How to Choose Keywords Customers Actually Search

Choosing keywords is simply working out the exact words and phrases your customers type into Google when they are looking for a business like yours, then making sure your website uses those words. You do not need paid software to do it well. Google itself, your own customer conversations, and a couple of free Google tools give you everything you need to find phrases that bring real enquiries. This guide walks a New Zealand owner through a repeatable process, from your first list of words to a working plan mapped to your pages.

The mistake most owners make is guessing. They optimise for the words they use in their trade, not the words a customer uses at their kitchen table. Get that gap right and the rest follows.

What does choosing keywords actually mean for a small business?

Choosing keywords means finding the real searches your customers make and matching your website pages to them, so you show up when someone is ready to call, book or visit. A keyword is any phrase a person types or speaks into Google, from “plumber in Tauranga” to “who can fix a leaking roof in Hamilton today”. When the words on your page match the words in the search, Google is far more likely to put you in front of that person.

For a small New Zealand business, this matters more than almost anything else you do online. If you use the same language your customers use, you appear at the moment they are deciding who to hire. If you guess wrong, or copy generic marketing buzzwords, you end up ranking for phrases that either nobody in New Zealand searches or that attract the wrong people entirely. That is wasted effort with no enquiries to show for it.

In 2026, searches are also more conversational than they used to be. People type full questions like “how much does a heat pump cost to install” and speak into their phones with phrases like “who installs heat pumps near me in Hamilton”. Search engines and AI answer summaries still read those words to decide which pages to show, so the language you choose is the bridge between a searcher’s question and your business. This is closely tied to how you get found on Google as a NZ service business, where keyword choice is the first lever.

Where do I find keywords before touching any tool?

Start with what you already know, straight from real customers, because their exact wording is the most valuable keyword source you have. Before you open a single tool, open a spreadsheet with three columns: the exact wording, where you heard it, and the problem behind it. In twenty to thirty minutes you can fill it with genuine search language.

Pull phrases from recent emails, contact forms, text messages, reviews and support calls. Write down the questions customers actually ask you, word for word, such as “can you do same-day tyre repair” or “how much does a bathroom reno cost”. Note the everyday language they use rather than your trade terms. A customer says “roof leak repair”, not “residential roofing remediation”. A homeowner says “heat pump”, while your supplier catalogue might say “reverse cycle split system”.

That last point is a real New Zealand trap. Local users search “heat pump” for home systems, while “air conditioning” skews toward commercial buyers. Australian and US content talks about split systems, reverse cycle and AC, and copying that terminology can quietly steer you toward the wrong words. The same goes for spelling. Kiwis and Google in New Zealand expect “colour”, “centre” and “optimise”, not the American forms, so lean into NZ English from the start.

These phrases become your seed keywords, a starting set of ten or twenty phrases you will test and expand in the next steps. If you have a site search box on your website, check its log too. Those queries are literally what visitors expected to find and typed in themselves.

How do I use Google itself as a free keyword tool?

Google gives you three free windows into real searches, and you can mine all of them in a few minutes without any software. Open an incognito window so your own history does not skew the results, then work through your seed phrases one at a time.

First, use autocomplete. Type a seed phrase slowly and do not press enter. Google suggests eight to ten completions based on searches people actually make. Push it further by adding patterns: “[service] near”, “[service] in”, “best [service]”, “[service] cost”, “[service] vs”. A landscape designer typing “garden design” might see “garden design Christchurch”, “garden design cost NZ” and “garden design small backyard”. Each suggestion is a phrase with real demand behind it, because Google would not surface it otherwise.

Second, run one of your better phrases as a full search and read the People Also Ask box. Expand each question and Google loads more. These are the follow-up questions your customers have, and each one is a candidate for an FAQ on a service page or a future blog article. Third, scroll to the bottom of the results and read the related searches. These reveal extra wordings and nearby topics.

If the same phrases keep appearing across autocomplete, People Also Ask and related searches, that is a strong sign real people use that language. One seed phrase can easily produce twenty to thirty unique variations this way, and ten seeds worked through carefully can generate a couple of hundred raw ideas in under an hour. Add every phrase a genuine buyer might type into your spreadsheet.

How do I check whether a keyword has real demand in NZ?

You do not need to run ads to see Google’s own demand data, and three free Google tools cover it. The goal is not a perfect number. It is confirming a phrase is not a ghost that nobody searches, and comparing which options have more interest in your part of the country.

Google Keyword Planner, inside a free Google Ads account, shows average monthly searches and suggests related keywords you had not thought of. Set the location to New Zealand, or better still to your region such as Auckland or Canterbury, so the numbers reflect your actual market. Without active ad spend, Planner reports volume in ranges like “10 to 100” or “100 to 1K” rather than exact figures, which is fine for choosing between phrases. For a local service, a term showing 10 to 100 searches a month is often plenty to be worth a page, especially when the intent is strong.

Google Trends does not give absolute volume but shows relative popularity from 0 to 100 over time and by region. You can compare up to five phrases at once and see which has more interest in New Zealand, and whether demand is seasonal. Air conditioning searches spike in summer, heat pump servicing in autumn. Favour phrases whose trend line is flat or rising rather than steeply declining.

Google Search Console is the most credible free source of all, once your site has any traffic. It lists every query where your site appeared, with impressions, clicks and average position, for up to sixteen months. This is your own data about what really brings New Zealanders to you, and we come back to it below.

A word of honesty about the numbers. New Zealand search volumes are small compared with the US, and many tools estimate from larger global datasets, so treat every figure as a rough guide. A phrase showing “no data” or “0 to 10” is not worthless. Long-tail queries are often under-reported yet collectively deliver most of the traffic and the best conversions. Pattern recognition beats chasing a precise number.

Should I target broad high-volume keywords or specific ones?

For almost every New Zealand small business, specific long-tail phrases beat broad high-volume terms, because they carry clearer intent and are realistic to rank for. A long-tail keyword is a longer, more specific phrase such as “after hours plumber Dunedin” or “fixed fee bathroom renovation Christchurch”, rather than a single word like “plumber” or “renovation”.

There are three reasons this matters more here than in the US. First, intent. Someone searching “same day washing machine repair Invercargill” is ready to call, not idly browsing. A broad term like “washing machine” could be a shopper, a repairer, or a student writing an essay. Second, competition. Broad national terms are dominated by large companies and directories, and a small local firm rarely outranks them. Specific phrases are winnable. Third, scale. New Zealand simply does not have the millions of monthly searches that make giant head terms worth chasing, so “Hamilton mortgage broker for first home buyers” is a smarter target than “mortgage”.

Layer local and real-world modifiers deliberately rather than waiting for Google to suggest them. Take your core service and add city (“landscape designer Auckland”), suburb (“landscape designer Ponsonby”), “near me”, urgency (“emergency electrician Christchurch”), and price or segment (“affordable”, “luxury”, “fixed fee”). Each of these is often worth its own focused page or at least its own section. Aim for a mix: a few broader phrases plus many specific buyer-style phrases across your key pages. If you want to go deeper on the local angle, local SEO versus national SEO for NZ businesses explains when each approach fits.

What does a real NZ keyword comparison look like?

A real example makes the trade-off concrete. Take a renovation business in Auckland and look at the actual monthly search figures from Google’s Keyword Planner, set to New Zealand. “Kitchen renovation Auckland” gets around 720 searches a month, which looks like the obvious target until you notice it is marked high competition, meaning established firms and directories already own it. Right beside it, “house renovations Auckland” gets around 480 searches a month at low competition. Almost as much demand, a fraction of the difficulty. That single comparison is the whole lesson: the biggest number is rarely the smartest target.

Go more specific and the pattern holds. “Bathroom renovation cost Auckland” shows only about 20 searches a month, but at low competition and with unmistakable buyer intent, because someone typing “cost” is pricing a real job rather than browsing. “Small bathroom renovation cost NZ” sits around 70 a month at medium competition. “Home renovation builders Auckland” is roughly 50 a month at low competition. None of these are big numbers by overseas standards, and that is exactly the New Zealand reality: a handful of genuine buyers a month on a phrase you can realistically rank for beats a few hundred on a phrase you never will.

Two honest caveats. First, without active ad spend the Keyword Planner reports demand in rounded bands rather than precise counts, so treat every figure as a ballpark, not a promise. Second, the competition label reflects advertiser competition, which tracks organic difficulty closely but not perfectly, so still read the actual results as your final check. The workflow that follows is simple: list your candidate phrases, note volume and competition side by side, and start with the ones that pair real intent with reachable competition, which for most New Zealand service businesses means the specific cost-and-location phrases rather than the head terms.

How do I check what keywords my competitors already rank for?

You can learn exactly which phrases work for similar NZ businesses just by searching Google and reading their pages, no tool required. Top-ranking pages already reflect what Google believes matches a searcher’s intent, so they are free market research.

Take three to five of your main phrases and search each one. Open the top five results that are businesses like yours, ignoring large marketplaces and directories unless you sell products. For each page, capture a few things in your spreadsheet: the page title shown in the browser tab, the main heading, the subheadings, and any phrases that repeat through the service descriptions and FAQs. Add a column and mark any keyword you see on two or more competitor sites.

When the same phrase appears across several strong competitors, treat it as a firm signal that the phrase brings traffic and enquiries. If three Christchurch builders all target “bathroom renovation cost Christchurch” in their titles and headings, that phrase matters, even if a free tool shows little volume for it. You are not copying their writing. You are learning the language that clearly earns them work, then covering it better in your own words. This is also a first look at why competitors outrank you on Google, which usually comes down to matching intent and depth, not tricks.

While you are reading those results, notice the format Google rewards. If the top pages are all long guides, a thin page will struggle. If they are all service pages, a blog post may underperform. Match the dominant format for that search.

How do I match each keyword to the right page and intent?

Every page should have one clear job: a single main topic, a handful of related phrases, and an intent it satisfies. Choosing keywords is useless until each phrase has a home, so this step turns your list into a plan.

Start by classifying each keyword by intent. Transactional phrases mean the person is ready to act, like “emergency vet Auckland” or “book heat pump install Wellington”. These belong on your service and contact pages. Commercial phrases mean they are comparing options, like “best fertility nutrition coach NZ”, and suit comparison or service pages. Informational phrases mean they are learning, like “how to reduce your power bill in a NZ winter”, and belong on blog or guide pages. Cover your transactional phrases first, because they bring the enquiries, then commercial, then informational content once the money pages exist.

Now map keywords to pages. Assign one primary keyword per page and two to five closely related variants. An emergency plumber page in Tauranga might own “emergency plumber Tauranga” as its primary, with “24 hour plumber Tauranga”, “urgent leak repair Tauranga” and “after hours plumber Tauranga” as variants. Put the primary phrase in the page title, the main heading, the first paragraph, the meta description, and an image alt text where it fits naturally. Search engines and AI answer summaries read this structure to decide when to show your page and for which searcher.

Two cautions. Do not force the exact phrase in over and over. Modern search is very good at spotting awkward keyword stuffing, and it reads natural variants and synonyms as the same topic anyway. And if you find yourself assigning fifteen unrelated ideas to one page, that is a signal you need more pages, not more keywords crammed onto one.

What do keywords have to do with AI search and answer engines?

Keyword choice still drives whether AI answer tools mention your business, because tools like ChatGPT, Google AI Overviews, Gemini and Perplexity read the same clear, well-structured pages that Google ranks. The words you choose and the questions you answer are how these systems understand what you do and when to cite you.

Two shifts are worth understanding, with the honest caveat that most hard figures here come from overseas research, not New Zealand data. Overseas studies suggest well over half of searches now end without a click, as people read an answer directly on the results page. That does not mean keywords stop mattering. It means being the source that gets quoted matters more, and that comes from answering real questions plainly. Broad “what is” and “how to” queries are the ones most likely to trigger an AI summary, while bottom-of-funnel searches with “cost”, “price”, “near me” or “book” are less affected and still send ready buyers to your site.

The practical move is to keep your transactional local phrases as the backbone, since they convert and are less exposed to answer boxes, while writing genuinely useful answer-first content for the informational questions your customers ask. Structuring a page around the exact question, then answering it in the first line, is what gets you quoted. For the detail, see how to write content that answer engines quote and how to get your NZ business cited in ChatGPT, Gemini and Perplexity. If you are still weighing whether any of this is worth your time, answer engines for small business is the plain-English case for a NZ owner.

How do I keep my keyword list working over time?

Keyword research is a monthly habit, not a one-off project, because searches, competitors and your own site all change. The businesses that win with free tools are the ones that spend an hour a month refining, not the ones that build a perfect list once and never touch it.

Once your site has traffic, Google Search Console becomes your best free keyword tool. Each month, open Performance, filter to the last three months, and sort your queries by impressions rather than clicks. Look for two patterns. First, near-miss keywords, where you have real impressions but sit in position 8 to 20. A focused edit to that page, using the exact phrase in a heading, can move you to page one relatively quickly. Second, surprise queries you never targeted but which clearly match what you offer. Those are ready-made ideas for a new page or section.

Build the habit into a simple 60 minute monthly sprint. Spend ten minutes adding new phrases customers used in recent calls and messages. Spend fifteen running autocomplete and People Also Ask for one or two core services. Spend fifteen in Search Console finding near-miss terms. Spend ten updating your keyword-to-page map. Spend ten deciding one change to publish, whether a new page, a sharper headline or an added FAQ. That single monthly change compounds over a year.

When you are unsure about a keyword, run a quick three-question test. Could a real paying customer say this out loud? Do the top results show businesses like mine rather than definitions or news? Does this match something I actually want to sell? If any answer is no, deprioritise it, however impressive the volume looks. Remember too that keywords set the direction, but content quality, a fast mobile-friendly site, and clear internal links between related pages decide whether you arrive. Your Google Business Profile is often the first thing to fix alongside this, and realistic patience helps, because SEO takes time to work for a NZ small business. Choose the words your customers actually use, give each one a home, and keep listening every month, and you will steadily show up where the enquiries are.