NazmIQ reads your PDF and image quotes, levels every trade package onto one common scope against your BOQ or budget, flags what each bidder has excluded, and ranks them with weighted scoring — then hands you a two-sheet Excel breakdown and a shareable report.
Bid levelling, bid tabulation, quote comparison — whatever your team calls it, this is the job: from raw tender returns to a defensible recommendation, without spreadsheets or manual data entry.
Vision AI reads multi-page PDF quotes and even phone photos of printed quotations — pulling every line item, rate, quantity, unit, currency and term. No manual entry.
Synonymous items ("60×60 LED Panel" vs "LED Panel 600*600") are levelled onto one row and grouped into sections, so every bidder is priced against the same scope — apples-for-apples, not guesswork.
Items a bidder didn't price are counted per package and fed into the score — then filled with a rate carried from the most complete quote, so every total covers the same scope. A "cheap" quote can't win by omission and eat your margin later.
Score bidders on price, delivery, payment terms, scope and brand — with weights you control — and get a ranked result with a clear recommended winner. Totals, VAT and currency are computed deterministically in code, not guessed by the AI.
Export a Summary sheet (financials, evaluation matrix, delivery timeline, risks) plus a detailed breakdown with live formulas — adjust a rate or quantity and every total recomputes. Plus an interactive web report.
Forward quotes to the NazmIQ bot, organise them by trade package, run /compare, and get the Excel back — all from site or your phone.
Upload the tender returns and let NazmIQ do the levelling, the math, and the ranking.
Drag in PDFs, snap a photo of a printed quote, or send them to the Telegram bot. Add your BOQ, budget or pricing schedule to level against if you have one.
NazmIQ extracts every line item from each bidder and levels matching items onto one scope — handling different descriptions, units, quantities and currencies, and marking what each has excluded.
Deterministic math computes totals, tax and savings — lump sums and rate-based items alike; weighted scoring ranks the bidders and surfaces a recommended winner with the reasons why.
Download the two-sheet Excel, open the interactive report, or share it — in your chosen currency, ready for sign-off.
Unpriced items are the reason two totals cannot simply be set side by side — and levelling them is the judgement an estimator makes by hand on every tender.
A supplier leaves eight items unpriced. Their column still adds up — to a smaller number than everyone else's — but it is not a like-for-like figure: it buys less work. Until those gaps are priced there is no sound basis for comparison, because the totals are answering different questions.
The scope doesn't disappear. It returns later as variations, at whatever rate the supplier quotes then.
Every unpriced item is filled with the rate from the most complete quote, and the cell is shaded so the borrowed figure is never mistaken for something the supplier offered. Ranking uses that levelled total; the as-quoted total stays on the page beside it.
The report names the supplier each rate came from, counts the gaps per bidder, and warns you outright when more than half of a levelled total is imputed rather than quoted.
It is the same thing experienced buyers do in a spreadsheet — carry across a rate from a supplier you trust, colour the cell so everyone can see it, and compare totals that finally cover the same scope. See it on a real report →
The one-page decision view.
Every line, every bidder, fully auditable.