Keyword Golden Ratio: Useful Filter or False Precision?

TL;DR A green spreadsheet cell cannot tell you whether someone wants appointment software or help chasing missing documents. I would use the keyword golden ratio to nominate queries for inspection, then judge the actual results, reader task, and contribution our site can afford. The calculation should inform a publishing decision; it should never make that decision alone.

Imagine a hypothetical founder building software that collects overdue client documents. A keyword spreadsheet highlights “automated client reminders.” The wording sounds relevant, the assumed ratio passes, and the founder starts outlining a post.

But reminders for what? Attending a meeting and submitting a missing file are different jobs. A mathematically correct calculation leaves that distinction completely unresolved.

That is my objection to treating a research shortcut as a publishing instruction. For a bootstrapped SaaS, the expensive mistake is spending scarce writing time on an answer the intended customer does not need.

What the keyword golden ratio actually calculates

Doug Cunnington's original explanation of the heuristic defines it as the number of Google allintitle results divided by local monthly search volume. His traditional convention looks for a ratio below 0.25 with monthly volume below 250. Those cutoffs belong to his method; they are not Google rules.

The keyword golden ratio formula is straightforward:

KGR = reported allintitle result count / local monthly search volume

In Cunnington's method, allintitle supplies a title-word search count. I would not describe that count as an exact-phrase census of every competing page. The method and its title-search procedure are documented in Cunnington's walkthrough.

If you arrived searching for “golden keyword ratio,” this is the calculation discussed here. I will use KGR keywords to mean candidates passing that conventional filter, without treating the label as evidence that our site can rank.

My limited case for the keyword golden ratio is procedural: it gives a researcher a repeatable way to shortlist candidates. The benefit ends where someone reads “shortlisted” as “approved.” There is no input for the page you intend to build, the product it supports, or the resources available to make it useful.

A hypothetical calculation exposes the sensitivity

Every input in this example is made up. These are arithmetic illustrations, not measurements of either query discussed in this article.

Hypothetical input	Scenario A	Scenario B
Assumed allintitle count	20	20
Assumed monthly volume estimate	100	50
Calculated ratio	0.20	0.40

The assumed numerator stays fixed. Halving the denominator doubles the ratio. Under Cunnington's below-0.25 convention, the first scenario passes and the second does not.

Nothing about the proposed article changed. We did not discover a stronger competitor, a worse product fit, or a missing example. We changed an input and crossed a decision boundary.

That does not make division wrong. It makes a rigid publishing rule difficult to defend when the denominator is uncertain. I would record both the estimate and the judgment it influenced, so a later revision does not silently rewrite the investment case.

A keyword golden ratio calculator can reproduce the arithmetic in a spreadsheet. Here is a copyable version, with A2 holding the reported count and B2 holding the volume estimate:

=IF(OR(A2="",B2=""),"missing input",IF(B2=0,"undefined",A2/B2))

Use nonnegative numeric inputs. A missing value stays missing; division by zero is undefined. Neither state should turn into a passing score. Extra decimal places would add display precision without supplying new evidence.

Where the keyword golden ratio leaves uncertainty

The denominator needs a definition

Google's Keyword Planner defines average monthly searches using the keyword and close variants, the selected month range, location, and Search Network settings. Its historical volume figures are rounded. These are Keyword Planner's documented characteristics, not a claim that every keyword provider uses identical methods.

My practical rule is to keep the provider, lookup date, market, and measurement definition beside the number. Do not combine a global estimate with a local estimate and interpret the resulting score change as new competitive evidence.

I would also resist rejecting a tightly relevant topic solely because a smaller denominator makes its ratio worse. Under the hypothetical arithmetic above, greater estimated demand improved the score without changing the assumed competition. That is a property of the formula, not independent evidence of an easier opportunity.

Title words do not define the competitive field

Google says RankBrain helps it return relevant content even when that content does not contain all the exact words in a search. Its ranking documentation describes multiple systems, including systems that understand relationships between concepts and queries. See Google's ranking systems guide.

My inference is that a title-restricted count cannot stand in for the ordinary results a prospective reader encounters. Relevant pages using different wording still belong in my review.

Nor does the formula evaluate the strength of the answers it counts. Imagine identical reported counts for a field of shallow summaries and a field of detailed, demonstrated workflows. KGR has no variable that distinguishes those hypothetical situations.

The keyword difficulty explainer covers a separate question: how to interpret a provider's difficulty score. Here, the issue is this specific title-count-to-volume shortcut. Substituting another unexplained number would leave the publishing judgment unfinished.

The ratio has no ranking clock

I would not attach a ranking deadline to a passing keyword golden ratio. Google explicitly says it does not guarantee crawling, indexing, or serving a page, even when that page follows Search Essentials. See Google's explanation of how Search works.

A threshold can define membership in someone's research method. It cannot, by itself, establish when your proposed page will earn visibility.

What current results reveal that the ratio misses

The examples below use a September 17, 2026, around 06:42 UTC capture of Google US-English desktop results through DataForSEO. The saved research evidence contains the returned entries and search settings. It is a dated sample, not permanent rankings.

No real allintitle counts or candidate search volumes were collected for these queries. This is an inspection of omitted considerations, not a test of predictive ranking accuracy or proof that either candidate satisfies KGR. Those limits accompany the research snapshot.

Google documents contextual variation in results, including location, language, and device, even without personalization. That is another reason to preserve the search conditions.

A checklist query asks for a usable deliverable

For “client onboarding checklist,” the sample includes guides and template destinations, including Rocketlane, Karbon, and Zapier. That is a result-composition observation from the checklist query capture.

The inspected Karbon template page describes sections for kickoff, systems setup, information collection, financial preparation, and review. It presents download and add-to-Karbon options. That describes the destination, without endorsing its outcome claims.

My inference: an instructional resource with something usable is a defensible starting format. A hypothetical document-collection SaaS would still need to decide which onboarding task it can explain convincingly.

The keyword golden ratio cannot tell me whether our proposed checklist offers a distinct contribution. I would require a specific scope and a worked example before approving it. Narrowing to document collection is a possible research direction, not a verified gap in these results.

A reminder query crosses product categories

For “automated client reminders,” the capture includes Appointment Reminder and SMS Client Reminders, alongside a YouTube entry, a Reddit discussion, vendor material, and explanatory resources. See the reminder query capture.

The inspected SMS Client Reminders homepage presents automatic text and email appointment reminders, with calendar connection or manual appointment entry. That supports identifying appointment scheduling as a task represented here.

My inference for our hypothetical document-chasing product: the broad phrase carries a meaningful risk of task mismatch. An article about missing files would need a stronger justification than shared words such as “client” and “reminders.”

Even if an assumed ratio passed, I would defer this broad assignment and investigate wording that explicitly names documents. That alternative remains unmeasured. The varied formats also give me no basis to declare the query easy, or to infer weak competition from the presence of a discussion.

These observations test the completeness of the planning shortcut. They do not demonstrate that KGR succeeds or fails at predicting rankings.

Use the filter without letting it own the calendar

I would keep the keyword golden ratio as an optional annotation inside a broader low-competition keyword research process. My sequence would be:

1. Name the customer problem. Reject candidates whose connection to the product requires an unrelated sales pitch.
2. Record the calculation and inputs. Preserve unknowns, provider definitions, and dates. Keep a passing label separate from approval.
3. Inspect ordinary results and destinations. Describe the represented tasks, formats, and answer quality using the SERP analysis worksheet.
4. Specify an affordable contribution. Identify the example, template, demonstration, or analysis we can actually deliver. Leave competitive strength unresolved where evidence is missing.
5. Assign the owning page. Update an existing answer when it already owns the task; create a new page only for a justified deliverable.

A keyword golden ratio tool would be useful to me only if I could inspect its inputs and carry uncertainty forward. A colored badge without that record would not earn a writing slot.

Here is the compact record I would copy beside a candidate:

Query and intended customer:
Product problem and reader task:
Allintitle query / reported count / capture date:
Volume provider / estimate / period / market:
KGR calculation, or missing input / undefined:
Threshold convention used:
Ordinary SERP capture and inspected destinations:
Observed formats and tasks:
My interpretation and unresolved evidence:
Contribution we can demonstrate:
Existing owning URL / proposed cluster:
Decision: write / update / defer / skip
Work budget and condition for reconsideration:

Let page ownership outrank a passing score

For hypothetical phrases such as “request missing client documents” and “follow up on incomplete client files,” I would first ask whether the same worked guide answers both. Separate spreadsheet rows do not oblige me to create separate articles.

Use a keyword map with explicit page owners to connect related tasks without commissioning repeated answers. The calendar should show the work needed: improve a guide, produce a missing template, investigate an uncertainty, or leave the slot open.

This is the planning problem I am building Boomranq around. It is in development, with the purpose of turning a product description into a winnability-first, 30-day SEO content calendar for small, low-authority SaaS sites. I want the plan to make the reasoning behind an assignment visible before a founder spends time drafting.

My verdict on the keyword golden ratio: keep it if it helps organize inspection. Refuse the false precision when a decimal starts deciding what your business should publish. The useful endpoint is a defensible page assignment with a clear customer task and an affordable contribution.