Writing Samples

How to Rank a Local Business on Google Without an Agency Budget

A complete, step-by-step local SEO guide, written the way I'd structure and publish it for a client's blog.

Why Local SEO Plays by Different Rules

National SEO is mostly a content and backlink competition. Local SEO adds a third factor that often matters more than either: proximity and proof of a real, active business at a real place. Google is trying to answer a more specific question than "what's a good page about this topic." It's trying to answer "which nearby business can actually solve this person's problem right now, and can I trust it." That's why a small business with a well-maintained Business Profile and a dozen recent reviews regularly outranks a bigger competitor with a fancier website and none of that.

In practice, this means local rankings come from three things working together: how well your Business Profile represents the business, how clearly your website answers the exact searches people make, and how much external, verifiable evidence exists that the business is real, active, and trusted. Miss one of the three and the others have to work much harder to compensate.

1. Get Your Google Business Profile Right First

For searches like "plumber near me" or "bakery in Westlands," Google leans heavily on your Business Profile, not your homepage. This is the single highest-leverage thing a small business can control, and most leave it half filled in.

Pick the most specific category available. "Plumber" beats "Contractor." Add secondary categories only if they're genuinely accurate, since a mismatched category can actively hurt relevance for the searches that matter most.

Fill in every available field. Hours, service area, attributes, a full business description, and a booking or contact link if you have one. An empty field is a missed signal, not a neutral one.

Use real photos, added regularly. Photos of the actual location, team, and finished work outperform stock imagery, and profiles that add new photos monthly tend to look more "alive" to both Google and potential customers than ones frozen in time.

Post updates and answer questions. The Posts feature and the public Q&A section are both underused. Answering your own most common customer questions there, before a stranger posts a wrong answer, quietly builds trust for anyone scanning the profile before calling.

Keep your name, address, and phone number identical everywhere. Even small mismatches (a spelled-out "Street" on your website versus "St." on your profile) can quietly weaken how confidently Google associates the two as the same business.

2. Build the Website Foundation Underneath It

Your Business Profile gets you into the conversation. Your website is what Google, and the customer, checks next to decide if you're actually a good fit.

One page per real service, not one page for everything. A generic "Services" page rarely ranks for anything specific. If people search "emergency plumber Kilimani" and "drain cleaning Nairobi," those deserve two separate pages, each answering that exact need early, in plain language, instead of one page trying to cover both and ranking for neither.

Add LocalBusiness structured data. A small block of schema markup on the site (name, address, phone, hours, service area) helps Google confirm what your Business Profile already says, and can unlock richer results like star ratings showing directly in search.

Write titles and meta descriptions like a person, not a keyword list. "Emergency Plumber in Kilimani, Nairobi | Same-Day Callouts" tells both Google and a searcher exactly what the page offers and where, in one line.

Don't ignore page speed and mobile layout. Most local searches happen on a phone, often while someone is standing in front of the problem they need solved. A slow page or a layout that breaks on mobile loses that customer before the content even gets a chance to help.

3. Find the Exact Words Your Customers Use

Good local content starts with knowing the real phrases people type, not the phrases a business owner would naturally use to describe themselves.

Start typing the service into Google and read the autocomplete suggestions. These are real, common searches, surfaced directly by Google, and they're free.

Check the "People Also Ask" box for the core service term. These questions make excellent subheadings and FAQ content, because they're confirmed real questions, not guesses.

Look at what's already ranking for your target terms. Not to copy it, but to see what those pages cover and where they're thin, vague, or outdated. That gap is usually where a smaller, sharper page can outperform it.

Match content types to intent. A "near me" search wants a fast answer and a way to contact you. A "how much does X cost" search wants a real, specific answer, even a range, not a "contact us for a quote" dodge. Giving a straight answer builds more trust than it costs in leads.

4. Build a Steady, Honest Flow of Reviews

Review count, rating, and recency all factor into local rankings, and they factor even more heavily into whether a stranger actually picks up the phone.

Ask right after a good outcome, with a direct link. Not weeks later, and not with a vague "please review us sometime." The easier and more immediate you make it, the more people actually follow through.

Reply to every review, good and bad. A short, genuine reply to a negative review, acknowledging the issue and how it was or will be handled, often does more for trust than the negative review does damage. Never argue publicly or dispute a real customer's experience.

Never buy, fake, or gate reviews. It violates Google's policies, risks the whole profile being suspended, and a steady trickle of five real reviews a month beats fifty fake ones from a service that disappears overnight, in every way that actually matters.

5. Get Listed and Linked Around the Web

Local rankings are also shaped by how consistently and credibly a business shows up beyond its own website and Google listing.

Claim and complete listings on the major directories. Local business directories, industry-specific listings, and map platforms beyond Google all count as evidence the business exists and is what it claims to be, as long as the details match everywhere.

Look for genuinely relevant local links. A mention from a local news site, a sponsorship of a community event, or a listing on a local business association's site all carry more weight than a paid link from an unrelated site ever will, and they're far less risky.

Partner with nearby, non-competing businesses. A florist and a wedding photographer, or a gym and a physiotherapist, can often link to and mention each other honestly, because the audiences genuinely overlap.

6. Track What's Actually Working, Every Month

Local SEO isn't a one-time setup. The businesses that keep improving are the ones that check a short, consistent set of numbers and adjust based on what they show.

Watch Business Profile insights. Calls, direction requests, and website clicks from the profile are a more honest signal of real demand than raw search impressions alone.

Check Search Console for what's actually being searched. It shows the real queries bringing traffic in, which often includes valuable searches nobody planned content around, and pages that are close to ranking but need a push.

Review, don't just publish and forget. Once a month, look at which pages are gaining ground and which have stalled, and spend the next month's effort on the ones most likely to move, rather than starting something new every time.

Common Mistakes That Quietly Undo All of This

• Using slightly different business name, address, or phone formats across listings
• Choosing a broad Business Profile category instead of the most specific accurate one
• Writing one page meant to rank for every service instead of separate, focused pages
• Letting the profile and website both go stale for months at a time
• Chasing backlinks from irrelevant sites instead of a handful of genuinely local ones
• Treating reviews as a one-time push instead of an ongoing, gentle habit

A Simple Monthly Checklist

• Add 2 to 4 new, real photos to the Business Profile
• Post one update and check for unanswered questions
• Ask the last 3 to 5 completed customers for a review
• Reply to every new review, good or bad
• Check Search Console for new or rising queries worth a page
• Confirm hours, contact details, and service area are still accurate everywhere

Frequently Asked Questions

How long does local SEO take to show results?
Most businesses see meaningful movement within 60 to 90 days of consistent work, though profile and review changes can influence rankings faster than new content takes to get indexed and settle in the results.

Do I need a blog to rank locally?
Not necessarily. Well-built service and location pages usually matter more than blog posts for local intent. A blog helps most when it answers real customer questions that don't fit naturally on a service page.

Is it worth paying for local SEO tools?
Not at first. Google Business Profile insights and Search Console are both free and cover most of what a small business actually needs to know. Paid tools become worth it once you're managing several locations or competing in a genuinely crowded market.

What Is Generative Engine Optimisation (GEO)? A Complete Guide

Written and structured following its own advice: a direct answer first in every section, then the reasoning behind it.

Why GEO Exists as Its Own Discipline

Search behaviour is splitting in two. A large share of searches still land on a traditional results page, where SEO's rules apply. But a growing share now happen inside a chat interface or an AI Overview, where the user never sees a list of pages at all, only a synthesised answer with a few citations underneath it. Ranking well in traditional search doesn't automatically mean being quoted in that answer, because the AI isn't ranking pages for a click. It's selecting a small number of trustworthy sources to summarise and picking exact phrases worth quoting. That's a different filter, and it rewards different things.

How an AI Answer Actually Gets Built

Most AI search tools work in two steps: they retrieve a handful of pages that seem relevant and trustworthy for the question, then generate an answer grounded in what those pages actually say. If a page never gets retrieved in the first step, it has no chance of being quoted in the second, no matter how well written it is. That's why GEO isn't only about writing style. It starts with the same groundwork as SEO, being genuinely relevant, well-structured, and findable, and then adds a second layer: writing in a way that's easy for the generation step to lift cleanly once the page has already earned a place in that shortlist.

What Actually Helps a Page Get Quoted

Lead every section with a self-contained answer. Put the direct answer to the heading's question in the first sentence beneath it, before any setup or context. An AI assistant summarising a page tends to lift the sentence that most cleanly answers the question, not the one buried three sentences into a windup.

Match headings to real questions. A heading phrased as an actual question people ask, like "How long does local SEO take to show results," gives an AI a near-perfect match between what it's looking for and what the page offers, far more reliably than a vague heading like "Timeline."

Use plain, specific language over marketing phrasing. "Reduces manual data entry by 80%" is quotable. "Streamlines your workflow" is not, because it carries no verifiable fact an AI can confidently repeat as true.

Back claims with real numbers, sources, or examples. AI systems are cautious about repeating unverifiable claims. A specific figure, a named example, or a cited source gives the model something concrete enough to trust and restate.

Build genuine topical depth, not a single lucky page. A site that thoroughly and consistently covers a subject across several well-linked pages reads as a credible authority. A single strong page surrounded by thin, unrelated content reads as a fluke, and AI systems weigh consistency heavily when deciding which sources to trust repeatedly.

The Technical Side: Making Content Easy to Read Correctly

Use clean, semantic HTML. Real heading tags, real lists, and real paragraph breaks make it far easier for any automated system, AI or otherwise, to parse where one idea ends and the next begins. A wall of unstructured text is harder to extract a clean quote from.

Add structured data where it fits. FAQ schema, article schema, and organisation schema all give machine-readable confirmation of what a page is and what it's answering, on top of what the visible text already says.

Make sure the page can actually be crawled and fetched. Content locked behind heavy client-side rendering, login walls, or blocked by robots.txt for AI crawlers simply never enters the pool of pages that could be cited, regardless of how good the writing is.

GEO and SEO, Side by Side

They share the same foundation. Genuine relevance, real expertise, accurate information, and a site Google already trusts enough to rank all carry over directly into whether an AI system trusts that same site enough to cite it.

They diverge in what "good" content looks like on the page. SEO content can build up to a point across several paragraphs and still rank well. GEO content that does the same often gets skipped in favour of a competitor that answered in the first sentence, because the AI is choosing what to extract and quote in the seconds it takes to generate a response, not rewarding a slow build-up the way a patient human reader might.

Common Mistakes That Keep Content Out of AI Answers

• Burying the direct answer under paragraphs of introduction and backstory
• Using vague, adjective-heavy marketing language instead of specific, checkable facts
• Writing one page in isolation instead of building genuine depth around a topic
• Phrasing headings generically instead of as the real question someone would ask
• Blocking AI crawlers entirely, then wondering why the content never gets cited
• Treating GEO as a one-time rewrite instead of an ongoing check against real AI answers

How to Actually Check If It's Working

You can't see GEO performance in a traditional rank tracker, because there's no ranked list to measure. The only reliable check is asking the real questions your content is meant to answer, directly in ChatGPT, Google AI Overviews, and Perplexity, and noting three things: whether your content shows up at all, how closely the phrasing matches what you actually wrote, and which competitor is being quoted instead when it doesn't. That last one is usually the most useful signal, since it shows exactly what a page needs to out-answer to earn the citation next time.

Frequently Asked Questions

Does GEO replace SEO?
No. GEO builds on the same foundation as SEO and adds a second layer for how content is written and structured. Dropping SEO fundamentals to focus only on GEO usually weakens both, since AI systems still lean on the same trust and relevance signals traditional search does.

Can you guarantee a citation in ChatGPT or AI Overviews?
No one can. These systems choose sources dynamically and the answer can change between two nearly identical questions. What's controllable is making a page as easy as possible to trust, extract from, and quote correctly when it is considered.

Is GEO only relevant for big brands?
No. Because AI systems favour a clear, specific, well-supported answer over sheer brand size, a small business that answers a niche question more precisely than a bigger competitor has a genuine, realistic shot at being the one quoted.

Fictional Product: "Ledgerly" (Invoicing Software for Freelancers)

A made-up product used purely to show copy range across platforms, angles, and character limits. Kept short on purpose: that's the actual skill being shown here.

Meta Ad, Pain-Point Angle

Meta Ad, Social Proof Angle

Google Search Ad (Responsive Search Ad Headlines)

Google Display Ad

TikTok Ad Script (15 Seconds)

Landing Page Hero

Landing Page Benefits Section

Turning Better Product Search Visibility Into a 90-Day Revenue Lift

Based on real work from my time at GrowthPulse Digital Agency. Client name and other identifying details are withheld under our client agreement.

Background

The client sold physical products online, in a category with several established competitors already running paid ads. They'd been running Facebook and Google ads for about a year through a previous freelancer, with results the client described as "fine, but expensive." Paid ads were bringing in sales, but every sale depended on an ad being live and being paid for. Nothing was compounding. When the budget paused for a week, so did the sales.

The Problem

The paid ads were converting reasonably well, but the products barely showed up when people searched for them directly on Google, by name, by category, or through Google Shopping. Anyone doing that kind of high-intent, ready-to-buy search was landing on a competitor's listing instead, which meant two things were both true at once: the client was paying to attract cold visitors through ads, while simultaneously losing warm, already-interested visitors to a competitor, for free, every single day.

Digging into why, the cause turned out to be unglamorous but common: the Google Shopping feed had vague, inconsistent product titles ("Item 4521" instead of the actual product name), several products sat in the wrong category, and the product pages themselves used marketing language that never matched the specific terms customers were actually searching for. None of it was broken exactly. It was just never built with search in mind.

The Approach

Rather than starting a separate SEO project, I treated the paid ads and the organic search presence as one connected system, since both were ultimately fighting for the same searches.

Rebuilt the Google Shopping feed properly. Every product got a clear, specific title matching real search terms, the correct category, and accurate, current images, replacing the vague listings that were live before.

Rewrote product pages around the same keywords driving the ads. If an ad was winning clicks for "waterproof running jacket," the page it landed on needed to say exactly that early and clearly, not describe it in vaguer brand language that happened to sound nicer.

Set up Facebook and Instagram shop listings to match. So the same consistent product information appeared everywhere a customer might encounter it, ads, shop tab, and organic search alike.

Shifted ad budget weekly toward what the data showed was actually working. Once accurate product-level data existed, budget stopped being split evenly and started following whichever products were both searched for and converting.

Obstacles Along the Way

The client's product catalog was large and had never been properly categorised, so the feed cleanup alone took longer than expected, closer to three weeks than the one originally planned. There was also an internal worry, understandable but ultimately unfounded, that rewriting product pages around plainer, more search-friendly language would make the brand sound "less premium." We tested that concern directly on a small batch of products first, rather than arguing about it, and the plainer, clearer pages converted at least as well as the original ones, which settled the question with data instead of opinion.

The Result

Within 90 days, the client saw a 62% increase in revenue. The bigger shift wasn't only the headline number, it was where that revenue started coming from. A meaningfully larger share of it was now people who'd searched for the product directly and landed on an accurate, well-matched page, rather than people who'd only ever seen a paid ad.

That mattered beyond the 90-day window. Free, high-intent search traffic doesn't disappear the moment a budget is paused, which meant the business was, for the first time, building something that kept working for it even in a slow ad week, instead of renting every single sale one click at a time.

What I'd Do Differently

I'd start the feed and product-page cleanup before touching ad budget at all, rather than running them in parallel. In hindsight, several early weeks of ad spend were still working against inconsistent product data, and fixing the foundation first would likely have shown a meaningful result sooner than the 90-day mark.