to 30 articles a month on Pro. That is far more than this plan needs, so cap it to what you will review. It has no backlink index, and, by GrowGanic's own account, the hosted copy of the free article stays out of Google until the plan is paid (connecting your site or a subdomain first avoids that). The free plan writes your first article, which is a cheap way to judge the quality before deciding. Try the free plan (affiliate link). Costs and combinations are in AI SEO tools for WordPress.
Mistakes that waste the first 90 days
- Publishing 100 pages in a month. Google warns against many pages without added value.
- Ten topics, one article each. A single well-covered topic gives you something to build on.
- Judging at week three. Google says to wait a few weeks, and some effects take months.
- Skipping About and Contact. They are the first trust signals a reviewer looks for.
- Buying links or fake mentions. Google calls inauthentic mentions unhelpful, and link schemes are spam.
Who should not follow this plan as written
- Local businesses with a Google Business Profile should add that first; this plan is for content sites.
- Health, legal and finance sites need a qualified, named author on every page, which slows the cadence.
- Sites with a team and a budget can run phases in parallel, but the order still holds.
The verdict, by situation
| If you are… | Do this |
|---|---|
| Solo, no budget | Free stack: Search Console, free keyword tools, 1 to 2 articles a week |
| A small business with expertise but no time | Interview yourself or your team for phase 2, then edit |
| Short of writing hours | Draft with a tool, review everything, add first-hand detail |
| On WordPress | Plugin and Search Console in phase 1, MCP audit in phase 3 |
| Unsure whether it is working | Use the day-90 scorecard before changing anything |
Frequently asked questions
How many articles should a new website publish in its first 90 days?
As many as you can make genuinely useful and review before publishing. For one person that is often 12 to 24 well-chosen articles, a figure that is our suggestion rather than a Google rule. Google asks for original, helpful content and warns against producing many pages without added value.
How long does it take for a new website to rank?
Google gives no timeframe. Its starter guide says changes can take from hours to months to show, and suggests waiting a few weeks before judging whether your work helped. Plan around indexing and first impressions in 90 days, not top rankings.
Do I need a sitemap for a new site?
Google calls submitting a sitemap optional, and says it mainly discovers pages through links from pages it already knows. A sitemap is cheap insurance and most CMSes make one, but it will not replace links to and between your pages.
How do I find keywords for a site with no data?
Start from questions your customers ask, check what already ranks by searching them yourself, and use free tools such as Google Trends and Keyword Planner for direction. Search Console starts showing your real queries once the site has impressions.
Should a new site use AI to write its articles?
AI is allowed, but Google's guidance asks for accuracy checks and non-commodity content, and warns against many pages without added value. Use a tool to draft or research, read every article, and add something only you know.
Sources
Google Search Central's SEO Starter Guide, its helpful content guidance and its generative AI content guidance; plus Rank Math's setup and MCP documentation and GrowGanic's pricing page. The schedule, counts and day-90 targets are ours.
Related
Best free SEO tools
Twelve free tools, with the real caps.
How to get backlinks the legal way
Seven ways to earn links without paying for them.
Search Console and Bing AI reports
What the free reports show, and what they leave out.
AI SEO tools for WordPress in 2027
Plugin, assistant and content engine: three stacks by budget.
Small business SEO automation
What to automate, what to keep, and what it costs.
Last reviewed . Google's guidance changes, so this page is re-checked monthly. Spotted one out of date? Tell me.