An SEO strategy is the set of decisions you make before you write anything: which searches are worth competing for, which page on your site answers each one, and how those pages fit together. Get that right and ordinary content ranks. Get it wrong and excellent content sits on page four, answering a question nobody asked or competing with a page you already own.
This is the part most businesses skip. They start producing (a blog post here, a service page there) and end up with a site full of pages that overlap each other and target terms with no commercial value. Strategy is cheaper than content, and it’s what decides whether the content pays.
Two neighbouring pieces have their own guides. Turning these decisions into scheduled work with a review rhythm and KPIs is SEO campaign management. And if a meaningful share of your customers are nearby, the geographic layer (profile, reviews, map pack) runs on different rules, which we cover in local SEO tactics for ranking in your area.
The Bottom Line
- Sort keywords by intent before volume. A term with 90 searches and buying intent beats one with 9,000 and none.
- One intent, one page. Two pages chasing the same search compete with each other and split the signal.
- Position one averages about 20% of organic clicks, and roughly 9.7% when an AI Overview sits above it (Advanced Web Ranking, 2026), so plan for visibility without a click.
- Fix what you already have first. Pages Google already trusts move faster than new ones.
- Strategy is a quarterly decision, not a permanent document. Re-check it when rankings, demand, or your services change.
Lower your delivery costs by 23%
How we reduce costs:
- No delivery vehicle expenses
- Optimized local routes
- Pay-per-delivery model
- Average 23% delivery cost reduction
Start with what you sell, not with keywords
The first step isn’t a keyword tool. It’s a list of the things you make money on, in order of margin.
Write down your service lines or product categories, the ones you want more of, and the customer who buys each. That list is the frame every keyword decision hangs on, because it tells you which searches lead somewhere. A wholesale bakery that wants more standing weekly accounts should not be optimizing for home baking recipes, however much search volume sits there.
Do the same for the customers you don’t want. Terms that bring in enquiries outside your delivery radius, below your minimum order, or for services you’re phasing out are a cost, not a win. You pay for them in time spent saying no.
How to choose keywords by search intent
Search intent is what the person wants to happen next, and it determines both whether a search is worth pursuing and what kind of page has any chance of ranking.
Four intents cover almost everything, and each maps to a different page type:
| Intent | What the searcher wants | Example | Page that ranks |
|---|---|---|---|
| Transactional | To buy or book now | “order corporate catering downtown” | Service or product page |
| Commercial | To compare before buying | “best wholesale bread supplier” | Comparison, pricing, or category page |
| Informational | To understand something | “how much does catering cost per person” | Guide or blog post |
| Navigational | To reach a specific brand | “rosewood bakery delivery hours” | Homepage or contact page |
Priority order for most businesses: transactional first, commercial second, informational third. That’s the reverse of how most content calendars get built, which is why so many blogs generate traffic and no revenue.
To check intent, search the term and look at what Google is already rewarding. If the top ten are all product pages, a blog post will not break in, and vice versa. Google’s own guidance is blunt about this: match the format the query is asking for (Google Search Central).
Volume is the last filter, not the first. For a business with a delivery radius, a term with 50 local searches a month can be worth more than a national term with 5,000, because you can serve one and not the other.
Map one keyword to one page
Assign every target keyword to exactly one URL, and write it down. A spreadsheet with keyword, intent, target URL, and current ranking is the whole artifact. This document is your strategy.
The rule it enforces: one intent, one page. When two pages target the same search, they compete with each other, Google picks one arbitrarily, and neither ranks as well as a single consolidated page would. This is the most common self-inflicted SEO problem, and it accumulates quietly as a site grows.
Where you already have two pages on one intent, choose the stronger, usually the one with more existing traffic and links, and make it the destination. Fold the useful content from the other into it rather than leaving both up to fight.
The mapping exercise also surfaces the gaps. Any high-priority commercial term with no page assigned is your next piece of work, and it’s a far better brief than “write a blog post about catering.”
Structure the site so the important pages are close to the top
Site structure is strategy made physical. The pages that matter most should be reachable in one or two clicks from the homepage, and the ones that support them should link up to them.
A workable shape for most delivery businesses: homepage, then a small number of service or category pages, then supporting pages beneath each: locations, use cases, and guides. Supporting content links up to the commercial page it supports; the commercial page links down to the best of it.
Two structural rules pay for themselves. Keep URLs short and descriptive, and don’t reorganize them without redirects. And make sure nothing important is orphaned. A page with no internal links pointing at it is a page you’ve hidden from both Google and your customers.
Decide how much on-page and technical work the strategy needs
On-page work is the fastest-returning lane in SEO because you’re improving pages Google already knows. For each mapped page: the target term in the title tag and H1, a description that earns the click, headings that reflect what the sections actually cover, and internal links from related pages using descriptive anchor text.
Technical work is mostly a threshold, not a competition. Your site needs to be crawlable, fast enough, readable on a phone, free of duplicate and broken pages, and marked up with basic structured data. Past that threshold, more technical work returns very little for a small site, which is worth knowing before someone sells you an audit with 400 findings.
If you’re weighing where to put limited hours: fix and re-optimize your existing commercial pages before writing anything new. The uplift arrives in weeks rather than quarters.
Plan for searches that don’t produce a click
The click-through economics of ranking have shifted, and a strategy written in 2022 assumptions will overpromise.
AI Overviews sit above organic results on a growing share of queries and absorb a meaningful share of the clicks. Tracking through 2026 showed desktop click-through on the top organic result dropping sharply, from around 45.6% in February to 14.6% by May on the SERPs measured, while mobile moved in the other direction (Seer Interactive, 2026). The numbers vary by query type and by study, but the direction is consistent.
Two implications for strategy. First, informational content has a weaker business case than it used to, because that’s where AI answers intercept most cleanly, which is another argument for weighting commercial and transactional terms. Second, being the source an AI answer draws on has value even without the click, and the way to be that source is the same as always: a clear, specific answer near the top of the page, attributable and easy to extract.
Pick your authority strategy honestly
Links and mentions from other sites remain a real ranking input, and they’re the lane where small businesses most often waste money.
Before you spend anything, get a baseline. A free website authority checker will tell you roughly where you stand against the competitors you’re trying to outrank, which is the difference between “we need links” and “we need to stop targeting terms four tiers above us.”
The version that works for an operating business is unglamorous: suppliers and partners who’ll link to you, your local press and neighborhood publications, industry associations and directories in your category, and the trade bodies your customers read. Earned mentions from businesses in your actual network beat anything you can buy, and they carry local relevance as a bonus.
If you do buy help, buy placement on publications with actual readers rather than links by the dozen. That’s the honest use of a guest posting service: access to editorial slots you couldn’t pitch yourself. Directory and listing consistency is a separate, cheaper job, and one you can hand to a local citation building service if the audit turns up a mess; it carries less ranking weight than links do, and the full picture of how it fits the local signals is in our local SEO guide.
What not to do: buy link packages priced per link, chase sites that exist only to host guest posts, or pursue links from irrelevant high-authority domains. These range from ineffective to actively risky, and they consume budget that would do more work in content.
One check worth building into the process if you outsource writing: run anything you didn’t write through a plagiarism checker before it goes live. Duplicated copy won’t rank, and on a small site one recycled page can undermine the credibility of everything around it.
How to tell whether the strategy is working
Check the strategy quarterly against three questions.
Are the mapped pages moving up for their assigned terms? Not all of them, but the trend across the set. Are the commercial pages earning clicks, not just the guides? And are those clicks producing inquiries or orders?
If rankings are improving and revenue isn’t, the mapping is wrong: you’re winning searches with no buying intent behind them. If nothing is moving at all after two quarters, the terms are probably too competitive for your current authority, and the fix is to drop down to longer, more specific variants and build up from there.
Setting those checks on a schedule, with a fixed KPI set and a baseline to compare against, is where the campaign management side takes over.
Frequently asked questions
What’s the difference between an SEO strategy and an SEO campaign?
The strategy is the set of decisions: which keywords, which pages, what structure. The campaign is the scheduled execution and measurement of those decisions. Strategy answers what to pursue; campaign management answers when the work happens and how you know it worked.
How many keywords should an SEO strategy target?
For a small business, 15 to 30 mapped terms is plenty to start, weighted toward commercial and transactional intent. A longer list mostly produces a longer list; the constraint is how many pages you can make good.
Is blogging still worth it for SEO in 2026?
Yes, but as support rather than as the main engine. Informational content is the most exposed to AI answers intercepting the click, so its job is increasingly to build topical depth and link internally to the pages that convert, rather than carry the traffic target by itself.
How often should I revisit my SEO strategy?
Review the keyword map quarterly and rebuild it annually, or whenever you add a service line, change your delivery area, or see a sustained ranking shift you can’t explain.
Putting it into practice
Open a spreadsheet and list your revenue lines. For each, write the two or three searches a ready-to-buy customer would type, then find the page on your site that should own each one. Note where there isn’t one.
What you’ll have is a short, ordered list of pages to fix and pages to build: a strategy derived from what you sell rather than from a keyword tool’s export. Work it in priority order, fix the existing pages before building new ones, and review the map every quarter.
