How to Build a SaaS Keyword Opportunity Score

Define the decision before choosing a formula. A discovery score ranks unreviewed phrases for further research. A content roadmap score ranks approved page opportunities. A quick-win score finds existing pages with a plausible path to near-term improvement. Mixing these objectives creates a number that cannot explain what action should follow.

Set eligibility rules outside the score. Exclude queries the product cannot serve, markets the company does not support, legally restricted claims, and phrases with an intent the proposed page cannot satisfy. Flag uncertain volume, ambiguous meaning, and missing position data as review needs rather than silently substituting false precision. Only qualified candidates should compete on the weighted dimensions.

Monthly volume approximates demand, but raw volume spans large ranges and can overwhelm a linear formula. Use a logarithmic or banded transformation so a phrase with 10,000 searches is rewarded without automatically outranking every 500-search commercial query. Store the source, market, device assumptions if known, and collection date. Volume estimates are directional and often less reliable in specialized B2B categories.

Tiptop uses a logarithmic volume component, which reflects diminishing returns as traffic estimates grow. Keep the raw number visible beside the normalized contribution. Add a confidence flag when the estimate is old, combines multiple countries, or contradicts first-party evidence. A score becomes misleading when a neat transformed value hides a weak underlying measurement.

Difficulty should represent the relative effort needed to compete, not the probability of ranking. If third-party result data is unavailable, phrase shape provides a transparent heuristic. Short head terms are usually broader and harder; additional qualifiers tend to narrow competition. Commercial modifiers can increase competition, while question, template, example, checklist, and tutorial language may identify more specific needs.

Tiptop starts at high difficulty for a head phrase, reduces the estimate for each additional term, adjusts upward for commercial wording, and downward for questions and useful resource modifiers. The result is bounded rather than allowed to produce impossible values. Validate the shortlist manually against the current results, domain strength, content quality, and page types. The heuristic is valuable because its logic is visible, not because it knows every competitor.

Classify intent from the phrase and confirmed search results. Transactional terms such as pricing, demo, trial, and sign-up deserve a larger money-intent contribution because the searcher signals an immediate action. Commercial terms such as best, versus, alternative, review, software, tool, and platform indicate active evaluation. Informational questions can still matter when they introduce a costly problem, enable product use, or build a necessary cluster.

Avoid treating every commercial modifier as equally valuable. A competitor alternative query may be relevant but require substantial proof and legal review. A pricing phrase can be valuable only if the company is willing to answer it. Add a zero-to-five business-fit rating based on ideal-customer relevance, product connection, conversion path, and expected contract value. Require a written reason for the highest and lowest ratings to discourage convenient scoring.

Once a phrase maps to an existing or proposed URL, add three company-specific inputs. Authority fit measures whether the site has relevant expertise, links, and supporting content. Current position distinguishes a page near the first page from a concept with no indexed asset. Effort estimates writing, design, product evidence, engineering, review, and promotion, not merely word count.

A practical roadmap score can allocate 25 points to business fit, 20 to traffic potential, 15 to intent, 15 to attainability, 15 to existing momentum, and 10 to inverse effort. Adjust weights before seeing candidate names. For a new domain, authority and cluster support may deserve more weight. For an established site with many near-page-one rankings, current position and low implementation effort can lead the quick-win model.

Suppose a phrase earns 23 of 25 for business fit, 13 of 20 for traffic, 13 of 15 for intent, 10 of 15 for attainability, 12 of 15 for current momentum, and 7 of 10 for inverse effort. Its total is 78 of 100. The breakdown says more than the total: the opportunity is commercially strong and already has momentum, while moderate demand and production work constrain it.

Use bands for workflow rather than truth claims. Scores of 75 and above may enter validation and briefing, 60 to 74 may remain qualified reserves, 45 to 59 may need a cheaper angle or stronger cluster, and lower scores may be declined for the current cycle. Do not describe 78 as a 78 percent chance of success. It is a relative rank produced by your evidence and weights.

Create a quick-win view by filtering for meaningful volume, lower estimated difficulty, a mapped URL, and a current position close enough to improve. Tiptop counts phrases with difficulty at or below 35 and volume of at least 100 as quick wins in its cluster summary. Treat that as a discovery rule. Verify that a suitable page exists, the intent matches, and the needed change is smaller than creating a stronger page from scratch.

After each cycle, compare scores with impressions, ranking movement, qualified visits, assisted conversions, delivery time, and page quality. Look for patterns across several releases. If high-volume informational pages repeatedly underperform commercial pages for the business goal, adjust the relevant weight. Preserve formula versions and original scores so the team can distinguish model changes from genuine changes in the opportunity.