Useful recommendations need visible reasoning.
SolutionRay separates facts, editorial judgment, hard requirements and preferences, so you can understand the result and challenge it.
Research current sources
We prioritize official pricing pages, feature documentation, legal terms, DPAs, security pages and subprocessor lists. We paraphrase them and record when a fact was checked.
Keep unknowns unknown
If data residency, pricing or a feature cannot be confirmed, it is marked "Not verified." We do not turn missing information into a "No," and we do not infer EU-only storage from GDPR support.
Apply hard filters first
Each category has its own hard requirements. Customer Support first checks a current qualifying plan's capabilities, agent capacity, recurring-usage completeness, commercial price and required documentation. Hosting checks workload support, mandatory operational capabilities, ongoing budget and explicitly required EU/EEA-only hosting. CRM & Sales checks mandatory capabilities, documented integrations, seat capacity, total active-user cost and explicitly required EU/EEA-only processing. Scores never override a hard conflict.
Score applicable preferences
Each branch activates only criteria that matter to the answers. Irrelevant criteria are removed entirely, then the remaining active weights are normalized to 100%. Automation, privacy or budget never receive points when the answer makes them irrelevant.
Show limitations
Every recommendation can include trade-offs and missing evidence. "Things to consider" is part of the result, not fine print.
Affiliate commission weight
Affiliate status never improves rank or breaks a tie. Ties use the visitor's most specific requirements, or are displayed as equally strong when they remain equal.
Different categories use different rubrics
Customer Support, Hosting and CRM & Sales do not share a generic scorecard. Every rubric uses the same 0-5 meaning: 0 is no verified capability; 1 is incidental; 2 is limited or add-on dependent; 3 is solid; 4 is strong; and 5 is unusually complete or category-leading.
CRM & Sales methodology
The CRM branch evaluates contact management, pipelines, ease of use, automation, email integration, outreach, reporting and forecasting, customization, integrations and API access, connected marketing and support workflows, scalability and SMB value. The selected primary job, sales-process complexity and top priority determine the active score. Automation weight changes with the answer and is removed when it is not important.
Mandatory capabilities and integrations pass only when supported by current official provider documentation. A missing source does not pass. Budget checks use the total for all active users at the cheapest researched plan that supports the selected workflow and must-haves. Plan name, per-seat basis, total and material limits are shown. GDPR documentation, an available European data region and verified EU/EEA-only processing remain separate facts.
Marketing Automation & Email methodology
The marketing branch evaluates campaign email, automation, customer journeys, segmentation, personalization, forms, ecommerce lifecycle fit, CRM connection, integrations, reporting, channel coverage, transactional email, ease, value and scalability. Only criteria relevant to the answers are active, and their remaining weights are normalized dynamically.
Ray selects the cheapest researched offering that documents the required automation depth, contacts, send volume, ecommerce or CRM relationship, channels and transactional email. Contact-based, send-based and add-on pricing is shown separately when it prevents a verified total. GDPR documentation, an available European region and verified EU/EEA-only processing remain separate facts.
Ecommerce Platforms methodology
The Ecommerce branch compares storefront and catalog strength, ease, technical control, integrations and API access, payments, international commerce, B2B, POS, unified retail, scalability, implementation support, regional relevance, value and migration. It activates only dimensions relevant to the business situation, catalog, retail model, technical preference, international plans and priority.
Catalog management follows one provider-independent 0-5 scale: 0 means no verified catalog capability; 1 covers a basic product list; 2 supports a small catalog with limited administration; 3 supports established catalog administration, variants and categories; 4 adds documented bulk import/export, price-list or rules support, and integrations or API access; 5 requires verified native multi-store catalog management, PIM-grade catalog orchestration, or equivalent enterprise catalog architecture across storefronts, regions or business units. Standard B2B accounts, price lists and purchasing rules alone do not reach 5. Product-count capacity alone does not determine this score.
Scalability measures a documented growth path for traffic, transactions, operations or markets without a forced platform migration. It does not award points for brand recognition, product count, price or integration count. B2B depth measures the native platform or a qualifying first-party B2B offering, such as company accounts, customer pricing, payment terms, buyer roles or approval workflows. A third-party extension can meet a selected B2B hard requirement only when it is explicitly included as the qualifying solution; it does not increase the native B2B score.
An optional integrated product can satisfy a hard functional requirement only when the qualifying offering documents that integration and its separate commercial basis is made clear. Native platform-depth scores, including catalog management, measure the platform itself and do not inherit a connected ERP, PIM or POS product’s score. Platform subscription, hosting, extensions, payment processing, POS and implementation are not treated as one false total. Ray identifies a qualifying commercial offering where public documentation supports one, then labels usage-dependent, quote-based and implementation-dependent costs clearly.
Hosting methodology
The Hosting branch evaluates managed hosting, WordPress, WooCommerce, custom-site and application support, developer control, scalability, performance tooling, backups, staging, security, support, ease, long-term value and geographic options. Workload type determines the comparison pool: managed WordPress and conventional SMB hosting are not ranked as direct substitutes for developer cloud or edge platforms.
For advanced cloud requests, Ray evaluates one coherent product, location and architecture configuration. It checks the selected country or city, documented regional service availability, multi-zone resilience where required, and required managed services such as Kubernetes, load balancing, private networking, storage or databases. A capability in one region cannot satisfy a request for another region.
Budget checks use the normal ongoing or renewal price. Known required components, such as a public IPv4 address, are added to the minimum configuration cost. Introductory promotions are labelled with their commitment and renewal basis. Usage-priced infrastructure keeps an explicit "usage-dependent" cost when the questionnaire cannot support a credible total. GDPR materials, EU storage, EU processing and strict EU/EEA-only processing remain separate checks.
Cross-currency Hosting budgets
Hosting prices remain in the provider’s published currency. When your selected budget uses another supported currency, Ray compares them through the European Central Bank (ECB) euro foreign-exchange reference rates from the ECB EXR dataset. These reference rates are normally published on working days around 16:00 CET; the most recent expected rate remains valid over weekends and TARGET closing days.
Cross-currency hard budgets use the stored ECB snapshot and a 4% safety margin on the converted cost. This avoids presenting a price that is only a few cents below a strict ceiling as safely affordable. The margin is not applied when the price and budget share a currency, and it never changes the provider’s native price. ECB rates are reference rates, not a prediction of your bank, card-network or settlement rate; tax treatment remains exactly as the provider publishes it.
Each saved result keeps the ECB source, rate date and converted decision context used when it was created, so it is not silently recalculated with later rates. If no current or acceptably fresh ECB snapshot is available, cross-currency hard-budget eligibility is left unverified. Known usage-dependent or explicitly unknown components remain unknown after conversion.
Hosting configuration pricing
Ray does not treat an hourly unit, free control plane or starting price as a full hosting configuration. A finite budget passes only when every mandatory component has a documented quantity and recurring price. Where an explicitly always-on resource has a verified hourly unit and no provider monthly cap, Ray normalizes it using 730.5 hours per month (365.25 days ÷ 12 × 24). Provider-published monthly caps take precedence.
Ray may show a known configuration floor when some usage remains open, but never uses that floor to pass a hard budget. Storage, databases, workers, serverless requests, traffic-based charges and other components stay unverified until the selected requirements supply a documented minimum or a real quantity. Native components are summed before any ECB conversion; the existing 4% cross-currency safety margin is applied once to the completed configuration. Tax treatment remains as published by each provider.
Text product-family modelling
SolutionRay scores documented product fit without a vendor penalty, then avoids duplicate entry points to the same effective subscription in one shortlist. LiveChat retains historical plans for existing customers, while its current new-trial path leads to Text. The current ChatBot.com and HelpDesk.com pricing surfaces also lead to the same Text subscription. For new-customer recommendations, Ray therefore uses the verified current Text plan rather than presenting these entry points as separate competing subscriptions.
Research scope and limitations
CRM & Sales information was checked against official provider sources on 17 September 2026. Customer Support and Hosting information retains its existing research dates. A selectable European region is not automatically treated as verified EU/EEA-only processing. No product is described as hands-on tested.