The cheapest content plan a local business will ever get is sitting in the competitor pages already ranking above it, and the job is to read them, group them by intent, and build only the pages that are worth building.
Keyword research sounds expensive until you reduce it to the practical question a garage door company, a barber, or a mobile notary actually has: what does the site above me answer that mine does not? A competitor may have separate pages for springs, rollers, openers, commercial doors, emergency repairs, and each city it serves, while the smaller owner has one generic home page that says "we do it all" and proves none of it. The gap is not just a missing keyword. It is missing proof that the business does that work, in that place, for that kind of customer.
This article is about that gap: what it is, how the industry sells the job of finding it, and how WALO turns it into a short approval-gated flow an owner can run from a phone. It keeps NU's rule throughout. Kooky till proven, receipts not verdicts, the record wins. Where a feature on WALO's site is a promise rather than a shipped tool, that is said plainly.
The problem owners actually have
Picture a Tuesday morning in Fremont. The owner of a two-truck garage door outfit has a slow week and opens Google on his phone to search "garage door spring repair Fremont." His site is not on the first page. Above him sit a national franchise, two directory listings, a competitor from the next town with a page titled exactly "Garage Door Spring Repair in Fremont, CA," and, further down, a site that uses his business name with a different phone number. He has one web page. It lists nine services in a bulleted paragraph and has a contact form.
He does not have a keyword problem in the way an agency would describe it. He has a page problem. The competitor has a page for the thing people search. He has a paragraph. Google's own guidance on how it ranks local results says results are "mainly based on relevance, distance, and popularity," and that relevance means "how well a Business Profile matches what someone is searching for." A dedicated page that says spring repair, shows spring repair, and answers spring repair questions is relevant. A line in a bulleted list is not.
The owner's second problem is that he does not know which of the dozens of possible pages would actually move the phone. Springs? Openers? Commercial? Emergency? "Garage door repair near me" versus "garage door repair Fremont"? He could spend a weekend writing all of them, and a good chunk of that weekend would be wasted on searches where a directory or a national brand will always win regardless of what he writes.
The third problem is time. The owner's day is driving, quoting, and installing. He has fifteen minutes in a truck between jobs. Any content plan that needs a laptop, a spreadsheet, and a subscription is a plan that will not happen.
How the industry handles it today
The standard answer to this problem is a keyword research suite. Ahrefs lists its Lite plan at $129 per month and Standard at $249 per month on its public pricing page. Moz Pro's Standard plan is $99 per month on its pricing page. BrightLocal, which is built for local businesses specifically, starts at $41 per month for a single location on its pricing page. Semrush publishes its plans on its own prices page. These tools are good. They are also designed for people whose job is SEO, with dashboards that assume you know what "keyword difficulty" and "SERP features" mean.
The feature these suites call "content gap" or "keyword gap" does roughly the same thing across vendors. You enter your domain and a competitor's, and the tool lists keywords the competitor ranks for that you do not. The output is a spreadsheet, often hundreds of rows, with estimated volumes and difficulty scores. For an agency managing twenty clients, that spreadsheet is the raw material of a monthly plan. For an owner in a truck, it is a wall.
There is also a data caveat the vendors are honest about if you read the fine print: volumes are estimates, rankings are sampled, and local results vary by the searcher's location. The only first-party source for what searches actually sent impressions and clicks to your site is Google Search Console's Performance report, which reports clicks, impressions, click-through rate, and average position per query, and only for queries where your site already appeared. That is a key limit worth saying twice. Search Console cannot tell you about the searches you have never appeared for. The competitor's pages can.
There is a cheaper, manual version of the same work that agencies have always done and that an owner can do with a notebook: open the competitor's site, write down every page title and every heading, and sort them into four piles. Service pages, which answer "do you fix this." Location pages, which answer "do you come here." Questions, which answer "how much, how long, what kind." Proof, which answers "have you done this before," usually photos, reviews, and job write-ups. A competitor with twelve service pages and a page per city has told you their whole plan in public. The limit of the manual version is not skill, it is time, and it is the step most owners never get to because the audit tools present the data as rows of keywords rather than piles of pages.
The payoff for closing a gap is measurable. Backlinko's analysis of roughly four million search results found that the first organic result has an average click-through rate of 27.6 percent and is about ten times more likely to be clicked than a result in position ten. The same study found that positions eight through ten get almost the same clicks, so moving from tenth to eighth is not a win, while moving from page two to the top three is. That is the honest economics of the gap: you are not chasing "more keywords," you are chasing a handful of searches where a real page can realistically reach the top three.
One more piece of the record matters before any owner writes a single page. Google's spam policies now name "scaled content abuse," defined as generating many pages "for the primary purpose of manipulating search rankings and not helping users." Google's February 2023 statement on AI content says the same thing from the other direction: automation used to generate content whose primary purpose is manipulating rankings violates the spam policies, while high-quality content is rewarded "however it is produced." The gap finder that spits out fifty thin city pages is not a plan. It is a liability.
How WALO does it
WALO is an AI assistant that lives in WhatsApp and has a web tool called SEO Pro at walohq.com/seo/. The product promise on that page is short: "Grade your site's SEO in 5 seconds. 20+ checks with a live score, category graphs, plain-English '?' explainers, and a one-click fix that writes your Google schema for you. Any site: yours or a competitor's." It is free and there is nothing to install.
The gap-finding flow on that page is a chain of four panels, and the order matters because each one feeds the next.
Rank Radar: see who is actually above you
The first panel is called Rank Radar. The copy on the page says it plainly: "Type the keyword you want to win. See the exact Google order, where you sit vs competitors, directories (Yelp, Google), and yeah... imposters." You type the keyword and an optional city, tap Show ranking, and WALO pulls the live Google order for that phrase.
This panel solves the first half of the owner's problem: it separates the results into the three kinds that matter. Competitors are other businesses like you. Directories are Yelp, Google's own listings, and aggregators that you cannot outrank with a page but can be listed on. Imposters are results using your brand name that are not you. For the Fremont owner above, that last category is not theoretical. The current NU article on rank alerts in this same batch covers the copycat problem in detail.
Competitor Gap: auto-pick the one that matters
The second panel is labeled "Competitor Gap - auto-pick from Rank Radar," and the instruction is to run Rank Radar first. The page describes what happens next: "WALO pulls the live search results, chooses the closest non-directory result above you, then compares both sites. You can still override it manually."
Two design choices here are the whole point of the feature. First, "closest non-directory result above you" means the tool does not waste the owner's attention comparing him to Yelp. It picks the business that is actually beating him. Second, the manual override exists because the auto-pick can be wrong, for example when the closest result above you is a national brand's city page you are never going to beat on authority. In that case the owner types the regional competitor he actually loses jobs to.
The comparison itself is between the two sites' pages, titles, headings, and signals, which is the "find" step from the original short version of this article: pull competitor titles, headings, and ranking phrases. The output is not a four-hundred-row spreadsheet. It is the difference between two sites.
Top keywords and edit plan: group by intent before you write
The third panel is "Top keywords and edit plan." Its description: "Auto-generate buyer keywords, title/meta copy, and FAQ ideas. Edit the plan before changing a real site." The inputs are the business name, an optional city, and an optional competitor URL. The output is a list the owner can edit.
This is the "group" step: separate service pages, local pages, FAQs, and article ideas. A good gap finder groups opportunities by intent, because some searches need a service page ("garage door spring repair Fremont"), some need a comparison article ("torsion vs extension springs"), some need a local proof page ("garage door repair in Niles district"), and some are not worth chasing because the competitor ranks with a directory or a national brand. The owner's job at this panel is to delete the lanes that are not his and keep the three or four that are. The panel is labeled an edit plan for a reason. Nothing changes on a real site from here.
The screenshot above also shows something useful that happens before the keyword panel: the audit's "Top fixes, most important first" list. In this capture of WALO's own site, the top two fixes are speed, with a phone score of 44 out of 100 and the main content appearing after 15.2 seconds, and the tool says what to do about it in plain words. A gap plan sits on top of a site that loads. If the audit says the site is slow, the plan has a step zero.
Landing Page Builder and Free Guide Page: produce the page, then approve it
The fourth step is producing something. SEO Pro has two different producers for two different kinds of gap.
For a service gap, there is the Landing Page Builder: "Enter the service or offer you want found for. WALO generates a paste-ready landing page and schema for the domain above. No fake claims; review before publishing." The tip under it says to run the audit first for better site context. This is the page that closes a "springs" or "commercial doors" gap. It is paste-ready, which means the owner or whoever runs the site still publishes it. WALO does not push it live from this panel.
For a topical gap, there is the Free Guide Page. The copy describes a different kind of asset: "WALO studies who ranks for your topic, their titles and meta tags, and what people ask on Reddit, then writes a 4,000+ word guide featuring you on NothingUnseen or Apiary, whichever genuinely fits your business. You approve it before anything goes public, and you can edit every word." The three steps are listed on the panel: scan the site (name, phone, what you sell, logo, pick the publication), confirm the details and topic, then generate and approve, which takes about two minutes and ends in a private preview link by email.
This matters for the gap problem because a competitor sometimes ranks with an article, not a service page, and an owner cannot beat an article with a service page. The guide gets published on Nothing Unseen or Apiary, both AMH publications, with the business featured in it. The honest framing is that this is a guide about the topic that features the business, not an advertisement wearing a guide costume, and the "no fake claims" rule applies.
A receipt on this feature, because NU prints receipts. The text-to-article pipeline behind the guide page was first run in September 2026 for a Nothing Unseen article about looking up a fictitious business name. The first draft invented three of its six source links and made one wrong claim about which agency holds DBA filings. The draft was corrected by hand, and the research and link-verification step was added before any second article shipped. Today every link in a draft is checked before the preview is sent, and the preview reports how many pages were read and how many links were verified. That is why the Free Guide Page says you approve before anything goes public. The approval step exists because the first run proved it had to.
The WhatsApp side: ask it from the truck
Everything above is the web tool. The same owner can run the short version from WhatsApp. WALO's free scan page shows the shape of the conversation: the owner texts "check my website SEO and tell me what to fix first," WALO answers with something like "Your title is clear, but you need a service page, FAQ schema, and a stronger internal link from the homepage. I can draft the page and show it before anything goes live," the owner texts "make the free page draft," and WALO replies "Draft ready... Nothing publishes until you approve."
That is the original "action card" idea from the short version of this piece: explain the move, show the source, let the owner approve, log the result. The owner does not need a marketing department to do basic operating work. He needs a tool that reads the competitor, says which three pages to build, drafts them, and waits.
Under the hood, in plain English
An owner does not need the architecture, but a smart owner wants to know where the data comes from and what the tool can and cannot see. Here is the chain without the secrets.
The audit fetches the page the way a search engine would and runs its checks: title, description, headings, links, images, load time, schema, local business signals. In the capture above it reported word count, image count, link count, H2 count, page weight, and load time for the scanned site. Some sites block automated fetches, and WALO's audit box says so up front: "protected sites like Shopify take a little longer." When a site blocks the audit entirely, the right behavior is to say the site could not be read, not to invent a score, and WALO's own site makes that commitment in several places, including the NAP check, which says "WALO does not list a public phone or address on the page, so this check will not invent one."
Rank Radar pulls a live set of Google results for the phrase. That is a lookup against real search results, which is why the FAQ on walohq.com says some rank data "may carry a small cost" as an add-on even though the core tools are free. Live results are a consumable resource.
Competitor Gap then runs the same audit on the competitor's site and lays the two next to each other. The keyword panel takes the two audits plus the business name and city and generates a buyer-keyword list with title and meta suggestions. The Landing Page Builder takes the chosen service and writes a page plus schema. All of this is drafting. None of it writes to the owner's live site.
The approval gate is the part that makes it a product rather than a chatbot. WALO's connect page states the rule for every connected account: "Connecting is not permission. WALO drafts first, shows exactly what changes, and waits. Nothing writes to code, DNS, payments or a public feed without your word." The four-step loop on that page is Ask, Stage, Approve, Verify. Site and code changes, when an owner does connect a site, run through Walrus Tusk, the AMH operations engine inside WALO, which does a dry run, a live health check, and an automatic rollback if the check fails. For the gap workflow specifically, most owners will never get to that step: they will take the paste-ready page and publish it themselves.
Logs exist at the level of what was drafted, what was approved, and what shipped. A plan without a log is a plan nobody can audit a month later, which is why the short version of this article called the last step "log the result." In practice the log is the owner's WhatsApp thread plus whatever the connected site records, and that is enough: a month from now the owner can scroll up and see which page was drafted on which day, what the audit said before it, and whether he approved it. That is also the honest answer to "did the SEO work?" You compare the Rank Radar order from before the page to the order after, with the page's publish date in between, and you stop guessing.
One more plain-English point about the AI itself. WALO's FAQ says the brain is "a fast, free large-language-model brain (Llama-class), with a fallback so it never goes dark," and that the owner does not manage any of it. For the gap workflow, the model's job is narrow: read two audits, group phrases by intent, draft a page from facts the audit found. The rule WALO repeats across its site is that it will not invent a price or a fact and will say so instead of guessing. That rule is what makes a drafted service page safe to read on a phone between jobs: the owner is checking true claims for tone, not hunting for fabricated ones.
What it costs and what is free
Only what the site states. SEO Pro's page says "Free - nothing to install." The free scan page says "Free scan. No coding needed." The Free Guide Page form says "Free, no signup." WALO's FAQ answers "Is WALO free?" with: "The core runs on free infrastructure and the first week is free. The AI brain, SEO checks, and content tools cost you nothing to run, only advanced add-ons (like some rank data or premium voice) have a small cost." The FAQ also says there are no per-message fees and no contract. WALO's home page describes three access tiers in general terms: a free core, Power AI "when needed" for heavier model usage and larger workflows, and Teams or Enterprise "for serious ops" with shared approvals and audit trails. No prices are listed for the paid tiers on the pages read for this article, so none are quoted here.
Compared with the industry: the gap-finding part of the job, which the subscription tools charge between $41 and $249 per month for, is free on WALO at the level of one keyword, one competitor, one plan. What you give up is the spreadsheet scale. If you manage twenty clients, buy the suite. If you own two trucks, you do not need it.
Honest limits and what is still being built
Rank Radar reports a live Google order for a phrase, and that order varies by location, device, and personalization. The Fremont owner should enter the city. A result "above you" in a tool is a sample, not a law.
Competitor Gap depends on Rank Radar having run, and its auto-pick can choose a site you cannot realistically beat. The manual override exists for that reason, and the owner should use it.
The keyword panel generates ideas. It does not know which of your services is most profitable. The owner's edit of the plan is a real step, not a formality.
The Landing Page Builder output is paste-ready, which means it is not published for you. Publishing requires either the owner's own site access or a connected site through WALO's approval flow. If the owner's site is built on a platform WALO cannot write to, the page gets pasted by a human.
The Free Guide Page publishes on AMH's own publications, Nothing Unseen or Apiary, not on the owner's domain. That is a real backlink and a real page about the business, but it is a different kind of asset from a service page on your own site. The honest use is topical gaps, not service gaps.
The audit cannot read every site. Protected storefronts and some directories block automated fetches. When that happens the correct output is "could not read," and the owner should expect that rather than a fake grade.
Some rank data may become a paid add-on, per the FAQ. Daily automated rank tracking is covered in the companion article on rank alerts, and it runs on Google Search Console data, which only exists for queries your site already appears for. A gap, by definition, is a query you do not appear for yet, which is exactly why the gap tool reads the competitor rather than your own console.
Finally, the whole thing is a tool, and a tool can be used to produce the fifty thin city pages Google's spam policy names. WALO's builder says "No fake claims; review before publishing." That line is doing a lot of work. The owner who ignores it is building a liability.
A 30-day playbook
Week one: measure. Run SEO Pro on your own site and read the "Top fixes, most important first" list. If speed is the top fix, do that first, because a new page on a slow site inherits the slowness. Pick the three searches that actually send jobs: service plus city, emergency service plus city, and your single most profitable service. Run Rank Radar on each with your city entered and write down who sits above you in each one, sorted into competitor, directory, and imposter. If there is an imposter, start the companion rank-alert playbook the same day.
Week two: compare. For each of the three searches, run Competitor Gap. Accept the auto-pick if it chose a real local competitor; override it if it chose a national brand. Read the comparison and write down the pages the competitor has that you do not. While you are on the competitor's site, note three things the tool cannot judge for you: whether their page shows real photos of the work, whether it names the neighborhoods it serves, and whether it answers the questions a customer asks on the phone (price range, how fast, what brands). Those are the proof elements your version has to match or beat, and they are the part a human owner writes better than any model. Then run the keyword panel with your business name, city, and the competitor URL, and edit the plan down to no more than four pages. Delete anything a directory will always win. Delete anything you do not actually do.
Week three: produce. For each service gap, use the Landing Page Builder with the audit already run, read the draft for any claim you cannot back, remove the ones you cannot, and publish the page on your own site with a link from your home page. For one topical gap, if there is one, run the Free Guide Page scan, confirm the topic, and review the preview when it arrives. Approve it only when every sentence about your business is true.
Week four: submit and watch. Put the new pages in your sitemap, link them internally, and submit them through the Indexing actions panel, which is its own article in this batch. Then re-run Rank Radar on the three searches and compare the order to week one. Expect nothing dramatic in one week; the point is to establish the baseline the rank alerts will measure against. Set the daily or weekly alert on the three phrases and stop checking by hand.
Repeat the cycle monthly, not weekly. Pages take time to be crawled, indexed, and ranked, and the owner who rewrites a page every week never finds out whether the first version worked.
FAQ
Is a keyword gap the same thing as a missing keyword? No. A gap is a missing page, or missing proof. If a competitor ranks for "commercial garage doors Fremont" with a page that shows commercial installs, the fix is a page that shows yours, not a sentence that mentions the phrase.
Which competitor should I compare against? The closest real business above you for the search you want to win. WALO's Competitor Gap auto-picks "the closest non-directory result above you" for that reason. If it picks a national brand you cannot realistically beat, override it with the regional competitor you actually lose jobs to.
Why not compare against Yelp or Google's own listing? Because you cannot outrank a directory with a page. You can be listed on the directory, and you should be, but that is a listing task, not a content gap.
How many pages should I build from a gap analysis? As few as actually match work you do and searches that send jobs. Four is a reasonable monthly cap for a small business. Fifty thin pages is the pattern Google's spam policies call scaled content abuse.
Does WALO publish the page for me? The Landing Page Builder produces a paste-ready page and schema; you publish it. The Free Guide Page publishes a guide featuring your business on Nothing Unseen or Apiary after you approve the preview. Changes to a connected site go through WALO's approval and verification flow.
Can I use this for a competitor's site instead of my own? Yes. SEO Pro says "Any site: yours or a competitor's." Auditing the competitor first is a legitimate way to see what their page does well before you draft yours.
What does the audit do if a site blocks it? Some protected sites take longer and some block automated reads entirely. The correct behavior, and the one WALO's site commits to in several places, is to say it could not read the site rather than invent a result.
Is this free? The audit, Rank Radar, Competitor Gap, keyword plan, Landing Page Builder, and Free Guide Page are described as free on walohq.com. The FAQ says some rank data may be a small-cost add-on.
Should every new page have FAQ schema? FAQ content is useful on a service page because it creates answer-ready text and long-tail phrases. But do not expect a rich result from it: Google's structured-data documentation was updated in 2023 to say the FAQ rich result is only shown for well-known, authoritative government and health websites. Write the FAQ for the customer, add the schema if your builder produces it, and do not count on the search-result decoration.
How long before a new page ranks? Nobody can promise a number, and a tool that does is guessing. Pages have to be crawled, indexed, and then evaluated against the competitor's. Set a monthly check, not a daily one, and use the rank alert to tell you when something moves.
Where does the ranking payoff come from? Backlinko's study of about four million results found the first organic result averages a 27.6 percent click-through rate and is about ten times more likely to be clicked than position ten. The gap is worth closing when a page can realistically reach the top three for a search that sends jobs.
Where to learn more
The other WALO product articles in this batch on Nothing Unseen:
- Rank alerts and competitor watch, for tracking whether the new pages move and whether a copycat is passing you.
- IndexNow, sitemaps, and the real limit of auto-submitting pages, for what happens after you publish.
- Email campaigns for local service, for the follow-up side of the same business.
- Shorts content engines and WALO Creators: the agentic content desk, for the creator side of WALO.
The product itself: WALO, WALO SEO Pro, and the free website SEO scan. WALO is made by AMH, Artificial Mind Hive, and the guide pages it writes land on Nothing Unseen or Apiary.
Sources (the record)
- WALO SEO Pro: Rank Radar, Competitor Gap, keyword plan, Landing Page Builder, Free Guide Page — walohq.com
- WALO free website SEO scan page — walohq.com
- WALO FAQ, including "Is WALO free?" and rank-data add-on note — walohq.com
- WALO Connect page: "Connecting is not permission" — walohq.com
- How to improve your local ranking on Google (relevance, distance, prominence) — Google Business Profile Help
- Search Console Performance report: clicks, impressions, CTR, average position — Google Search Console Help
- Influencing your title links in search results — Google Search Central
- Control your snippets in search results — Google Search Central
- Local Business (LocalBusiness) structured data — Google Search Central
- FAQ (FAQPage) structured data, including the 2023 limit to authoritative government and health sites — Google Search Central
- Creating helpful, reliable, people-first content — Google Search Central
- Spam policies for Google web search, including scaled content abuse — Google Search Central
- Google Search's guidance about AI-generated content (Feb. 2023) — Google Search Central Blog
- JavaScript SEO basics (why pages that render only in the browser can look empty to crawlers) — Google Search Central
- Google organic CTR study: #1 result averages 27.6% CTR, about 10x position 10 — Backlinko
- Ahrefs pricing (Lite $129/mo, Standard $249/mo) — Ahrefs
- Moz Pro pricing (Standard $99/mo) — Moz
- BrightLocal pricing (from $41/mo for one location) — BrightLocal
- Semrush plans and pricing — Semrush
- Local Consumer Review Survey 2026 — BrightLocal
Originally published on Nothing Unseen, an Artificial Mind Hive (AMH) site. AMH is owned by Austin Little. Product pictures are screenshots of the live sites; tap any picture to open the product.