Reading Time: 12 minutes

How to Choose the Best Keywords for SEO Without Wasting Months on the Wrong Ones

Six months. That’s roughly how long a lot of businesses spend chasing a keyword before they admit it was never going to pay off.

Maybe it was too competitive. Maybe it brought traffic that never bought anything. Maybe it ranked beautifully and then Google’s AI Overview started answering the question right on the results page, so nobody clicked through. Whatever the reason, the cost is the same: months of content, links and budget poured into the wrong hole.

So let’s fix that before it starts. This is a practical walk through how to choose the best keywords for SEO, the way an experienced strategist actually does it, including a few places where the usual advice falls apart.

Start with money, not with a tool

Most keyword research begins by opening a tool and typing in a seed term. Understandable. Also backwards.

Start instead with three questions about your business:

  1. What do we sell that has the best margin or lifetime value?
  2. Who buys it, and what problem are they trying to solve in the moment before they buy?
  3. What would they type, or ask an AI assistant, at that moment?

Write those answers down in plain language. They become your filter. Every keyword you consider later should connect back to one of them. If it doesn’t, it’s probably a vanity term, nice for traffic charts and useless for revenue.

A family law firm, for instance, might find “how does divorce work” has huge volume. Great traffic. But the person searching it is often months away from hiring anyone, if they ever do. “Contested divorce lawyer Nassau County” has a fraction of the volume and a far higher chance of turning into a consultation this week.

Volume is not value. That’s the first lesson, and honestly the one people forget most.

Understand the long tail, really understand it

Here’s a number that changes how you think about this. Ahrefs found that 94.74% of keywords get 10 or fewer searches a month. Almost 95%. The tiny slice of high-volume terms everyone fights over is the exception, not the rule.

Ahrefs also notes that roughly 15% of daily Google searches are brand new, never searched before. People keep inventing new ways to ask things, especially now that AI tools encourage longer, more conversational questions.

What does this mean for you?

It means a page built around one well-chosen topic can rank for hundreds of related low-volume phrases. Ahrefs’ data shows the average top-ranking page also ranks in the top 10 for nearly 1,000 other relevant keywords. You’re not really choosing a single keyword. You’re choosing a topic cluster, with one primary phrase to anchor it.

Read search intent before anything else

Type your candidate keyword into Google. Look at what’s actually ranking. This one habit will save you more wasted months than any tool.

Ask yourself:

  • Are the top results blog posts, product pages, local listings, videos or tools?
  • Is there an AI Overview at the top? How complete is it?
  • Are the ranking pages from big publishers, small businesses, directories, forums?
  • Does the results page show a map pack, shopping ads, “People also ask” boxes?

If you want to rank a service page and the whole first page is how-to articles, you’re fighting the intent. Google has decided what searchers want, and it isn’t a sales page. You can either write the guide or pick a different keyword.

Intent usually falls into a few buckets:

  • Informational. “What is a fiduciary,” “how to fix a leaky faucet.”
  • Commercial investigation. “Best CRM for small agencies,” “Invisalign vs braces cost.”
  • Transactional. “Buy standing desk,” “book plumber Brooklyn.”
  • Local. “Emergency dentist near me,” “therapist Upper West Side.”

Map each keyword to one of these, and match the page type accordingly. Simple. Rarely done.

Judge difficulty honestly

Every keyword tool gives you a difficulty score. Useful, but treat it as a rough signal, not gospel.

A better gut check: look at the sites ranking in the top five. Are they much bigger and more authoritative than yours? Do they have pages that are clearly better than what you could produce? If yes to both, that keyword is a long-term goal, not a this-quarter win.

Newer or smaller sites usually do better targeting phrases where the current top results are thin, outdated or poorly matched to intent. You’ll find these more often in the longer, more specific phrases. Not always, of course. Sometimes a two-word term is wide open in a niche market. You just have to look.

And a reality check from Ahrefs: only about 1.74% of newly published pages reach the top 10 for at least one keyword within a year. Most pages never rank. Choosing achievable targets isn’t timid; it’s survival.

Don’t trust search volume blindly

Here’s something a lot of marketers don’t realize. Ahrefs compared Google Keyword Planner’s volume estimates with real data and found Keyword Planner overestimated volume 54.28% of the time.

Third-party tools have their own quirks too. All of them are estimating.

So use volume to compare keywords relative to each other, not to forecast exact traffic. If one phrase shows 1,200 searches and another shows 90, the first is probably bigger. Whether it’s really 1,200? Nobody knows for sure. Your own Search Console data, once pages start getting impressions, will tell you the truth faster than any tool.

Factor in how AI answers change the math

This is newer, and it matters more every month.

Some queries now get answered so completely by Google’s AI Overviews or by ChatGPT that very few people click through. Pew Research Center found users clicked a traditional result on just 8% of visits when an AI summary appeared, versus 15% when there was no summary.

Does that mean you should avoid informational keywords entirely? No. But be selective.

Keywords that tend to hold up well even with AI answers:

  • Queries that need local or personal context (“best orthodontist for adults in Queens”)
  • Comparison and decision queries where people want depth
  • Questions where trust matters and people want to see the source (legal, medical, financial)
  • Anything with transactional or booking intent

Keywords more likely to lose clicks:

  • Simple definitions
  • Quick facts and conversions
  • Basic “how to” queries with one-line answers

Even then, being cited inside an AI answer has value. Semrush found visitors arriving from AI search converted at 4.4 times the rate of typical organic visitors. Fewer clicks, but warmer ones.

Dominate the Search Results

Stop losing clients to your competitors. Get a high-impact SEO & PPC strategy designed specifically for criminal defense law firms.

Build a shortlist with a simple scoring method

Once you have a pile of candidate keywords, score each one from 1 to 5 on four things:

  • Business value. How directly does this lead to revenue?
  • Intent match. Can we build the type of page Google wants for this?
  • Achievability. Can we realistically compete with the current top results?
  • Click potential. Will people still click through, given AI answers and SERP features?

Add the scores. The top of the list is where you start. It isn’t scientific, and that’s fine. It forces you to weigh trade-offs instead of chasing the biggest number on the screen.

You might notice the highest-volume keywords rarely end up at the top. That’s usually a sign the method is working.

Use the sources people forget

Tools are great. But some of the best keyword ideas come from places no tool scrapes:

  • Sales calls and emails. Write down the exact phrasing prospects use. Not your industry jargon, theirs.
  • Customer service tickets. Recurring questions are content gold.
  • Reddit and niche forums. How do real people describe the problem when nobody’s selling to them?
  • “People also ask” boxes. Click a few and they expand endlessly.
  • Your own Search Console. Queries where you already get impressions but sit on page two are often the fastest wins of all.
  • Competitor pages. Not to copy, but to spot gaps they haven’t covered.

That last one deserves a note. Running a proper SEO competitor analysis often reveals profitable keywords competitors rank for by accident, without a dedicated page. Those are openings.

A quick worked example

Say you run a small pediatric dental practice in Brooklyn. A tool shows “dentist” with enormous volume. Tempting. Also hopeless, and mostly irrelevant, because many of those searchers aren’t parents, aren’t local, and aren’t ready to book.

Work the process instead. Your best-value service might be new patient exams for kids under six, since those families stay for years. Parents in that moment are probably asking things like “pediatric dentist Park Slope accepting new patients,” “first dentist visit for toddler,” or “kids dentist open Saturday Brooklyn.”

Check the results pages. The first one shows a map pack and local practice sites; perfect for a service page. The second shows blog posts and an AI Overview; good for a helpful guide that links to your booking page. The third is local again, with fewer strong competitors.

Score them, and suddenly the plan writes itself. One location-focused service page, one parent-friendly guide, and a Google Business Profile that clearly lists weekend hours. None of these phrases looks big on its own. Together, they bring in exactly the families you want.

Map one primary keyword to one page

A mistake we see constantly: three or four pages all half-targeting the same phrase. Google gets confused about which one to rank, and often ranks none of them well.

Keep a simple spreadsheet. One row per page, one primary keyword, a handful of secondary phrases. Before writing anything new, check whether an existing page already covers it. Often the better move is improving that page instead of publishing another.

How many times should the keyword appear?

Quick tangent, since people always ask. Forget old rules about keyword density. Use the primary phrase where it reads naturally, typically in the title, early in the page and perhaps once more, and then write about the topic thoroughly. Related terms and synonyms matter far more than repetition. Stuffing the exact phrase reads badly, and modern search systems understand meaning, not just matching strings.

Revisit, because keywords drift

Markets change. New products launch, language shifts, AI features roll out to new query types. A keyword plan built in January may look stale by September.

A light review each quarter is enough for most businesses. Check which pages gained or lost impressions, which new queries started appearing in Search Console, and whether any target keyword now triggers an AI Overview that didn’t before. Adjust. Keep moving.

So what does this look like in practice?

Pull it together and the process is something like this:

  1. Define what you sell, to whom, and the moment before they buy.
  2. Brainstorm from sales conversations, service tickets, forums and Search Console.
  3. Expand with a keyword tool, focusing on topics rather than single phrases.
  4. Check intent and the live results page for each candidate.
  5. Score for value, intent match, achievability and click potential.
  6. Map one primary keyword per page, secondary terms alongside.
  7. Build or improve pages, then measure real impressions and conversions.
  8. Review quarterly and adjust.

It’s not glamorous. It doesn’t need to be. Done well, it’s the difference between content that quietly brings in clients for years and content that just sits there.

Where to go from here

Learning how to choose the best keywords for SEO is half strategy, half discipline. The strategy is knowing what your buyers actually want. The discipline is saying no to the shiny high-volume terms that won’t pay the bills.

If you’d like help building that plan, Digital Drew SEM runs keyword strategy as part of every SEO engagement in NYC and beyond, combining live search data with what’s happening inside AI answers. Our data-driven SEO approach ties keyword choices to revenue, not just rankings, and our SEO copywriting team turns that research into pages that rank and convert. Prefer to keep it in-house? Start with a free AI SEO audit to see which keywords you’re closest to winning.

Running paid search too? Keyword insights from a well-managed Google Ads campaign can show you which terms convert before you invest months in organic content. When you’re ready, talk to the Digital Drew SEM team.

Frequently Asked Questions 

1. How do I choose the best keywords for SEO for my business?

Start by identifying your most profitable products or services and understanding what potential customers search for before making a purchase. Evaluate keywords based on search intent, business relevance, competition, and click potential. Prioritize keywords that attract qualified visitors who are more likely to become customers rather than focusing only on high search volume.

2. What is more important for SEO, keyword search volume or search intent?

Search intent is often more important than search volume because it reveals what users want to accomplish. A high-volume keyword may generate traffic without conversions, while a lower-volume keyword with commercial or transactional intent can attract customers ready to take action. The best SEO strategy balances search demand with relevance, user intent, and conversion potential.

3. How can I find low-competition keywords that are easier to rank for?

Look for specific long-tail keywords related to your services, products, or target locations. Use keyword research tools to identify opportunities, then examine Google’s top-ranking results to assess competition. Keywords with outdated content, weak competitors, or poorly matched search intent may offer better ranking opportunities. Google Search Console can also reveal keywords where your existing pages are close to reaching the first page.

4. How do Google’s AI Overviews affect SEO keyword selection?

Google’s AI Overviews can answer informational searches directly, reducing the number of users clicking traditional search results. Businesses should consider whether a keyword still encourages visitors to explore a website. Local searches, detailed comparisons, service-related queries, and purchase-focused keywords may offer stronger click potential. Creating trustworthy, comprehensive content can also improve opportunities for visibility within AI-generated answers.

5. How many keywords should I target on a single webpage for SEO?

Each webpage should focus on one primary keyword supported by relevant secondary keywords, related phrases, and natural language variations. Rather than repeating the same keyword excessively, create comprehensive content that addresses the user’s search intent. Assigning a distinct primary keyword to each page also helps prevent keyword cannibalization, where multiple pages compete for the same search query. Review keyword performance regularly using Google Search Console.

Drew Blumenthal

Drew Blumenthal is the founder and CEO of Digital Drew SEM, a results-driven, performance-focused digital marketing agency based in New York.

SEO
November 9, 2026
Ready to Grow?

Let's have an honest
conversation.

You've read enough websites. Tell us about your business and we'll tell you honestly whether we're the right digital marketing agency in NYC for the job. No hard sell. No fluff.

Request A Free Consultation

This field is for validation purposes and should be left unchanged.

Let’s See If We’re a Fit

We’re selective about who we take on — and we want to make sure we’re the right team for your goals before either of us commits. Tell us about your business.

Name(Required)
Scroll to Top