Free SEO Tools for Small Business: A Practical Guide

The best free SEO toolkit for a small business starts with Google Search Console, PageSpeed Insights and a short list of pages that matter to revenue. Add a crawler when you need to find technical problems, analytics when you need to measure enquiries or sales, and local tools when customers visit you or you visit them. You do not need to open every dashboard every week.

This guide compares nine tools and tool categories, explains the access requirements people often miss, and shows how to turn their reports into a first-month work plan. The goal is a functioning measurement and improvement process with no required software subscription.

Choose your free toolkit by the job

For a new five-page service website, start with Search Console and a PageSpeed check of the homepage and main service page. For a larger existing site, add Screaming Frog. For a shop or service-area business, move Business Profile near the top of the list. For an online store, set up meaningful purchase measurement before using visits as your success metric.

Cloudflare Web Analytics is an alternative inside the analytics section if a simpler traffic and performance view is enough. Keep a single working issue list across these tools: three reports describing the same broken page represent one problem.

01 Google Search Console: start with your own search data

Use it for: Google visibility, page-level search trends and indexing investigation. Access: a Google account plus verified ownership or permission to the property. Installing Google Analytics is not a prerequisite. Start at Google Search Console.

The Performance report shows clicks, impressions, CTR and average position, with dimensions such as pages, queries, countries and devices. Average position depends on the reporting context; it is not a fixed daily rank for every searcher. Google's Performance report documentation explains these metrics and filters.

Your first useful check

1. Choose a recent, complete reporting period and compare it with an equal-length previous period. Keep the search type, country and device filters consistent.
2. Open Pages and pick a service, category or product page that matters to the business.
3. Filter to that page, then open Queries. Check whether the visible searches match the job the page should answer.
4. Inspect the same URL in URL Inspection if its indexing status is unclear. Record the result and the date.
5. Write one action: repair an indexing barrier, clarify the offer, improve an answer, or leave the page unchanged while collecting more evidence.

Example output: “The boiler repair page appears for repair-intent queries, but the page does not explain the service area or how to book. Add those details and record the change date.” This is a defensible content action without promising that the edit will improve rankings.

What the report cannot prove

Google omits anonymized queries and does not expose every query row, so a missing phrase is not proof of zero demand. Read the data limitations before adding query rows together. Clicks also are not the same measurement as analytics sessions or enquiries.

URL Inspection distinguishes Google's stored indexing information from a live fetch. A successful live test does not establish that the URL is indexed or guarantee future inclusion; see Google's URL Inspection guidance. For the larger workflow, follow what to check in Search Console before paying for SEO.

02 Website analytics: connect visits with useful actions

Use it for: understanding landing pages and actions after arrival. Search Console supplies search reporting without a visitor analytics tag. Add on-site measurement when you have a question such as “Which organic landing pages generate successful quote requests?”

Google Analytics 4 for events and outcomes

The standard Google Analytics product is free. You still need a property, a web data stream, a correctly installed tag and appropriate event setup. A dashboard that collects pageviews alone does not automatically measure a booking, a completed purchase or a qualified lead. Google's setup guide covers the starting configuration.

Choose one business action first. For a service company, use a confirmed enquiry rather than any click on the contact button. For a shop, use a completed purchase with correctly implemented transaction data. Test that action before relying on the report, and distinguish a measurement test from genuine customer activity.

Google calls important measured actions key events. Marking an event does not repair a faulty trigger: a “success” event firing on a failed form will still produce misleading numbers.

1. Find the Landing page report. Navigation can differ between report collections.
2. Add Session source / medium, or use an Organic Search comparison for the question you are answering.
3. Read sessions alongside the relevant key event, using the same dates.
4. If visits exist but outcomes do not, check the event implementation and the page's booking or buying journey before concluding that SEO traffic is poor.

The Landing page report documentation describes the session-scoped dimensions. Do not expect its session totals to match Search Console clicks exactly: they measure different things, and collection conditions differ.

A simpler alternative: Cloudflare Web Analytics

Cloudflare Web Analytics provides free traffic and real-user performance reporting with an account and a JavaScript beacon; moving DNS or using Cloudflare's proxy is not required. Cloudflare describes the service as privacy-first and states that it does not collect or use visitors' personal data. Consider it when pageviews, visitors and performance answer your questions. Choose an event-oriented setup when the decision depends on specific completed actions.

Practical limit: an analytics setup only helps if its collection and event definitions are understood. Check the applicable consent requirements and provider controls before implementation; Google's consent documentation covers its EEA-related use cases. Keep customer names, email addresses and form contents out of analytics event labels and page URLs.

03 PageSpeed Insights: turn a slow page into a fix

Use it for: investigating a specific page's loading, responsiveness and layout stability. Access: paste a publicly accessible URL into PageSpeed Insights. Start with an important service or product page, not just the homepage.

PSI combines a Lighthouse lab test with Chrome User Experience Report field data where available. The field section covers a trailing 28-day period; it may show URL data, fall back to origin data, or lack sufficient samples. Origin data describes the wider site, not necessarily the exact page you entered. These distinctions come from Google's PSI documentation.

Use the result in this order

1. Select mobile or desktop deliberately and record it.
2. Check whether the real-user section describes this URL or the origin.
3. Review LCP, INP and CLS where present, then use lab diagnostics to investigate likely causes.
4. Choose one concrete change, such as reducing a large above-the-fold image or investigating a script that blocks interaction.
5. Retest after the fix under comparable conditions. Treat lab runs as diagnostics; real-user reporting needs time to reflect new experiences.

Example handoff: “The main service image is much larger than its displayed size. Export an appropriately sized image, preserve its aspect ratio and dimensions, then compare the page before and after.” This is more actionable than “get performance to 100.”

Practical limit: a green lab score does not establish a good experience for all visitors or a ranking improvement. Missing field data does not establish a fast page. Save the report context and send the developer the specific finding. The implementation guide to fixing Core Web Vitals goes deeper.

04 Keyword Planner: build a small, relevant keyword map

Use it for: keyword ideas and Google advertising-oriented demand estimates. Access: Google Ads account setup. Google's current Keyword Planner instructions explicitly require billing information and completed setup to access basic features such as keyword ideas. Calling it a free tool should not hide that requirement.

If you do not want that setup, begin with your Search Console queries, customer questions, Bing Keyword Research and Trends. Do not launch an advertising campaign just to follow an SEO checklist.

Research a service you actually sell

1. Open Discover new keywords and enter a small group of related services or your website.
2. Set the relevant geography, language and network context before interpreting the results.
3. Remove jobs, training, unrelated products and places you do not serve.
4. Group close variants by the customer task: repair, installation and maintenance may need different answers; tiny wording variations often do not.
5. Assign each useful group to an existing page first. Propose a new page only when the intent needs a distinct answer.

Example output: a plumber maps “boiler repair” and closely related repair phrases to one repair page, while an installation quote belongs to a separate installation page. A research note saying “customer wants repair availability and booking details” is more useful than a list of 200 loosely related terms.

Practical limit: average monthly searches are estimates influenced by settings and close variants. Preserve any ranges shown instead of turning them into exact counts. The Competition column describes advertisers, and bid estimates concern ads; neither measures organic ranking difficulty. Google's metric definitions explain this. Do not add every close variant's volume into a traffic forecast.

05 Bing Webmaster Tools: get a second search perspective

Use it for: Bing search performance, URL investigation, keyword research and a browser-based technical scan. Access: an account and verified-site access for site-specific reporting. The official overview lists the available tools.

Start with your important landing pages. Compare their Bing visibility with the reader task each page should serve. Different engines can produce different observations; a page appearing in one engine does not prove the other has indexed it.

Use research and scanning separately

Bing Keyword Research provides related ideas, question keywords and trend context, with country, language and device filters. Keep its figures labelled as Bing data. Use a useful question to improve an existing page rather than automatically opening a new article brief.

Site Scan crawls a selected scope and reports issues. The quota is allocated to the site automatically, so check the allowance shown in your account rather than assuming a universal monthly URL limit. Set a bounded scan and review affected URLs before applying a recommendation.

Example output: “Site Scan flagged several pages with missing titles; they share one template. Check the template once and fix the cause.” The report is a starting point for verification, not permission to rewrite every flagged page.

If your account offers AI reporting or Topics Portfolio, treat it as a separate measurement task. Availability and citation units need their own interpretation; the Bing Topics Portfolio guide shows a documented FloxoLab example. A citation count is not a visit or a lead.

06 Google Trends: decide when to publish or refresh

Use it for: seasonal patterns and relative changes in search interest. Access: the public Google Trends interface. It is particularly useful for services with a planning season, weather sensitivity or a recurring annual buying cycle.

Compare related terms with the same geography, time range, category and search context. Use a multi-year view to look for repetition, then a shorter view for current timing. Check whether you selected a search term or a topic before comparing entries.

Example output: a seasonal service business records “refresh the service page and booking information before the recurring interest peak.” That is a planning hypothesis to check against actual enquiries and capacity, not a claim about next month's bookings.

Practical limit: Trends normalizes interest to a 0–100 scale for the selection. A value of 100 is a relative peak, not 100 searches. Low-volume terms may appear as zero or lack a graph, and unusual spikes need context. Google's data FAQ explains these limits. Broaden the period or geography when appropriate, but record that change: national seasonality is not proof of demand in one neighbourhood.

07 Screaming Frog: find technical problems on a small site

Use it for: a desktop crawl of discoverable URLs, response codes, titles, metadata and directives. Access: download SEO Spider for Windows, macOS or Linux. The free edition permits 500 URLs per crawl. It is an ongoing restricted edition, not a short trial.

The free version can expose broken links, redirects, missing or repeated titles and meta robots directives. Saving and reopening crawl files, JavaScript rendering, scheduling, integrations and advanced configuration require a licence. Check the official general guide and feature comparison. A free crawl is not equivalent to the complete paid audit workflow.

A first crawl worth doing

1. Enter the canonical homepage URL of a site you own or are authorized to audit.
2. Let the crawl finish, or record explicitly if it stopped at its limit.
3. Review internal client and server errors, then inspect the source links pointing to affected URLs.
4. Review important indexable pages for missing or confusing titles and unintended directives.
5. Export the relevant rows and write an owner, action and verification step for each actual problem.

Example output: “Three service pages link to a retired booking URL. Update those links to the current booking page and confirm the destination works.” Fixing the source links is more precise than adding a redirect without checking why the links remain.

Practical limit: 500 URLs does not mean 500 sellable pages: assets and URL variations can consume crawl capacity. An unlinked page may be missed, a stopped crawl is incomplete, and JavaScript-dependent content needs a method the free edition cannot provide. Do not label every duplicate title or noindex URL an error without checking its purpose.

Use our Screaming Frog review to understand the free-versus-paid boundary, or the crawl and Search Console audit workflow to connect crawler findings with Google's reporting.

08 Google Business Profile: the local-business foundation

Use it for: managing an eligible business's presence in Google Search and Maps. Google provides Business Profile at no charge. Eligibility matters: online-only businesses are not eligible simply because they want local visibility. Check Google's eligibility rules for your business model.

For an eligible shop, clinic, restaurant or service-area business, start with accuracy: the real business name, suitable categories, current hours, contact information, services and a useful website destination. Resolve customer confusion before buying a local ranking subscription.

Make the first review about customer actions

Check the profile as a customer would, then use Performance for the available metrics and reporting period. Google's Performance documentation explains access and the interactions reported for verified profiles. Keep website clicks, call-button clicks and other interactions distinct. A call-button click is not confirmation of an answered call or a sale.

Example output: “The profile sends visitors to the homepage even though a useful location page exists. Check the destination and make sure the address, hours and booking information agree.” For a multi-location business, review each real location's details rather than copying every field blindly.

Practical limit: profile performance is not a map-grid rank tracker. One manual Maps search cannot describe visibility across an entire service area. For recurring location-based checks, use the decision framework in our local SEO tools guide. The broader small-business local SEO guide explains the work behind the tools.

09 Rich Results Test: validate a specific markup change

Use it for: checking structured data supported by Google's rich-result features. Paste a public URL or a code sample into the Rich Results Test. Use it when you change a relevant page template or markup implementation.

Inspect the detected item types and open each error. Check that the markup describes visible page content. After changing a template, test a representative live page as well as the code: the deployed page can differ from the snippet you intended to publish.

Example output: “The product template is missing a required property for the intended supported feature. Confirm the correct value exists in the product data, fix the template and retest an affected URL.” Do not manufacture review ratings or offer details to satisfy a validator.

Practical limit: Google-specific validation is narrower than validating all Schema.org vocabulary. Google's testing-tools overview distinguishes the Rich Results Test from Schema Markup Validator. A valid result does not guarantee a rich result, indexing or better rankings, as the structured-data policies explain. For screenshots and a complete walkthrough, use the Rich Results Test guide.

Turn reports into one prioritized work list

The expensive part of free tools is often the time spent collecting overlapping warnings. Keep the evidence with the decision. A useful task describes the affected URL, what was observed, why it matters, who can fix it and how completion will be checked.

Start with barriers affecting important customer pages: failed URLs, accidental indexing restrictions, broken booking paths or unusable mobile content. Next improve relevant pages already appearing in search. Schedule lower-impact metadata tidying after those tasks. A crawler's severity label helps investigation; your affected pages and business context determine priority.

Download the first-month SEO workbook (.txt) for a blank page inventory, issue register, research map and review checklist. Copy the fields into the document or task system you already use; no additional paid tool is needed.

A practical first-month SEO plan

This schedule is a suggested order of work, not a promise of results within 30 days. A new site may have little search data by the end of the month. Completing measurement and fixing a verified problem are useful milestones even when traffic remains too small for conclusions.

Week 1: access, important pages and a baseline

Get access to Search Console and Bing Webmaster Tools. List the homepage and the service, product or location pages customers need most. Check their indexing information. If an eligible Business Profile matters, review its accuracy. Decide whether you need on-site outcome measurement and define the action before installing more tags.

Deliverable: a priority-page list, access notes and a dated baseline. Record “not enough data” where appropriate rather than filling the gap with estimates.

Week 2: technical obstacles and the customer journey

Run a bounded crawler check and PageSpeed tests on the key page types. Inspect the actual pages before assigning work. Fix a short list of confirmed barriers, such as broken internal links, unintended noindex directives or a difficult mobile booking path. Keep a copy of the original observation and the deployment date of each fix.

Deliverable: a few closed tasks with verification evidence, plus a clearly separated backlog. A shorter, verified list beats hundreds of unreviewed warnings.

Week 3: improve one page and map the next opportunity

Use visible Search Console queries and customer questions to choose an existing page. Consult Keyword Planner or Bing for related wording, and Trends if timing matters. Improve the page's answer, service details, examples or next step. Avoid creating near-identical pages for every phrase. Validate structured data if the change affects a supported implementation.

Deliverable: one substantive page improvement and a small keyword-to-page map. Record why an additional proposed page is needed before writing it.

Week 4: review changes and set the next cycle

Check that the fixes remain live, review the issue register and save another report with the same filters. Distinguish successful implementation from later changes in impressions, clicks and genuine enquiries. If the comparison period crosses a seasonal shift or a tracking change, record that limitation.

Deliverable: a short review: what changed, what was verified, what remains uncertain and which action comes next. Use the guide to reading a monthly SEO report when you need to assess someone else's reporting.

When paying for a tool becomes reasonable

Buy capacity for a repeated task you already understand. A subscription is easier to assess when you can say “we need weekly tracking for these locations” or “our crawl stops before reaching the product catalogue.” Buying more charts without a decision process adds work.

• The crawl is incomplete: investigate a larger crawl allowance or the rendering capability your site needs.
• You need scheduled rank checks: define the keywords, search engine, devices, locations and update frequency first.
• You need competitor or backlink research: assess the relevant provider's coverage and sample outputs. Third-party estimates do not replace your own search data.
• You manage several sites: calculate the time spent repeating exports and assembling reports.

Before upgrading, write down the current allowance, the exact paid feature required and a representative task for evaluating it. A trial helps test that task; it is not a permanent free operating plan. Our SE Ranking review, documented Mangools test and Screaming Frog review provide narrower comparisons when one of those jobs becomes necessary.

A simple break-even estimate is monthly hours saved × your value per hour, compared with the subscription and setup effort. Use your own numbers and include the time needed to interpret the new reports. Better coverage can justify a tool, but it does not guarantee traffic or sales.

Common questions

What is the best free SEO tool for a small business?

Start with Google Search Console for your own Google search and indexing data. Then add the tool for the immediate problem: PageSpeed Insights for performance, a crawler for technical investigation, or Business Profile for an eligible local business. No single free tool handles every task.

Can I do SEO without a paid subscription?

Yes. You can establish search measurement, inspect important URLs, research relevant topics and fix many small-site problems with this toolkit. The constraints are your implementation time, data availability, crawl capacity and the depth of competitor research you need.

Does free Keyword Planner require a payment card?

Google's current documentation requires billing information and completed Google Ads account setup for basic Keyword Planner access. The payment methods and setup screens can depend on the account. Check that requirement before choosing it; Bing research, Trends and your own queries are useful starting alternatives.

Is Screaming Frog free for a 500-page website?

The limit is 500 URLs in a crawl. Images, scripts, stylesheets and URL variations can consume capacity, so a site with 500 HTML pages can exceed it. The free edition also restricts features independently of size, including JavaScript rendering and saving crawl files.

Do I need Google Analytics to appear in Google?

No. Use Search Console to investigate Google's discovery and indexing. Add website analytics to answer questions about what happens after arrival. A correctly defined outcome is more useful than installing analytics simply because an SEO list tells you to.

Why does PageSpeed show no real-user data?

The URL or origin may not have enough qualifying CrUX samples. Use the lab test for investigation, record that field evidence is unavailable and avoid presenting the lab score as a measurement of every visitor's experience.

Will these tools get my business into AI answers?

They can help investigate technical access, page quality and search observations. None guarantees AI citations or recommendations. Keep citations, search clicks and customer enquiries as separate measurements; use a documented observation method before making an AI visibility claim.

Sources checked

Official product documentation checked November 5, 2026. Recommendations and worked task examples are FloxoLab's editorial guidance. This refresh did not run a new account-level comparison or measure a traffic uplift.

• Google Search Console: Performance report, data limitations and URL Inspection.
• Google Analytics: free product overview, setup, Landing page report, key events and consent settings.
• Cloudflare: Web Analytics scope and setup.
• Google: PageSpeed Insights and Trends data FAQ.
• Google Ads: Keyword Planner access and workflow and estimates and metric definitions.
• Bing: Webmaster Tools overview, Keyword Research and Site Scan.
• Screaming Frog: free-edition restrictions and feature comparison.
• Google Business Profile: getting started, eligibility and Performance metrics.
• Google Search Central: testing tools and structured-data policies.