How many pages should I touch in my first month of SEO?

Five to ten pages. A small set done properly teaches more than fifty edits you cannot evaluate, and it keeps the monthly review short enough to happen.

What SEO work should I not do alone?

Site migrations, template changes across thousands of URLs, and tracking changes for paid campaigns. Those break quietly and cost more to repair than to outsource.

How Do I Do SEO? A Seven-Day Start Plan for Founders

WebSEOGrow

Self-Talk Card

How Do I Do SEO? A Seven-Day Start Plan for Founders

How do I do SEO? You do it in the order a stranger would read your site: find the ten pages that already deserve traffic, make each one obvious to a first-time visitor, link them together, then watch one number for ninety days. That is a complete plan. Everything else, from tools to audits to dashboards, is support for those four moves rather than a replacement for them.

No account signup, no crawl budget talk, no ten-tab dashboard. One page at a time.

You ask

Where do I actually start? There are forty things on every checklist.

You ask

How long until this works, and how will I know it is working?

You ask

Do I need a tool subscription, or can I do this in a spreadsheet?

You ask

When is doing it myself the wrong call instead of the brave one?

four questions, one honest answer

What is the shortest honest answer to how do I do SEO?

You do SEO by matching one page to one question a buyer types, making that page the clearest answer available, linking it from pages that already get crawled, and measuring whether the people who arrive ask for a quote. Repeat on the next page. Nothing in that loop needs a big budget, only a named owner and a fixed review date.

The four questions running in your head, answered plainly

Every founder who tries SEO alone asks the same four things, usually at 11pm. Here is what each question is really about, and what to do with the answer.

Where do I start?

Not with a tool, and not with a new blog. Start with the pages that already exist and already earn something: your product pages, your service pages, your most-visited article. Those pages have history, links and traffic you can improve this week.

A new site has no history, so the same work takes months to show. The fastest way to learn SEO is to change a page that already has an audience.

Do this: list ten pages. Write the one question each page should answer, in the words a customer would use.

How long does it take?

Page-level changes on pages that already rank can move within four to eight weeks. New pages on a site with modest authority take longer, and pretending otherwise is how people quit in month two.

Give the work one quarter before judging it. Judge the leading indicators monthly, though: impressions on your target query, pages indexed, and how many people reach the quote form.

Do this: put a review date in the calendar before you start, not after you get impatient.

What does it cost me?

The shape of the budget matters more than the size. Most of the cost is your own hours, then a crawl tool, then writing help if you cannot write clearly in English. Software is the smallest line and the one people buy first.

You can run the first month on a free crawl, a free search console and a spreadsheet. Spend money when a specific task is blocked, not because a comparison page said so.

Do this: price your own hour. If ten hours of your time costs more than the task, buy the task.

How do I know it is working?

Not by ranking screenshots. Use two numbers: qualified sessions from non-branded queries, and quote requests. Both are visible in free tools once tracking is honest, and both move for reasons you can explain.

Add one diagnostic: which pages gained or lost position this month. That is where your next hour should go.

Do this: one event, one source, one monthly page. Anything more becomes a hobby.

Your first seven days, one hour at a time

A week is enough to build the habit and see the first signal. Each day has one output, because a day without an output is a day you will repeat.

Day	Task	Time	Output you keep
Day 1	Write down what you sell, in the sentence a buyer would repeat to a colleague	45 min	One offer sentence plus five customer questions
Day 2	Pick ten queries you can plausibly answer, and name the page for each	60 min	One query to page map
Day 3	Crawl your own site and read the index report without fixing anything yet	45 min	List of unindexed pages and broken links
Day 4	Rewrite one title and the first screen on your five most valuable pages	90 min	Five rewrites ready to publish
Day 5	Add three internal links into each of those pages from pages that already get traffic	60 min	An internal link note for your developer or CMS
Day 6	Confirm the quote form, chat and phone click each fire once per session	60 min	One working conversion event
Day 7	Book a monthly review and decide which four numbers you will read	30 min	Calendar block and a one-page review template

If a day slips, skip the next task rather than doubling up. The point of the week is a repeatable loop, not a perfect week.

When to stop doing it yourself

Self-guided SEO fails for structural reasons, not for lack of talent. Three tests tell you whether you have hit a structural wall. If two of them are true, the honest move is to hand over the loop, not to work later at night.

You cannot ship the change

A title rewrite needs a developer ticket, a staging review and a two-week wait. When the loop from idea to live page takes a month, the feedback never accumulates and you are guessing by quarter two.

The site changes faster than you can document

New products, new templates, new URLs every week. Someone has to own the queue and decide which changes ship with SEO considered, or the same errors reappear every release.

Nobody owns the numbers

Reporting is a file someone opens when a director asks. If no person is accountable for qualified sessions and quote requests, the work drifts toward whatever is easy to show instead of what pays.

Test one is about access, test two is about cadence, test three is about accountability. You can fix any of them without an agency, and you should know which one is broken before you hire.

Halfway through the week and stuck on the crawl?

Send the domain and the report you do not understand. We will tell you which finding actually matters and which one is noise, without a pitch attached.

The minimum path that still produces results

Strip SEO down and five things remain. Everything else is either acceleration or distraction, and you can tell the difference by asking whether it changes a page a buyer will read.

One intent per page. A page that tries to answer three questions ranks for none of them clearly. Split it and link the pieces.
A first screen that answers. The heading and the opening paragraph should settle the visitor's question before they scroll. This is also what earns the featured result.
Internal links with your own words. Link from pages that already have traffic, using the phrase a customer would search, not "click here".
Honest technical basics. Indexable, fast enough, no duplicate URL doing the same job as another. If you are unsure what a crawl report means, start with Whats Is SEO? for the vocabulary and come back to the report with it open.
A review you actually attend. Same four numbers, same day each month, one decision per page reviewed. Doing this alone puts you ahead of most small teams.

What we take over, and what stays yours

Handing over does not mean handing over everything. Most clients keep the offer, the subject knowledge and the final approval, and pass us the parts that need a routine: the crawl, the page rewrites, the link decisions and the monthly review.

We usually start with the ten pages you already picked during your seven days. That means the first month builds on your work instead of restarting it, and you can compare our judgement against yours on pages you understand.

If the bottleneck is page-level clarity rather than tooling, our On-Page SEO Services work is the smallest useful engagement: a defined list of URLs, one intent each, rewritten to answer in the first screen.

What comes back to you

The query to page map, the crawl file, a rewrite sheet in your own CMS fields, and a review template with the four numbers already wired.

If you later bring it back in house, you keep the working files instead of a login you do not control.

Questions people ask before their first DIY month

Do I need paid tools to start?

No. A free crawl, the search console you already have and a spreadsheet cover the first month. Buy a tool when a specific report blocks a decision, not before.

Should I start with a blog or my product pages?

Product and service pages first. They already have a reason to exist and often have history. A blog is a long game you can add once the commercial pages answer clearly.

How many pages should I touch in month one?

Five to ten. A small set done properly teaches you more than fifty edits you cannot evaluate, and it keeps the review meeting short enough to actually happen.

What if I write badly in English?

Write the answer in your own words first, even roughly, then have it edited for clarity. Buyers forgive simple sentences. They do not forgive a page that never answers the question.

Is it worth doing SEO myself for three months?

Yes, if you can ship changes and hold the monthly review. Three months of honest iteration teaches you what to buy later, and it makes any agency conversation far more useful.

What should I not do alone?

Site migrations, template-level changes across thousands of URLs, and anything that touches tracking for paid campaigns. Those break quietly and cost more to repair than to outsource.