How we calculate estimates

The calculation pipeline

All five calculators share a single pipeline. Calculators contribute declarative data — a benchmark reference, adjustment factors, itemised cost blocks and breakdown shares — and never perform their own arithmetic.

• Your inputs are validated against the calculator definition.
• A benchmark band (low / typical / high) is selected for your scope and your market. Markets modelled separately, such as India, use their own ranges rather than a converted global figure.
• A scope quantity is applied where scale is continuous, such as screens or modules.
• Multiplicative adjustment factors are applied for size, design, platform and complexity, with the combined factor clamped so stacking answers cannot run away from the benchmark.
• Where a market has no dedicated benchmark, a market factor is applied in the base currency.
• Itemised cost blocks are added for each feature or service you selected, scaled to the size of the project.
• The estimate is positioned against budget, professional and premium agency levels derived from the same range.
• The result is allocated across breakdown categories.
• Only then is the figure converted to your display currency and rounded.

Why ranges rather than a single number

Real prices for identical scope vary by supplier, seniority mix, commercial terms and timing. A single number implies a precision that does not exist. Every result therefore reports a low, typical and high figure, and rounds to a step appropriate to the magnitude of the number, so we never present false precision.

How complexity affects price

Complexity enters the calculation in two distinct ways. Structural complexity — the type of product, its design ambition, its security posture — is modelled as a multiplicative factor on the whole estimate, because it affects every part of delivery. Discrete functionality — payments, chat, dashboards, reporting — is modelled as an additive block, because it is a bounded piece of work that can be added or removed.

How geography affects price

Country selection applies a market factor to the base estimate, expressed relative to a global blended baseline of 1.00. This factor is applied inside the calculation, in the base currency, before any conversion. Currency selection is presentational only and never changes the underlying estimate.

Currency treatment

All calculations are performed in USD as the base currency. Conversion happens as the final step. Current conversion source: Indicative reference rate table (not live FX), effective 2026-01-01. Because conversion is separate from methodology, changing currency never changes the underlying result — only its presentation.

Benchmark data

Benchmark records carry a category, subcategory, geography, currency, low/typical/high values, unit, source, effective date, review date, methodology version and status. Only records with published status represent reviewed market data.

The benchmark set in this build is indicative-2026.04. PriceInsight360 currently uses a structured, indicative benchmark dataset to power and test the pricing engine. These figures are indicative and are not presented as a survey of actual supplier quotes. Verified market records can be added and dated as the dataset grows, and any result drawing on indicative data is labelled as such on screen.

Versioning

Each result records the calculator version, methodology version and benchmark set used to produce it. Historical methodology is never modified in place: changes are released as a new version so past results remain interpretable.

Determinism and the role of AI

The engine is a pure function: the same inputs always produce the same output. There is no randomness, no clock dependency and no model inference in the numerical path. AI may later assist with explanation or advisory features, but it will never determine the core estimate.

Limitations

• Estimates are indicative and are not quotations, valuations or professional advice.
• Benchmark data in this build is an indicative dataset, not verified supplier quotes.
• Unclear requirements, legacy systems and regulated domains routinely exceed the high figure.
• Scope proxies such as page, screen and module counts cannot capture domain difficulty.
• Nothing on this site is influenced by any vendor, affiliate or advertiser.