AkiliCamp is coming: practical AI and tech learning, online.Meet AkiliCamp.Join the early list

Insights

What search engine marketing means for your website

· 3 min read · Caleb Kellah

I built Kwarry, a free keyword research tool that returns real Google Ads Keyword Planner data for Kenya, Uganda, Tanzania and Rwanda: monthly searches, a 12-month trend against last year, buyer and question intent, and an opportunity score for each term. That is where search engine marketing has to start, with real numbers on what people type into Google rather than guesses. This page walks through what that work actually covers, what solid groundwork looks like, and how you can tell if any of it is working.

What search engine marketing actually covers

People use the phrase loosely. Some mean the ads that sit above search results. Others mean organic SEO: the content, structure and technical work that earns a place in results on its own.

Either way, it starts with the same question: what are people actually typing into Google, and what already ranks for it. Before you spend time on either one, it helps to know two things about your own market: how many people are actually searching for what you offer, and how much competition already exists for that search. Without those two numbers, a marketing plan is a guess dressed up as a strategy.

Answering that question properly, for your market and your competitors, is the heart of my SEO and marketing research work.

Research before you write or launch anything

Kwarry answers that question with real numbers instead of guesswork. Type in a term and you get the monthly searches, how that has moved over the past year, how much competition there is, whether people are searching to buy or to ask a question, and a score for how good an opportunity it is.

Not every search means the same thing. Someone typing a question is often still learning. Someone typing a specific product or brand name is closer to acting. It helps to look at your own list of terms the same way: group the ones that sound ready to act separately from the ones that still sound like someone learning.

My AI content pipelines use the same idea at a larger scale. Each of the eight pipelines picks its topics from live search demand, volume, difficulty and what is already ranking, before anything gets drafted.

If you are planning a site or a content push, pull the numbers for your own terms first. Kwarry is free to use for exactly that.

Protecting what you already have

Gaining new ground is one part of this work. Keeping what you already have is another.

When I moved a safari marketplace I built from WordPress to a new build, I crawled every one of its 2,245 live URLs first and checked that titles, descriptions and canonical URLs matched, before anything went live. If you ever move a site to a new build, list every page that already shows up in search first, including the ones you might forget. It is easy to forget an old post or a seasonal page that still quietly brings in search traffic. A migration can wipe out years of that visibility if nobody checks the small things.

The same care applies from day one. A poetry journal I built for a writer launched with author schema, a sitemap, Search Console and Analytics already in place. That gave it something to measure from day one.

Checking whether it is working

None of this means much if you cannot see whether it is working. My results page publishes live Search Console data, clicks, impressions and four-week comparisons, for client sites, refreshed automatically every week.

Reading that kind of data takes care. I compare like-for-like periods before I read any change in search numbers as meaningful. A longer weekend, a holiday or a site outage can shift numbers for reasons that have nothing to do with marketing. Compare the same stretch of days against a similar period from before, rather than only last week. What good SEO services actually look like covers how to tell the difference. The same care matters for the content behind it: AI for business starts with who checks the AI's work looks at why someone still needs to check what an AI pipeline publishes.

If you want the fuller method, Signal is where I write about keyword scoring, topic selection and migration checks in more depth. If you have a search engine marketing project in mind, tell me about it.

← All insights Talk to me about your project →