AI use case · Construction

AI tender and BOQ analysis for contractors, with pricing left to the estimator.

For estimators and tendering teams who read every tender pack line by line against a fixed submission date. AI turns the BOQ into structured items, summarises the conditions and flags gaps. Your estimator prices and decides.

Ahmed Khaled, FounderLast updated:

Discuss this use case

A short note is enough. Phone and email let us reach you.

About: AI & Intelligent Automation

Tell us about the workflow, who uses it and what is difficult today. Please leave out passwords, confidential records and sensitive personal information.

Add company and other details (optional)

After you send this, choose a time for a free 30-minute discovery call. If no time suits, we get back to you as soon as possible to arrange one.

What is AI tender and BOQ analysis?

AI tender and BOQ analysis reads a tender pack, including the bill of quantities, specifications and conditions, into structured items. It compares BOQ lines with your past jobs and rates, summarises penalties, retention, payment terms and key dates, and flags missing scope. Each finding links to its source. The estimator reviews it, prices the work and decides whether to bid.

The problem

Tender packs are large, and the submission date does not move.

UAE tenders often arrive as a mix of Excel and PDF BOQs, specifications, drawing lists and contract conditions, sometimes with Arabic forms from a government client. The estimator has to read all of it, build a priced BOQ and raise clarifications before the submission date.

The risk sits in the details. A provisional sum, a liquidated damages rate, a retention clause or an item missing from the BOQ can decide whether the job is worth bidding. When several tenders overlap, the reading gets rushed and those details are found after award.

  • Re-keying BOQs

    BOQ lines are copied from PDFs into the estimating sheet by hand, with units and quantities checked line by line.

  • Conditions read under pressure

    Penalties, payment terms, retention and insurance requirements are spread through the pack and summarised from memory.

  • Scope gaps found late

    Items shown in the specification or drawings but missing from the BOQ surface during the project, not during pricing.

How it works

One tender pack, from download to a reviewed bid summary.

An illustrative workflow. The cost database, rate sources and summary format are agreed with your commercial team.

  1. Intake

    The tender pack is uploaded as received: Excel or PDF BOQs, specifications, conditions, addenda and drawing lists.

  2. BOQ extraction

    AI reads each BOQ line into item number, description, unit and quantity. It keeps the bill and section structure and links every line to its page.

  3. Comparison with past work

    Lines are matched to similar items from your past jobs or cost database, so the estimator sees reference rates where they exist. Where none exists, the line says so.

  4. Conditions summary

    Penalties, retention, payment terms, insurance, bonds and key dates are summarised, each with a link to its clause.

  5. Gap flags

    Items in the specification or drawing list but not in the BOQ, unclear units and conflicting quantities are flagged as possible clarifications.

  6. Estimator review

    The estimator checks the summary, prices the work, decides which clarifications to send and makes the bid or no-bid call.

Who does what

What the AI does · What a person checks.

What the AI does

  • Reads Excel and PDF BOQs into structured lines with units and quantities

  • Links every line and condition to its page in the tender pack

  • Matches BOQ items to similar items in your past jobs or rates

  • Summarises penalties, retention, payment terms and key dates

  • Flags missing scope, unclear units and conflicting quantities

  • Drafts clarification questions for the estimator to edit

What a person checks

  • Every rate, mark-up and price in the bid

  • Risk judgement on conditions, penalties and programme

  • Which clarifications go to the client and how they are worded

  • Any BOQ line the extraction marks as low confidence

  • The bid or no-bid decision

First pilot

A first pilot on tenders you have already priced.

We test on your own tender packs before anyone talks about results. Scope, timeline and price are confirmed in a written proposal.

  • Past tenders as the test set

    Use tenders your team has already priced, so the extraction and gap flags can be checked against what the estimator found.

  • One trade or project type

    Start with the work you bid most often, such as fit-out or MEP packages, where BOQ formats repeat.

  • One rate source

    One cost database or set of past BOQs as the reference, clean enough to compare against.

  • Agreed acceptance

    Agree line-level checks and what counts as a useful gap flag with your estimators, before the evaluation runs.

Data, hosting and Arabic

Where tender data goes, and how Arabic is handled.

  • Hosting

    The application and stored data can be hosted on Azure UAE North, on private servers managed by Vloud, or in your own cloud account. The AI model is used through a provider agreed per project, and its processing region may be outside the UAE.

  • Confidential rates

    Your rates and past bids are commercially sensitive. They stay in the agreed environment, and access is limited to your tendering team.

  • Arabic

    Arabic tender forms, conditions and bilingual BOQs are tested as their own sample, not assumed to behave like English ones.

  • Retention

    What is kept after a tender closes, and for how long, is agreed with your commercial manager. Documents from lost bids can be deleted on request.

Honest limits

Where it does not work yet.

  • Pricing decisions and risk judgement stay with the estimator. The workflow shows reference rates; it never sets a price.
  • We do not quote accuracy or time savings before testing on your tender packs. Results are measured on a held-out sample first.
  • Rate comparison is only as good as your cost data. If past BOQs use different descriptions or units, cleaning them is part of the scope.
  • Scanned BOQs with merged cells, handwritten amendments or poor print quality need a person to check them.
  • It reads documents, not drawings. Quantity take-off from drawings is not part of this workflow.

What drives cost and timeline. The number of tender types and BOQ formats, Excel versus scanned PDF, languages, the size and state of your cost database, the summary format your team needs and hosting requirements. We don't publish a price list; scope, timeline and price come in a written proposal after the discovery call.

Evidence

What we can show you.

Every example is labelled: delivered work, our own product, a demonstrator, or an illustration with synthetic data.

  • Our product

    ArkanERP BOQ control

    ArkanERP, our ERP for construction and project-driven companies, keeps the bill of quantities connected to procurement and project accounting. It is our own product, not an AI tender project, and shows where a structured BOQ goes after award.

  • Delivered work

    BOQ in ArkanCS at Al Mukab AlFudhi

    On Al Mukab AlFudhi Engineering Consultancy's G+5P+15 project, the BOQ is imported from Excel into ArkanCS, validated and linked to variations and payments. It is construction software work; no AI tender analysis is claimed.

  • Synthetic data

    The illustrative document workflow

    A step-by-step walkthrough of extraction, checks and review, shown with a synthetic invoice rather than a tender. The same pattern of values linked to their source and exceptions sent to a person applies to BOQ lines.

Before you start

Questions buyers ask.

Does it read Excel and PDF BOQs?

Yes. Excel BOQs are read directly, keeping bills, sections and item numbers. PDF BOQs are read page by page, and scanned or badly formatted pages are flagged for a person.

Can it handle Arabic tender documents?

Yes, Arabic, English and bilingual packs. Arabic forms and conditions are tested as their own sample during the pilot, since layouts and scan quality vary between clients.

Can it link to our cost database?

If your cost database or past BOQs can be exported or reached through an API, yes. We review the data first, because inconsistent descriptions and units affect the comparison more than the AI does.

Will it price the tender for us?

No. It shows reference rates from your own past work where they exist and flags lines with no match. Rates, mark-ups and the final price are set by your estimator.

Who designs the workflow?

Our founder is a mechanical engineer with a master's in construction management, with experience across MEP, tendering and construction software. The workflow follows how an estimator reads a pack, not how a generic document tool does.

Where is the data hosted?

The application and stored data can be hosted on Azure UAE North, on private servers managed by Vloud, or in your own cloud account. The AI model is used through a provider agreed per project, and its processing region may be outside the UAE.

Who owns the code?

You own the source code and IP of the custom workflow we build for you. Our own products stay ours, and any licence terms are agreed separately.

Next step

Test it on your own documents.

Tell us what the work looks like today. We reply with the questions we need answered before a pilot can be scoped.

Discuss this use case

A short note is enough. Phone and email let us reach you.

About: AI & Intelligent Automation

Tell us about the workflow, who uses it and what is difficult today. Please leave out passwords, confidential records and sensitive personal information.

Add company and other details (optional)

After you send this, choose a time for a free 30-minute discovery call. If no time suits, we get back to you as soon as possible to arrange one.