How Do I SEO? Start With What You Can Do Today
How do I SEO my own site without hiring anyone? Start with three time boxes. Today, fix the pages you already have: titles, openings, and the one page that almost ranks. This week, fix how the site is built so people and crawlers can find your best pages. This month, publish two pages aimed at real questions and earn a few links to them. Everything else - tools, dashboards, audits - is support for those three boxes, not a substitute for them.
You can run the first box in one afternoon with no tool subscription.
If you only read four lines
- Fix one page that already gets impressions.
- Answer the question in the first paragraph.
- Link three related pages into it.
- Publish one new page a month, no more.
This is the loop I still run on my own projects. It is boring, and it works.
How do I SEO? The short answer
Answer: take one page that earns impressions but few clicks, rewrite its title, answer its main question in the opening paragraph, add three internal links to it, and request indexing. Repeat that loop every week. It does more than any tool subscription, and it is the only loop that gets cheaper as you learn it.
The three boxes I actually use
SEO advice feels endless because it is written for every site at once. You have one site, a handful of pages that matter, and a calendar. These three boxes fit in it.
Today Box 1
About 90 minutes- List your top 20 pages by impressions
- Rewrite the three worst titles
- Add a plain answer to three openings
- Open the key page on your phone
This week Box 2
Two to three hours- Link your money page from five pages
- Compress the images that load first
- Make sure nothing important is three clicks deep
- Build a two-column tracking sheet
This month Box 3
Six to eight hours- Publish two pages for real questions
- Ask five sites for a mention
- Review which pages moved and which did not
- Delete or merge one dead page
Today: fix what you already have
The first box is the one people skip, because it feels like admin rather than strategy. It is the only box that can move numbers this month. Start in search console and sort pages by impressions, not clicks. The pages at the top of that list are the ones search engines already consider relevant. Your job is to make them worth choosing.
Rewrite the title, not the whole page
A title that leads with the phrase people search for and ends with a reason to click does most of the work. Keep it under 60 characters including your brand. Change three titles today, then wait two weeks before judging. Rewriting twenty at once makes it impossible to know which change did anything.
Put the answer in the opening paragraph
Most pages open with context: who we are, why this matters, what we will cover. A reader who searched a question wants the answer first and the context after. Write the answer as a plain sentence, then a short paragraph that expands it. This single habit is the difference between a page that gets read and a page that gets scrolled past.
Check the page the way a customer does
Open it on your phone, on mobile data, not office wifi. Count the seconds until you can read something useful. If the page jumps around while loading, or the first thing you see is a cookie banner, fix that before you touch the copy. Speed and stability are part of the answer too.
This week: make the site easy to find
Structure is the quiet multiplier. A great page three clicks from the homepage is treated as a minor page, by readers and by crawlers. Open your site and count the clicks from the homepage to the page that makes money. If it is more than two, that is the week's project.
Internal links are the cheapest tool you have. Every page you publish should link to the two or three pages it naturally leads into, and those pages should link back. Do this by hand for your ten most important pages and you will feel it in the crawling before you feel it in the rankings. Avoid sitewide footer link blocks - they spread attention rather than focus it.
Then look at what loads. Compress the images above the fold, remove the carousel nobody clicks, and check that the page works with scripts blocked. You do not need a perfect score on a speed test. You need a page that opens fast enough that a buyer on a train does not give up.
This month: publish less, aim better
Two pages a month beats eight. Each one should answer a question a buyer actually types, and each should have a clear next step at the end. Before you write, check whether you already have a page on that topic. Refreshing an existing page almost always costs less than publishing a new one that competes with it.
Then earn a mention. Five thoughtful requests beat fifty templates: one podcast note, one supplier page, one industry roundup, one directory that your buyers actually use, one partner who will introduce you. You are not collecting links, you are collecting places where the right person can find you. Track each one in the same sheet as your pages, with a column for referral sessions.
At the end of the month, review honestly. Which page moved, which did nothing, and why? A page that did not move usually aimed at a query nobody searches, or answered a different question than the one that was asked. Both are fixable, and both are cheaper to fix than to replace.
The daily 30 minutes
One habit, three blocks. It is short enough to keep on a bad day, which is the only test that matters.
Should I hire someone? Three lines
I am biased, so use these as tests rather than opinions. Cross even one line and the maths usually favours getting help.
You have not published anything in three months
The plan is not failing because of knowledge. It is failing because nobody has the hours. Hiring a writer fixes a capacity problem; buying another audit does not. Fix capacity first, then decide whether you still need advice.
You cannot name your ten most valuable URLs
If the site is large enough that you cannot say which pages earn impressions or revenue, the risk is not a missing tactic, it is a missing map. That is a technical project, and it is usually cheaper done once by someone who has done it fifty times.
The site pays for itself and you are the bottleneck
When organic traffic already produces revenue and every improvement waits on your calendar, the cost of delay is measurable. At that point you are not buying SEO, you are buying your own time back. Compare it to the margin you lose per month of waiting.
What to ignore for now
If you are at the start, skip these without guilt: a full technical audit of a twenty-page site, a link-building retainer, an AI content pipeline, and any dashboard that takes a day to build. They are real work, but they are the wrong work before you know which pages deserve attention. Learn the vocabulary first - our plain-language answer to Whats Is SEO? covers the terms in one sitting - and then spend your hours on pages, not tools.
When you are ready for someone to run the page-level work with you, that is exactly what our On-Page SEO Services are built around: titles, headings, internal links and answer-first copy, handed back with a note on every change.
Want a second pair of eyes on your list?
Send three URLs and the queries you want. We will tell you which one to fix first, and why - no pitch attached.
Questions I get asked most
How do I SEO my site for free?
Use the free tools first: search console for impressions and positions, a page speed check, and your own analytics. Then spend time, not money: rewrite titles, answer the main question in the first paragraph, and link related pages together. Paid tools save hours, not ideas.
How long before I see results from my own SEO work?
For a page that already ranks on page two, expect movement in 2 to 6 weeks after a real rewrite. New pages take 3 to 9 months in a competitive niche. If nothing moves in 90 days, the problem is usually the target query, not the effort.
Do I need a paid SEO tool to start?
No. A spreadsheet, search console and analytics cover the first three months. Buy a paid tool when you have more than 100 pages to track or need competitor data at scale, not before.
What should I fix first on a brand new site?
Indexation, then navigation, then copy. Confirm the pages are indexed, make sure every important page is two clicks from the homepage, and write one page per service with a clear answer at the top. Links can wait until there is something worth linking to.
Can I do SEO myself with no experience?
Yes, at the page level. Titles, headings, internal links and clear answers are learnable in a week. What takes longer is judgement about which pages deserve the work, and that comes from reading your own search console data every week.
How do I know if my SEO is working?
Track three numbers monthly: impressions, average position for your target queries, and conversions from organic sessions. Traffic alone can rise while quality falls. If impressions rise and position improves but conversions do not, fix the page, not the channel.