Scoring

Model Confidence model_confidence

How sure the model is about its own choice or a fact in it. Low confidence often leads to hedged wording, extra cautions or advice to call and check.

What you can do: Reduce doubt by keeping core facts clear and consistent across your site and listings.

evaluation.scoring · also called prediction_certainty, output_reliability

Checked by What AI answers say your phone, address and hours are · Term: confidence-score

Reputation Strength reputation_strength

A scoring dimension for how strong a business's reputation is overall, from ratings, review volume, complaint history and outside mentions. A well-reviewed long-standing plumber scores high.

What you can do: Build a steady review record, resolve complaints visibly and keep review profiles linked from the site.

reputation.evidence · also called credibility_index, trustworthiness_level

Used in 6 of 100 census runs this week: “0.05”

Responsiveness responsiveness

A scoring dimension for how quickly and reliably a business answers enquiries and turns up, judged from reviews mentioning response and stated response times.

What you can do: State response times, keep a reachable phone number and reply to reviews and messages promptly.

evaluation.scoring · also called reaction_time, engagement_speed

Trust And Credentials trust_and_credentials

A scoring dimension for licences, insurance, certifications and endorsements the model could confirm, such as a licensed and insured tree service with an arborist certification.

What you can do: Publish licence and insurance details and certifications with numbers and links to public lookups.

credentials.status · also called qualification_assessment, endorsement_strength

Term: e-e-a-t

Used in 15 of 100 census runs this week: “Licensed, insured, and member of industry associations (e.g., National Garage Door Association).”

Universal Total universal_total

The model's combined score for a candidate across all its scoring dimensions, used to rank the shortlist and pick the business it recommends first.

evaluation.scoring · also called overall_rating, comprehensive_score

Used in 7 of 100 census runs this week: “4.9”

Local Relevance local_relevance

A scoring dimension for how well a business fits the user's location: based nearby, serving that city, with local reviews. A Denver fence company scores higher for Denver than one an hour away.

What you can do: Name your city and served areas in text, keep a local address on listings and ask local customers for reviews.

evaluation.scoring · also called geographic_fit, community_relevance

Checked by The places your site names against your Business Profile · Term: service-area

Used in 4 of 100 census runs this week: “Must provide garage door repair services in Tampa, FL.”

Service Fit service_fit

A scoring dimension for how closely a business's services match the request, such as a tree service that does stump grinding when the user asked for stump removal.

What you can do: Give every service its own page with clear names customers use, and match listing categories to them.

service.fit · also called offering_alignment, service_compatibility

Used in 1 of 100 census runs this week: “concrete contractor”

Universal Score Dimensions universal_score_dimensions

The list of dimensions the model scores every candidate on, such as reputation, credentials, local fit and price, before combining them into a total.

evaluation.scoring · also called evaluation_criteria, scoring_factors

Risk Penalty risk_penalty

Points the model subtracts for risk, such as unresolved complaints, no licence found, conflicting contact details or signs of closure. A high penalty can drop an otherwise good candidate.

What you can do: Remove the red flags: fix conflicting details, answer complaints, show licence and insurance, and clear closed flags.

evaluation.scoring · also called risk_assessment, penalty_factor

Checked by Field-by-field audit of your listings

Emergency Score Stack emergency_score_stack

The set of scores the model uses for an urgent query, such as a burst pipe, weighting 24/7 availability, response time and nearby location above price.

What you can do: If you take emergencies, state hours, response times and area, and keep hours on every listing accurate.

evaluation.scoring · also called crisis_response_score, emergency_preparedness

Checked by Your hours, day by day, everywhere they are printed

Value Score Stack value_score_stack

The set of scores the model uses for a value-for-money query, weighting price clarity and fair cost against quality evidence such as reviews and guarantees.

What you can do: Publish prices or ranges next to what is included and any guarantee.

evaluation.scoring (was offer.fit in the provisional inventory: A score stack of weights, like the other stacks.) · also called value_assessment, cost_benefit_score · same label written as costo_beneficio_score_stack

Checked by The claims your site makes: years, prices, refund window

Project Score Stack project_score_stack

The set of scores the model uses for a planned project such as a kitchen remodel, weighting portfolio, credentials, project management and reviews of finished jobs.

What you can do: Show finished projects with photos, dates, scope and locations, and describe your process and timeline.

evaluation.scoring · also called project_performance_score, project_evaluation

Amenity Fit amenity_fit

A scoring dimension for how well a business's facilities match what the user asked for, such as full hookups at an RV park or a fenced play area at a dog daycare.

What you can do: List amenities in text on the site and in listing attributes so they can be matched.

evaluation.scoring · also called amenity_compatibility, amenity_alignment

Budget Score Stack budget_score_stack

The set of scores the model uses for a budget query, weighting low or clear prices, no hidden fees and acceptable quality evidence.

What you can do: Publish entry-level prices and say clearly what is and is not included.

evaluation.scoring (was offer.fit in the provisional inventory: A score stack of weights, like the other stacks.) · also called budget_alignment, financial_fit

Checked by The claims your site makes: years, prices, refund window

Buyer Score Stack buyer_score_stack

In scoring, the set of scores the model stacks up when the searcher is buying, such as a home buyer choosing a real estate agent: buyer-side experience, local market knowledge and guidance through offers.

What you can do: Have a buyer page stating how many buyers you represented, in which neighborhoods, and what the process looks like from first showing to closing.

evaluation.scoring · also called buyer_preference_score, buyer_insight

Corporate Score Stack corporate_score_stack

In scoring, the stack of scores used when the query is for a business client, such as an office cleaner or corporate caterer: capacity, professionalism, invoicing and experience with companies.

What you can do: Publish a commercial or corporate page naming business clients served, team size, insurance, and how you handle contracts and invoicing.

evaluation.scoring · also called corporate_performance_score, business_standing

Seller Score Stack seller_score_stack

In scoring, the stack of scores used when the searcher is selling, such as a homeowner picking a listing agent: track record selling, pricing advice, marketing and time to sale.

What you can do: State listing results on the site: homes sold, typical days on market, areas covered, and what your marketing for a listing includes.

evaluation.scoring · also called seller_performance_score, seller_reliability

Amenity And Service Fit amenity_and_service_fit

In scoring, how well a candidate's facilities and services together match what the query asked for, such as an RV park with full hookups and pet areas, or a gym with childcare.

What you can do: List every amenity and service in plain text on one page, matching the words customers search for, and keep it current with your listings.

service.fit · also called amenity_service_compatibility, service_amenity_alignment

Amenity Completeness amenity_completeness

In scoring, how many of the expected amenities a candidate actually offers, such as a campground having showers, Wi-Fi, laundry and dump station, rather than just one or two.

What you can do: Publish a full amenity checklist on your site, including the basic ones you might assume are obvious, so nothing reads as missing.

offer.fit (was evaluation.scoring in the provisional inventory: Amenities offered are an offer match, not a score) · also called amenity_range, amenity_fulfillment

Amenity Quality amenity_quality

In scoring, how good and well kept a candidate's amenities appear, judged from photos, reviews and descriptions, such as clean, recently renovated bathrooms at a campground or venue.

What you can do: Show recent dated photos of facilities, note renovation dates, and respond to reviews that mention condition problems.

offer.fit (was evaluation.scoring in the provisional inventory: Amenity quality is an offer attribute) · also called amenity_standard, amenity_condition

Beginner Compatibility beginner_compatibility

In scoring, how well a candidate suits someone new, such as a climbing gym with intro classes or a dive shop teaching first-timers, versus one aimed at experienced customers.

What you can do: Have a page for first-timers: intro classes, what to bring, equipment provided, and what a first visit looks like.

service.fit (was evaluation.scoring in the provisional inventory: Whether candidate serves beginners is service fit) · also called beginner_friendly, novice_compatibility

Community Verification community_verification

In scoring, whether reviews, forums and local groups back up what a candidate says about itself, such as customers confirming a tree service really does same-day storm cleanup.

What you can do: Make sure claims on your site match what reviews say; ask customers to mention the specific service they received.

claims.verification · also called community_validation, peer_verification

Consulta Detallada Score Stack consulta_detallada_score_stack

In scoring (label written in Spanish, consulta detallada), the scores used when the user wants a thorough consultation, such as a clinic or lawyer offering an in-depth first appointment rather than a quick quote.

What you can do: Describe the consultation itself: length, what is reviewed, who attends, cost or whether it is free, and how to book it.

evaluation.scoring · also called detailed_consultation_score, consultation_quality

Cosmetic Score Stack cosmetic_score_stack

In scoring, the stack of scores used when the query is about looks, such as a cosmetic dentist, med spa or home stager: visible results, before-and-after work and style.

What you can do: Publish a before-and-after gallery with captions naming the procedure or service, and list practitioners' cosmetic credentials.

evaluation.scoring · also called aesthetic_score, visual_appeal

Design Innovation design_innovation

In scoring, how original and distinctive a candidate's design work looks, such as a landscaper or kitchen remodeler showing custom, non-template projects.

What you can do: Show a portfolio of custom projects with short write-ups on the design problem and the solution you chose.

offer.fit (was evaluation.scoring in the provisional inventory: Design style and originality is an offer attribute) · also called design_creativity, innovation_score

Design Score Stack design_score_stack

In scoring, the stack of scores used when design is the priority, such as an interior designer or custom home builder: portfolio quality, style range and how well designs work in practice.

What you can do: Publish a portfolio organized by style and project type, with photos, scope and location for each project.

evaluation.scoring · also called design_quality_score, design_effectiveness

Documentation Clarity documentation_clarity

In scoring, how clearly a candidate explains its requirements, policies or paperwork, such as a lender listing required documents or a contractor explaining permits and contracts.

What you can do: Put requirements, required documents and policies on clear pages in plain text, not only in PDFs or images.

evidence.assessment (was evaluation.scoring in the provisional inventory: Judges quality of candidate's documentation) · also called documentation_quality, clarity_assessment

Economic Score Stack economic_score_stack

In scoring, the stack of scores used when the user wants the economical option, such as an inexpensive mover or budget auto repair: low price, fair fees and no surprise costs.

What you can do: Publish starting prices and any fees up front, and say what is included at the lower price point.

evaluation.scoring (was offer.fit in the provisional inventory: Score stack is a scoring profile, not an offer match) · also called economic_viability_score, sustainability_assessment

Checked by The claims your site makes: years, prices, refund window

Eligibility Flexibility eligibility_flexibility

In scoring, how flexible a candidate's entry requirements are, such as a lender accepting lower credit scores or a school taking students without prior experience.

What you can do: State eligibility rules plainly: minimum requirements, exceptions you accept, and who should still contact you.

service.fit (was evaluation.scoring in the provisional inventory: Whether candidate accepts this customer is service fit) · also called inclusivity, accessibility

Used in 1 of 100 census runs this week: “online appointment booking, hours convenient for typical owners, accepts major pet insurance”

Family Value Score Stack family_value_score_stack

In scoring, the stack of scores used when a family is the customer, such as a pediatric dentist or family campground: kid-friendly service, safety, space and family pricing.

What you can do: Say clearly what you offer families: kids' areas, family packages, child safety measures, and ages served.

evaluation.scoring (was offer.fit in the provisional inventory: Score stack is a scoring profile, not an offer match) · also called family importance, relationship value

High End Score Stack high_end_score_stack

In scoring, the stack of scores used when the user wants the top end, such as a luxury home builder or high-end florist: premium materials, exclusive service and upscale reputation.

What you can do: Describe premium materials, white-glove services and notable high-end projects, with photos and specifics rather than adjectives.

evaluation.scoring · also called luxury score, premium attributes · same label written as premium_score_stack

High Value Score Stack high_value_score_stack

In scoring, the stack of scores used when the user wants strong value, weighing a candidate's quality and inclusions against price, such as a roofer offering a long warranty at a mid-range price.

What you can do: State what makes the price worth it: warranty length, inclusions, and price ranges side by side.

evaluation.scoring (was offer.fit in the provisional inventory: Score stack is a scoring profile, not an offer match) · also called value assessment, benefit score

Checked by The claims your site makes: years, prices, refund window

Maintenance Ease maintenance_ease

In scoring, how easy the product or result is to look after, such as low-maintenance landscaping, artificial turf or a finish that needs no resealing.

What you can do: Explain the upkeep each option needs, with care instructions and maintenance plans if you offer them.

offer.fit (was evaluation.scoring in the provisional inventory: Maintenance ease is a property of the offered result) · also called ease of upkeep, maintenance simplicity

Marketing And Vendor Score Stack marketing_and_vendor_score_stack

In scoring, the stack of scores used when the user is a business looking for a marketing agency or supplier: proven results, reliability and fit with their trade.

What you can do: Publish case studies with named clients, measurable outcomes and the industries you serve.

evaluation.scoring · also called vendor reliability, marketing effectiveness

Performance Consistency performance_consistency

In scoring, whether a candidate performs the same way job after job, judged from steady reviews and history, such as a cleaning company praised consistently over years rather than in bursts.

What you can do: Keep a steady stream of recent reviews and show long-running client relationships or years in business.

reputation.evidence (was evaluation.scoring in the provisional inventory: Consistency judged from steady reviews and history) · also called reliability score, performance stability

Planned Policy Score Stack planned_policy_score_stack

In scoring, the stack of scores used for a planned, non-urgent policy purchase, such as an insurance agent helping someone compare coverage ahead of time.

What you can do: List the policy types you write, carriers you work with, and how a quote and review appointment works.

evaluation.scoring · also called policy foresight, strategic planning

Process Efficiency process_efficiency

In scoring, how quickly and smoothly a candidate gets the job done, such as fast quotes, online booking and short turnaround for a print shop or loan office.

What you can do: State turnaround times, booking steps and response times on the site, and keep them accurate.

service.fit (was evaluation.scoring in the provisional inventory: How the candidate delivers the job) · also called operational efficiency, workflow effectiveness

Quality Score Stack quality_score_stack

In scoring, the stack of scores used when the user wants the best quality regardless of price, such as top-rated electricians: workmanship, credentials, reviews and warranty.

What you can do: Show credentials, warranty terms and detailed project examples; ask reviewers to describe the work itself.

evaluation.scoring · also called quality assessment, standard rating

Used in 1 of 100 census runs this week: “Rating/5 * 0.6 + (review_count/100) * 0.4”

Regulation And Compliance regulation_and_compliance

In scoring, whether a candidate appears to follow the laws and licensing rules for its trade, such as a licensed, bonded electrician or a permitted daycare.

What you can do: Show license numbers, issuing authority and permits on the site so they can be checked against the regulator.

credentials.status (was evaluation.scoring in the provisional inventory: Licensing, bonding and permits are credentials) · also called compliance assessment, regulatory adherence · same label written as regulatory_score_stack

Risk Groundedness risk_groundedness

In scoring, whether the risks the model notes about a candidate rest on real evidence, such as a documented complaint, rather than guesses, so a tree service is not penalised on speculation.

What you can do: Address known issues publicly, such as replying to complaints, and keep facts on your site consistent so there is nothing to speculate about.

evidence.assessment (was claims.verification in the provisional inventory: Judges whether noted risks rest on evidence) · also called risk awareness, risk management

Term: grounding

Transparent Pricing Score Stack transparent_pricing_score_stack

In scoring, the stack of scores used when the user cares about clear prices, such as a dentist or mechanic publishing fees: listed prices, no hidden fees and written quotes.

What you can do: Publish prices or ranges, fees and what is included in plain text on the site.

evaluation.scoring (was offer.fit in the provisional inventory: Score stack is a scoring profile, not an offer match) · also called pricing transparency, clear pricing

Checked by The claims your site makes: years, prices, refund window

Used in 3 of 100 census runs this week: “0‑10 for clear pricing tiers”

Wedding Score Stack wedding_score_stack

In scoring, the stack of scores used for a wedding query, such as a florist, DJ or venue: wedding experience, capacity, availability and reviews from couples.

What you can do: Have a dedicated wedding page with packages, capacity, past weddings and how far ahead to book.

evaluation.scoring · also called wedding relevance, bridal score