SEO for IT companies and MSPs: the pages, proof and tracking that bring leads

Why SEO for IT companies works differently

A homeowner with a burst pipe searches once and calls the first plumber who answers. An IT buyer researches in stages, usually with colleagues, and often would rather not speak to a salesperson at all. Gartner reports that 75% of B2B buyers prefer a sales experience without a sales rep. It describes B2B buying as a set of "buying jobs" rather than a straight funnel: problem identification, solution exploration, requirements building and supplier selection. Buyers move between them in no fixed order, and most return to at least one of them before they purchase.

Three consequences follow for a technology services website:

• The site has to hold the first conversations. Scope, stack, process, security posture and how an engagement starts belong on the page, because many buyers decide whether to shortlist you before they speak to anyone.
• Queries are specific, and many are rare. Buyers search for products by name (Microsoft 365, Intune, Azure, AWS, Kubernetes, Snowflake), for error messages, for migration paths and for compliance terms. A query that almost nobody else types can still lead to a contract.
• Several people read the same site. An IT manager checks the technical detail, a finance lead checks the commercial model and an executive checks who you are. Each of them needs a page that answers their question.

Map search intent to the pages buyers need

Google's Search Quality Rater Guidelines (the September 2025 edition) describe four kinds of search intent: Know (find information or explore a topic), Do (accomplish a goal), Website (reach a specific site or page) and Visit-in-person (find a business or a category of businesses). Put them next to Gartner's buying jobs, and the page types an IT services site needs follow:

One page answers one intent. Two pages chasing the same query compete with each other, and a page that tries to serve every stage serves none of them well. Before anything is written, record the query, the buying job and the page that will answer it.

Build one page per service

A single "Services" page that lists ten offerings in ten paragraphs cannot rank for ten different searches, and it cannot answer a requirements question about any of them. Give each service its own URL, title and main heading, and make a services hub page that links to all of them.

A service page that works for search and for a buying committee covers seven things:

1. Who it is for and the problem it solves, in the buyer's words.
2. What is included, and what is not. A clear exclusion earns more trust than a longer list.
3. The stack by name: the platforms, tools and vendors you actually work with. Buyers search by product name, so this is also where much of the page's vocabulary comes from.
4. How an engagement runs: the first step, what the client provides, what they receive and when.
5. The commercial model: fixed fee, retainer, per user or per device, even when prices are agreed per client.
6. Proof: links to the case studies and articles for this service.
7. The questions buyers ask, answered plainly, and a clear way to get in touch.

Titles deserve more care than most teams give them, because Google notes that the title link is often the main thing people use to decide which result to click. Its guidance on title links asks for descriptive, concise titles, no keyword stuffing, no boilerplate repeated across pages and the brand added concisely. Managed IT services for small offices | Example IT does that. IT Solutions | IT Company | Best IT Services | Example IT repeats itself on every page and says nothing about this one.

Add industry pages only where you have real depth

Industry pages ("IT support for law firms", "cloud hosting for healthcare") answer requirements-building searches, and they persuade when they show that you know how that industry works. A useful industry page names the software the industry runs on (practice management systems, electronic health records, point-of-sale), the rules it works under and what they mean for the work, the constraints that change delivery (after-hours change windows, data residency, audit evidence) and the case studies you have in that field.

Get the rules right, because buyers in regulated industries check. The US HIPAA Security Rule sets national standards for electronic health information created, received, used or maintained by a covered entity or its business associate. PCI DSS applies to every entity that stores, processes or transmits cardholder data, and to those that could affect the security of the environment where it lives, which includes service providers as well as merchants.

The test for an industry page is simple: replace the industry name with another one. If the page still reads correctly, it is a template, not an industry page, and a set of them sits close to what Google's spam policies call doorway abuse. Build pages for the few sectors where you have clients and cases. List only the certifications, audits and partner statuses the company actually holds, and link to where a buyer can check them.

Create location pages only where you have a real presence

Location pages answer supplier-selection searches that include a place, and they matter most to MSPs that send engineers on site. They work when an office stands behind them: an address, people who work there, the services delivered from it and the clients it supports. Google's Business Profile guidelines set the same bar for the map listing. A virtual office (a mailing address you do not operate from) is not eligible, and an office in a co-working space counts only if it has clear signage, is staffed by your team and receives customers during business hours.

Pages for towns where you have no office are what Google's spam policies describe as doorway abuse: pages targeted at specific regions or cities that funnel users to one page, or substantially similar pages that sit closer to search results than to a browsable site structure. The keyword stuffing policy names the other common shortcut, a block of text listing the cities and regions a page is trying to rank for.

For the Business Profile, consistent name, address and phone number, reviews and LocalBusiness markup on each office page, see the local SEO guide.

Write case studies that are specific without naming the client

Case studies answer the question every shortlist asks: have you done this before? Many IT clients do not allow their names to be published, and a case study can still convince without the name if it is specific about the work. Keep the client's profile generic and the solution detailed:

This is also what search engines look for. Google's helpful content guidance says it helps readers to know how a piece of content was produced, and gives the example of showing how tests were done, with evidence of the work involved. Its rater guidelines tell raters to judge experience from the main content itself, as well as from what the business says about itself and what independent sources say about it. An architecture and a sequence of decisions are that evidence; adjectives are not.

Check the contract's confidentiality clause before you publish, and send the draft to the client even when they are unnamed. The details that make a case study convincing are also the ones a client recognizes.

Publish technical articles that answer real problems

Problem-identification searches are where many IT buyers first meet a provider: an error they cannot fix, a migration they are planning, a comparison they need to make. Articles and documentation that answer those well bring in the engineers and managers who later run the evaluation.

Google's self-assessment questions describe what helpful content of this kind offers: original information or analysis, a substantial and complete description of the topic, something beyond the obvious and first-hand expertise, for example from having actually used a product or service. In practice:

• Write from the work. Name the versions, the commands you ran, what failed and why. That first-hand detail is what a summary of other people's articles cannot copy.
• Cite primary sources (vendor documentation, RFCs, standards) inline, where each fact is used.
• Link each article to the service it supports, and link the service page back to its best articles. Organizing articles into a topic cluster around each service is covered in the semantic SEO guide.
• Do not write to a word count. Google says plainly that it has no preferred word count.
• Update when the facts change, and only then. Google lists changing a page's date to make it look fresh, without substantially changing the content, as a warning sign.

For software and product companies, the same applies to documentation. Public docs, API references and migration guides answer Know and Do queries better than marketing pages can, as long as they are public, crawlable pages rather than content behind a login.

Show who wrote it and why it can be trusted

Google's helpful content guidance asks creators to consider Who, How and Why. For the Who, it asks whether it is self-evident who wrote a page, whether pages carry a byline where one is expected, and whether bylines lead to more information about the author, and it strongly encourages accurate authorship information. Google calls the qualities it looks for E-E-A-T (experience, expertise, authoritativeness and trustworthiness), says trust matters most, and says E-E-A-T is not itself a specific ranking factor but something its systems identify through a mix of factors.

For an IT company, that translates into four things:

• Named authors with real credentials. A byline on every article and case study, linking to an author page that gives the person's role, what they work on, the certifications they actually hold, and their talks, open-source work and profiles elsewhere.
• An about page that says who stands behind the site. The rater guidelines tell raters to start with the "About us" page or the creator's profile page, then look for independent reviews, references and news about the business.
• Contact details that work: an address you operate from, a phone number and people a buyer can reach.
• Sources. Google's link guidance notes that external links citing your sources can help establish trustworthiness.

For what E-E-A-T is and is not, see how Google determines whether a website is relevant and authoritative.

Add the structured data that fits each page

Structured data states, in a form machines read reliably, what a page already says. Five schema.org types cover an IT services site:

Google asks for the most specific subtype of Organization and LocalBusiness that fits. For an IT firm's office that is usually plain LocalBusiness: none of the direct subtypes on schema.org's LocalBusiness page describes an IT services company, and schema.org deprecated ProfessionalService because it was confused with Service.

A service page could carry this, in JSON-LD, with the company described in full once on the home page and named here by the same @id:

{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "Service",
      "name": "Managed IT services",
      "serviceType": "Managed IT services",
      "description": "Monitoring, patching, help desk and backup for small and mid-sized offices.",
      "url": "https://www.example.com/services/managed-it/",
      "provider": {
        "@type": "Organization",
        "@id": "https://www.example.com/#organization",
        "name": "Example IT"
      },
      "areaServed": [
        { "@type": "Country", "name": "United States" },
        { "@type": "Country", "name": "Canada" }
      ]
    },
    {
      "@type": "BreadcrumbList",
      "itemListElement": [
        {
          "@type": "ListItem",
          "position": 1,
          "name": "Services",
          "item": "https://www.example.com/services/"
        },
        { "@type": "ListItem", "position": 2, "name": "Managed IT services" }
      ]
    }
  ]
}

Three rules keep structured data useful:

1. Mark up only what is visible. Google's general structured data guidelines forbid marking up content readers cannot see, and misleading content such as fake reviews.
2. Expect no guarantees. Google does not promise that valid markup produces a rich result. Its list of supported features (last updated in June 2026) includes Organization, Local business, Article and Breadcrumb, but neither Service nor FAQ, so FAQ markup on a service page earns no rich result.
3. Test it. Run the Rich Results Test for Google's features and the Schema Markup Validator for everything else, then check a deployed page with URL Inspection in Search Console.

AI search needs nothing extra. Google's page on AI features says there are no additional requirements to appear in AI Overviews or AI Mode, no special schema.org markup and no AI text files, and that structured data should match the visible text. AI Overviews in search covers how those answers choose their links.

Get the technical basics right on a large service site

Services multiplied by industries and locations, plus case studies and a few years of articles, add up quickly. Google describes a site of about 500 pages or fewer as small, and says a small, well-linked site may not need a sitemap. Past that size, or while the site is new and has few links pointing to it, submit an XML sitemap and watch index coverage in Search Console.

Three more things decide whether a large service site is crawled and understood.

Internal links. Google can generally crawl only <a> elements with an href, it asks that every page you care about have a link from at least one other page on your site, and it wants anchor text that describes the target. On a service site, the hub links to every service, each service page links to its industry pages, case studies and articles, and each of those links back to its service. Menus built from buttons with click handlers, and case studies reachable only through a filter widget, are the usual gaps.

Canonical URLs. The same page often answers at several addresses: with campaign parameters such as ?utm_source= or ?gclid=, with and without a trailing slash, or as a print view. Google ranks the methods for choosing one by strength: redirects and rel="canonical" links are strong signals, and listing a URL in a sitemap is a weak one. Do not use robots.txt for canonicalization, since Google may still index a disallowed URL without its content. How the Google search engine works explains how Google indexes a page and chooses its canonical.

Speed. Google recommends good Core Web Vitals, which measure real-world experience: Largest Contentful Paint within 2.5 seconds of the page starting to load, Interaction to Next Paint under 200 milliseconds and Cumulative Layout Shift under 0.1. On a service site, audit the third-party scripts first (chat widgets, booking embeds, tag managers carrying many tags), because each one runs on every page.

Measure leads, not traffic

Traffic is the easiest number to grow and the least useful one for an IT company. What matters is whether the right people get in touch and whether they become clients. Three tools, joined on the landing page, answer that:

In GA4, any event can be marked as a key event, and reports such as Landing pages then show a key events column. For lead generation, Google recommends a set of events meant for business-to-business sales and for conversions that happen offline: generate_lead when a form is submitted, qualify_lead, working_lead and disqualify_lead as sales works the lead, and close_convert_lead or close_unconvert_lead at the end. Sending them fills the Lead acquisition report, which counts new, qualified and converted leads by channel, including organic search.

Once your server has accepted the contact form, the page can record the lead. The event reference recommends a value when the event is a key event, and requires a currency whenever a value is sent:

// After your server has accepted the contact form
gtag("event", "generate_lead", {
  lead_source: "contact_form",
  value: averageLeadValue, // what a lead is worth to you, on average
  currency: "USD",
});

The later stages happen in your CRM, not in the browser. GA4's Measurement Protocol accepts events from a server over HTTP, which is how offline outcomes are tied to online behaviour. For server-to-server integrations, Google now recommends building on its Data Manager API.

One limit shapes the whole setup: Search Console data cannot be tied to individual visitors. When you connect Search Console to GA4, its metrics combine only with the landing page, device and country dimensions, so the landing page is the key that connects a query to a lead. Store it with every lead. Search Console keeps 16 months of data, so export it if you want a longer history.

Once leads are counted, the next gains usually come from pages that already rank but are not clicked. Untapped SEO opportunities shows how to find them in Search Console.

What to avoid

Most of what goes wrong on IT company sites comes from producing pages faster than the company produces work:

• City and industry pages that differ only by the name swapped in. Google's spam policies call city pages of this kind doorway abuse, and the pattern is the same with an industry.
• Spun case studies: one project rewritten as five, with the sector or the city swapped. Google's scaled content abuse policy covers many pages generated mainly to manipulate rankings with little value, however they were made, and its examples include automated rewrites such as synonymizing.
• Invented testimonials and results. In the United States, the FTC's final rule, announced on August 14, 2024, bans testimonials that claim to come from someone who does not exist, including AI-generated ones, or from someone with no real experience of the business.
• Keyword-stuffed titles and footers that list every service and every city.
• Bought links. Buying or selling links for ranking purposes is link spam under the same policies.
• Promises of rankings. Google's own advice on hiring an SEO says no one can guarantee a first-place ranking.

Our SEO service covers technical SEO audits, structured data, Core Web Vitals and content planned from search demand, measured in Search Console, plus local search profiles for businesses that serve an area. It does not include link buying or guaranteed rankings.