There is a particular kind of frustration that hits a business owner about three months after a landing page goes live. The page converts well. The paid traffic is working. Somebody suggests "we should get organic traffic to this too," and a check in Search Console returns one of two answers: the URL is not in the index at all, or it sits in a state called Crawled – currently not indexed, which reads like a technicality and is actually a verdict.
The page was not unlucky. It was built that way.
Almost everything that makes a landing page convert well makes it hard to index. Stripping the navigation removes distraction — and severs the page from the site's link graph. Keeping copy tight respects the visitor's attention — and produces a page with nothing to rank for. Running three headline variants improves conversion rate — and creates three near-identical URLs. Building the page in a dedicated tool ships it in an afternoon — and frequently publishes it on somebody else's domain.
None of those are mistakes in isolation. Each is a deliberate, defensible conversion decision. Taken together they produce a page that no search engine has any reason to index, and the sixteen articles you will find on this topic mostly do not mention it, because they are describing a page that ranks and quietly assuming it is the same page that converts.
This article is about the conflict. What actually stops landing pages indexing, which of those causes you can fix without hurting conversion, which require you to accept a genuine trade, and where the page should physically live — because that last decision turns out to govern more of the outcome than any on-page factor.
It is written for Canadian businesses running lead-generation or product landing pages, and for the people who build them.
One note on sources: Google's own SEO Starter Guide is the correct starting point for general on-page work and this article does not attempt to restate it. What that guide does not cover is landing pages specifically — the structural tension described above — which is the gap this piece occupies.
The two jobs, and why one URL rarely does both
A landing page can be built to do one of two things well. Businesses usually want both from one URL and are rarely told they are incompatible.
| Built to convert | Built to rank | |
| Navigation | Removed, to eliminate exits | Present, so the page sits in the link graph |
| Copy length | Short — decision-focused | Longer — covers the query properly |
| Internal links | Avoided, they leak attention | Necessary, they carry crawl signals and context |
| Variants | Several, tested continuously | One canonical URL |
| Entry point | Paid click, email, social | Search result |
| Visitor state | Already interested, pre-qualified | Cold, comparing, researching |
| Success measure | Conversion rate | Impressions, position, then conversion |
| Typical lifespan | Weeks to a campaign cycle | Years |
Read the last two rows together and the conflict becomes obvious. A paid visitor arrives already interested, so the page's job is to remove friction. An organic visitor arrives cold and comparing, so the page's job is to answer questions the paid visitor had answered before they clicked. Those are different pages for different people, and trying to serve both from one URL usually produces a page that converts worse than the paid version and ranks worse than the organic one.
The resolution is usually two pages, not a compromise. An organic page designed to rank, with navigation, depth and internal links, living permanently in your site structure. A paid page designed to convert, stripped down and iterated on, deliberately kept out of the index. Each doing one job properly. The section on canonicals and noindex below covers how to keep them from competing.
There are exceptions. A well-built service page — the kind that describes one offering thoroughly, answers the obvious objections, and ends with a form — genuinely does both jobs, and most businesses would be better served by more of those and fewer campaign-specific landing pages. But the exception is a page that was designed as a permanent site page from the start, not a campaign page retrofitted for search.
Why conversion-optimised pages fail to index
Six causes, in rough order of how often they are the actual explanation.
1. The page is an orphan
This is the big one and the one least understood.
Search engines discover pages by following links. A page in your sitemap with no internal links pointing at it is technically discoverable and practically invisible — it has no context, no signal of importance, and nothing indicating it belongs to your site's subject matter. The industry term is an orphan page.
Now recall the first decision in conversion-optimised landing page design: remove the navigation. In one stroke that removes the site-wide links out of the page, and because landing pages are usually built as standalone campaign assets, nothing links in either. The page is attached to your site by nothing but a sitemap entry and a shared domain.
An orphan page is the single most common profile behind Crawled – currently not indexed. Google found the URL, fetched it, evaluated it, and concluded there was no reason to keep it. That state is not an error and there is nothing to fix in the usual sense — it is a judgement about value, and the absence of any internal link is one of the clearest signals of low value a site can send about its own page.
The fix that does not hurt conversion: links in are what matter, and they cost you nothing. Link to the page from relevant blog posts, from a related service page, from a resources index, from your footer if it is a permanent offer. You can leave the page's own navigation stripped and still fix the orphan problem entirely, because the problem was inbound links, not outbound ones. This is also the argument for maintaining the kind of topical content that can carry those links — our piece on whether blogging helps SEO covers that side.
2. The page is thin
A landing page with a headline, three bullets, a testimonial and a form contains perhaps 120 words. There is nothing there to match a query against. Not because of a word count threshold — there isn't one — but because the page does not demonstrably answer anything.
This one is a genuine trade, and it is worth being honest about it rather than pretending a magic layout solves it. You can add depth below the fold, which most conversion testing finds costs little because the people who were going to convert immediately already have, and you can use progressive disclosure — expandable FAQ blocks, specification tables — so the content is present in the HTML without cluttering the visual design. Both help. Neither turns a 120-word page into a page that ranks for a competitive commercial query.
3. Variants have created duplicates
A page with three headline variants at three URLs is three pages Google must choose between, and its most likely conclusion is that two or all three are duplicates not worth indexing.
Google's own guidance on split testing is specific and easy to follow: put a rel="canonical" on each variant pointing at the original URL, so the test does not fragment your signals. If you are running a redirect test, use a 302 rather than a 301, because a 301 tells Google the move is permanent and you are going to take the variant down next month. Run tests for a defined period and clean up afterwards — a test left running for two years is not a test, it is a duplicate content problem with a reason attached.
4. Parameters have multiplied the URL
Paid traffic arrives with tracking parameters. Every unique combination is technically a distinct URL, and a page that has accumulated hundreds of parameterised variants in the wild can present as a mess of near-duplicates.
The fix is a self-referencing canonical on the clean URL. It is one line, it is the default in most content management systems, and it is missing more often than you would expect on pages built outside the CMS.
5. Something is telling Google not to index it
Worth checking before anything else, because it is common and embarrassing.
A noindex robots meta tag left over from staging. A robots.txt rule blocking the campaign directory. A canonical pointing at a different page — often the homepage, because a template set it. A page behind a login or a cookie wall. On landing page builders specifically, a default "hide from search engines" setting that was on when the page was created.
Search Console's URL Inspection tool reports exactly which of these applies, and it takes about fifteen seconds. Do this first.
6. The page is not on your domain
This one deserves its own section, because it is the decision with the largest consequences and the least discussion.
Where the page lives: the decision that governs everything else
Landing page builders are genuinely good products. They ship pages fast, they make testing easy, and they remove a dependency on a developer. The question is not whether to use one; it is what address the resulting page has.
| Placement | Example | What it means for search |
| Your own subdirectory | yourbusiness.ca/offers/spring/ | Part of your site. Inherits and contributes to your domain's standing. Fully in your control. |
| Your own subdomain | offers.yourbusiness.ca | Yours, and indexable, but Google may treat it as a somewhat separate property. Workable; adds a variable you did not need. |
| Builder's domain | yourbusiness.builderpages.com | Not your site. Any authority the page earns accrues to the builder's domain, not yours. |
The third row is the one that matters, and a surprising number of businesses are in it without knowing.
A page at yourbusiness.builderpages.com is, from a search engine's perspective, a page on the builder's website. Links people give it point at the builder's domain. It cannot inherit your domain's standing and cannot contribute to it. If you stop paying the subscription, the URL stops existing and anything pointing at it breaks. For a two-week campaign page nobody will ever link to, that is a reasonable trade for convenience. For a permanent offer you hope will earn organic traffic, you have spent budget building an asset on rented land.
Subdirectory versus subdomain is a smaller question than SEO discussion generally implies. Google is capable of understanding that a subdomain belongs to the same organisation, and plenty of large sites run subdomains successfully. But "capable of understanding" is not the same as "treats identically," and for a small business site with modest authority, keeping everything in one place under one domain removes a variable for no cost. If there is no operational reason for a subdomain, use a subdirectory.
Most good landing page builders support publishing to a subdirectory of your own domain — usually through a reverse proxy or a CMS plugin — and this is worth the configuration effort. It is the difference between a page that is part of your site and a page that merely mentions your name.
Where hosting actually enters the picture
This is the honest version of the hosting-and-landing-page connection, and it is more limited than hosting marketing suggests.
Your host does not make a landing page rank. What your host determines is whether the page can be served from your own domain at all, whether server response time puts a floor under the page's loading metrics, and whether the page stays available during the traffic burst a campaign produces.
That last point is the one businesses discover the hard way. A landing page is the one page on your site guaranteed to receive concentrated traffic — that is what a campaign is. If the page is server-rendered on a plan with limited concurrency, the ad spend that was supposed to produce leads produces timeouts instead. A well-cached static landing page is extremely cheap to serve; a database-driven page with a form and personalisation is not. Knowing which one you have built matters before you buy the traffic.
The page speed side of this is a real factor in page experience but it is a floor, not a lever: a fast server cannot rescue a page carrying six tracking scripts and an uncompressed hero image. We cover that relationship properly in our guides to hosting and SEO and Core Web Vitals and hosting, and there is no value in restating it here.
Where Canadian hosting for SEO makes a defensible difference is narrower and worth stating precisely. For a business whose customers are in Canada, a Canadian origin reduces server response time for the real visitors whose experience feeds Google's field measurements, and it keeps form submissions — which on a lead-generation page means personal information — inside Canadian jurisdiction. Neither of those is a ranking lever. Both are real. That is the honest case for web hosting in Canada on a lead-generation page: not a ranking advantage, but a shorter path for the visitors Google actually measures and a simpler answer about where the contact details went. Providers running Canadian facilities — 4GoodHosting among them, from Vancouver and Toronto — close that second question at the infrastructure layer. And if you are running paid traffic to a page hosted offshore while collecting Canadian customers' contact details on it, the privacy question is live regardless of the search question. Our Canadian data centres page and our guide to PIPEDA and Canadian data residency cover that side.
A/B testing without creating a duplicate content problem
Short section, because the rules are simple and widely broken.
- Canonical every variant to the original. One line per variant. This is Google's own stated guidance and it resolves the duplication question completely.
- Use 302, not 301, for redirect tests. A 301 declares the move permanent. You are going to reverse it.
- Do not serve different content to crawlers than to users. Cloaking is a spam policy violation, and "we were just testing" is not a defence. Whatever proportion of traffic sees a variant, crawlers should be in the same pool.
- End the test. Implement the winner at the canonical URL and remove the variants. Tests that never end become permanent duplicate URLs.
- Do not test the URL structure itself on a page you want to rank. Changing the address of a page that has accumulated signals in order to test a headline is a poor trade.
The doorway page line, and how to stay on the right side of it
Here is where a very common Canadian small business tactic runs into Google's spam policies, and it is worth being direct about it.
The tactic: build a landing page for each city or service area you serve. Accountant in Mississauga, Accountant in Brampton, Accountant in Oakville, and so on, often twenty or forty of them, generated from a template with the city name swapped in.
Google's spam policies name this pattern explicitly. Doorway pages are pages or sites created to rank for specific queries that funnel visitors to the same destination without offering distinct value. The characteristics named include multiple pages where the content varies only by the location inserted, and pages that exist to capture query variations rather than to serve a visitor.
The practical consequence is usually not a penalty. It is simpler and more common: the pages do not index. Google crawls them, recognises them as near-duplicates of each other, picks one or none, and discards the rest. The business concludes that local SEO does not work, when what actually happened is that it built forty pages that say the same thing.
The honest test
One question distinguishes a legitimate location page from a doorway page: could a visitor in that city tell this page was written for them, and would it be wrong if you swapped the city name?
A legitimate Mississauga page names the neighbourhoods served, references the local context that actually differs — municipal requirements, the specific local market, travel and service logistics — includes staff, projects or clients in that area, and would read as inaccurate with a different city substituted. A doorway page would read perfectly well with any city name in it, which is exactly the problem.
The corollary is uncomfortable but useful: most businesses should have far fewer location pages than they think, each substantially better. Three genuinely differentiated city pages will outperform forty templated ones, and they will index. Our local SEO in Edmonton guide works through what genuine local differentiation looks like.
The same logic applies to service-variant pages. If Emergency Plumbing, 24 Hour Plumbing and Urgent Plumbing are three pages describing one service, they are competing with each other for one query and Google will resolve that competition by indexing at most one.
The technical checklist that actually applies to landing pages
The generic on-page checklist is covered well by Google's starter guide and does not need repeating. These are the items specific to landing pages, where the usual advice does not reach.
| Check | Why it matters on a landing page specifically |
| At least two internal links pointing in | Fixes the orphan problem, which stripped navigation created |
| Included in the XML sitemap | Necessary but not sufficient; a sitemap is not a substitute for links |
| Self-referencing canonical on the clean URL | Absorbs the parameterised variants paid traffic generates |
| Variant canonicals pointing at the original | Prevents test variants competing with the page |
| No stray noindex from staging or builder defaults | The most common single cause, and the fastest to check |
| Served from your own domain, ideally a subdirectory | Determines whether the page is part of your site at all |
| Crawlable without JavaScript execution for primary content | Builder-generated pages sometimes render the offer client-side |
| One clear H1 matching the search intent, not the ad copy | Ad headlines are written for a pre-qualified audience |
| Breadcrumbs, if the page sits in a category | Gives the page a position in the structure it was stripped out of |
| Appropriate structured data | Product, Service, FAQPage or LocalBusiness as the page actually warrants |
| Distinct title and meta description | Templated location pages routinely share both, which is a duplication signal in itself |
| Form does not block content rendering | A page whose content appears only after a form interaction has little to index |
Two notes on structured data, because it is frequently over-applied on landing pages. Mark up what is genuinely on the page and visible to the visitor. FAQPage markup on questions that are not actually on the page, or Review markup a business has written about itself, are misuses that produce no benefit and some risk.
Content that can do both jobs
If you do decide one URL must serve both audiences, the content has to be structured for two readers rather than averaged between them.
Match the intent, not the campaign. Your ad headline speaks to someone who has already seen three touchpoints. Your H1 and opening paragraph have to work for someone who typed a query and knows nothing about you. Those are usually different sentences, and the organic one has to come first.
Answer the question above the fold. If the query is "commercial snow removal Winnipeg cost," the page should establish what it costs and how pricing works near the top. Burying that behind a form is the single most reliable way to lose both the ranking and the visitor.
Put the depth below the conversion point. Visitors ready to act convert in the first screen. Everyone else is still deciding, and the material that helps them decide — specifications, comparisons, objections, process, proof — belongs below, where it adds indexable substance without obstructing the people who were already convinced.
Use expandable sections for genuine depth. Content inside collapsed accordions is present in the HTML and indexable. It is a legitimate way to carry 1,200 words on a page that visually reads as 300.
Include real proof. Named clients, specific numbers, actual photographs, a verifiable address. This is what a templated page structurally cannot have, which is what makes it the strongest differentiator available.
How to diagnose your own landing page
Five steps, in order, using tools you already have.
- Run URL Inspection in Search Console. Paste the exact URL. It will tell you whether the page is indexed, whether Google has a different canonical in mind, whether a crawl was blocked, and what the last crawl found. This answers most cases outright.
- Read the coverage state literally. Crawled – currently not indexed means Google evaluated it and declined. Discovered – currently not indexed means it has not been crawled yet, often a crawl priority signal. Duplicate, Google chose a different canonical names the page it preferred, which is usually the clue you needed. Excluded by noindex tag is self-explanatory and means stop reading and go fix that.
- Count the internal links pointing at the page. Site search, a crawler, or Search Console's internal links report. If the answer is zero, you have found your problem and it is cheap to fix.
- Check what the page looks like with JavaScript disabled. If the offer, the headline and the body copy disappear, you are asking Google to index an empty container.
- Compare the page against the query you want. Read it as a stranger who typed that query. If it does not visibly answer the question, no technical fix will make it rank, and the work is editorial rather than technical.
Canadian specifics for lead-capture pages
Four things that apply here and are absent from every one of the reference articles.
Consent on the form. A landing page form that adds people to a marketing list is a consent event under Canada's Anti-Spam Legislation. Express consent must be knowingly given — which means the consent request has to be separate and clear, not buried in terms, and not pre-checked. Sending a subsequent marketing email to someone who filled in a quote form without having consented to marketing is where businesses get into difficulty. The distinction matters practically: a reply about the quote they requested is a different category from a newsletter. Our guide to building an email list under CASL covers the mechanics.
The form data is personal information. Name, email, phone, and often a description of the visitor's problem. PIPEDA obligations attach the moment you collect it — purpose, retention, safeguards, access. A landing page built in a third-party tool means that tool now holds it, which is a service provider relationship you should be able to describe if asked. Where the data sits geographically is part of that description.
French, if you are targeting Quebec. Quebec's Charter of the French Language imposes requirements on commercial communications directed at Quebec consumers, and these were expanded by the 2022 reforms. A landing page running paid traffic into Quebec is commercial communication. In search terms, the correct implementation is a genuine French page at its own URL with reciprocal hreflang annotations between the two versions — not machine translation on the same URL, and not a language switcher that changes content without changing the address. The legal requirements here are specific and a business targeting Quebec should get its own position confirmed rather than relying on a summary.
Local signals for a national business. If you serve all of Canada from one location, a .ca domain, a real address, and consistent business information do more geotargeting work than any amount of city-name insertion. That is also the honest answer to "should we build a page for every province": almost certainly not.
When to bring in outside help, and how to evaluate it
The supplied brief asked for a section on SEO agencies. Here is the version that is actually useful, which is a set of criteria rather than a recommendation.
Nobody can honestly claim to be the best SEO agency in Canada, and a provider who does has told you something about how they handle evidence. What a business can do is evaluate a provider on things that are checkable.
Ask what they will measure and when. A credible answer names specific metrics, a baseline, and a realistic timeframe. Organic results from technical and content work typically take three to six months to appear, and an agency promising rankings in weeks is describing something else.
Ask them to diagnose before they propose. Any provider should be able to tell you why your current pages are not indexing before telling you what they would build. If the proposal is identical to what they would have proposed for a different business, they have not looked.
Ask who owns what. The domain, the hosting account, the analytics property, the Google Business Profile, the content. Agencies that build on their own accounts create an exit cost, and you should know about it at the start rather than at the end.
Ask about the tactics they will not use. A provider who volunteers that they will not build forty templated city pages, buy links, or generate bulk content has demonstrated that they know where the lines are.
Be wary of guaranteed rankings. Nobody controls Google's results. A guarantee is either meaningless or is being met through methods you would not approve of.
Check whether the advice survives contact with your own site. Run the five diagnostic steps above yourself first. You will then know enough to tell whether a proposal is addressing your actual problem, which is the most useful position from which to hire anyone.
If you want that diagnosis done rather than done to you, our own professional SEO support works from exactly this sort of audit — and you should hold it to the criteria above like any other provider.
Common mistakes
Assuming the sitemap is enough. A sitemap tells Google a URL exists. Internal links tell Google it matters.
Building on the builder's domain. Convenient, and the authority accrues to them.
Keeping A/B variants live forever. A test with no end date is a duplicate content problem.
Using the ad headline as the H1. Written for a warm audience, indexed for a cold one.
Building forty city pages from one template. They will not index, and the conclusion drawn will be the wrong one.
Putting the answer behind the form. Loses the ranking and the visitor.
Leaving `noindex` on from staging. Check it first, every time. It costs fifteen seconds.
Marking up FAQs that are not on the page. No benefit, some risk.
Buying traffic before load-testing the page. A campaign is a concentrated traffic event aimed at your least-cached page.
Frequently asked questions
Why is my landing page not indexed by Google?
The most common cause is that nothing on your site links to it. Conversion-optimised landing pages usually have the navigation stripped out and are built as standalone campaign assets, so they end up as orphan pages — present in the sitemap, connected to nothing. After that: the page is too thin to answer a query, a noindex tag was left on from staging or by a builder default, a canonical points somewhere else, or the page is on the builder's domain rather than yours. Check Search Console's URL Inspection tool first; it names the cause in most cases.
What does "Crawled – currently not indexed" actually mean?
Google found the URL, fetched it, evaluated it, and decided not to index it. It is not an error and there is no setting to change. It is a judgement about value, and for a landing page the usual drivers are an absence of internal links, insufficient substance, or near-duplication with another page on your site.
Should landing pages be set to noindex?
Campaign-specific pages that exist only to receive paid traffic, and A/B test variants, are reasonable candidates — though canonicalising variants to the original is generally better than noindexing them. Permanent pages describing a service or offer should be indexable. The deciding question is whether the page will still exist in a year.
Does removing the navigation hurt SEO?
It removes the page's outbound site links and, more importantly, usually means nothing links in either. The fix is not to restore the navigation — it is to add two or three contextual links pointing at the page from elsewhere on your site. That resolves the orphan problem without changing the page's design at all.
Subdomain or subdirectory for landing pages?
A subdirectory on your own domain, unless there is an operational reason not to. Google can handle subdomains, but a subdirectory removes a variable for no cost. The distinction that genuinely matters is neither of these — it is whether the page is on your domain at all, rather than on a landing page builder's.
How do I run A/B tests without a duplicate content problem?
Put a rel="canonical" on each variant pointing at the original URL. Use a 302 rather than a 301 for redirect tests, since the change is temporary. Do not serve crawlers different content from users. End the test and implement the winner at the canonical URL.
Are city-specific landing pages against Google's guidelines?
Templated ones that differ only by the place name fall within what Google's spam policies describe as doorway pages. In practice the usual outcome is not a penalty but a failure to index — Google recognises them as near-duplicates and keeps at most one. The test: would the page be wrong if you swapped the city name? If not, it is a doorway page. Genuinely differentiated location pages, with local specifics a visitor would recognise, are fine and useful.
Can one page rank and convert?
Sometimes, and it takes deliberate structure: the organic answer above the fold, the conversion point next, and the depth below it. But a well-built permanent service page does this far more reliably than a campaign page retrofitted for search. For most businesses the honest answer is two pages — one built to rank, one built to convert — with the campaign page kept out of the index.
Does hosting affect whether a landing page ranks?
Not directly. What hosting determines is whether the page can be served from your own domain, what floor server response time puts under your loading metrics, and whether the page stays up during the traffic burst a campaign creates. That last one is the practical risk: a landing page is the page on your site most certain to receive concentrated traffic.
How long should a landing page be?
There is no threshold. The question is whether the page answers the query it is targeting, which for a competitive commercial term usually takes more than a headline and three bullets. Add depth below the conversion point, where it costs nothing from the people who were going to convert immediately.
Key takeaways
- Conversion optimisation and indexation pull in opposite directions. Stripped navigation, short copy, multiple variants and third-party hosting are all deliberate conversion decisions that cost you the organic listing.
- The most common reason a landing page does not index is that nothing links to it. A sitemap entry is discovery, not endorsement.
- Fixing the orphan problem requires inbound internal links, not restored navigation — so it costs nothing in conversion terms.
- Crawled – currently not indexed is a judgement about value, not an error with a setting to change.
- Where the page lives governs more than any on-page factor. A page on a builder's domain builds the builder's authority, not yours. Publish to a subdirectory of your own domain.
- Canonical every A/B variant to the original, use 302 for redirect tests, and end tests rather than leaving them running.
- Templated city pages that differ only by place name are what Google's spam policies describe as doorway pages. The usual penalty is simply that none of them indexes.
- Three genuinely differentiated location pages beat forty templated ones, and they index.
- Check for a stray noindex before anything else. It is a fifteen-second check and a frequent cause.
- A lead-capture form is a CASL consent event and a PIPEDA collection. Separate, unchecked, clearly worded consent is not optional.
- Quebec's language requirements apply to commercial communications aimed at Quebec consumers; the correct search implementation is a genuine French page at its own URL with reciprocal hreflang.
- No provider can substantiate being the best SEO agency in Canada. Evaluate on what is checkable: diagnosis before proposal, named metrics and timeframes, account ownership, and the tactics they refuse to use.
Conclusion
The reason landing page SEO advice is so uniformly unhelpful is that it assumes away the actual problem. Sixteen articles will tell you to write a compelling title tag and compress your images, and all of that is true, and none of it addresses a page that Google has already looked at and decided not to keep.
The real question is structural. Is this page part of your website, or a standalone artefact that shares your brand name? Does anything on your site point at it? Is there enough on it to answer the query you have in mind? Is it one page or three near-identical ones? Is it at an address you control?
Those five questions account for most of the difference between landing pages that earn organic traffic and landing pages that only ever consume ad budget. They are also all answerable in an afternoon, using Search Console and a look at your own link structure, before spending anything on content or on help.
And the most useful conclusion is often the uncomfortable one: that you do not need a better landing page, you need fewer of them. One page that genuinely answers a query, properly linked from the rest of your site, hosted on your own domain, with real proof on it, will do more than a dozen templated variations — and it will still be working in three years, which the campaign pages will not.
Before changing anything, run the five diagnostic steps in this article on your most important landing page: URL Inspection, the coverage state, a count of internal links pointing in, a look at the page with JavaScript disabled, and an honest read against the query you want it to answer. Most pages fail on the third step, and that one is free to fix.
If the diagnosis points at infrastructure — a page you cannot publish to your own domain, a server that will not hold up under campaign traffic, or form data sitting outside Canada — that is worth resolving before the content work, because no amount of on-page optimisation fixes an address. 4GoodHosting runs Canadian infrastructure from data centres in Vancouver and Toronto, so pages and the data they collect stay on your own domain and under Canadian jurisdiction. Our Canadian web hosting range covers the tiers that matter for campaign traffic, and if you would rather have the diagnosis done properly, our professional SEO support starts from the audit described above rather than from a package.
