TenderHer Procurement Intelligence

TenderHer finds real public tenders across Nigerian procurement portals, reads the actual tender documents, pulls out the requirements with evidence, and tells you how ready your business is to bid — then serves it all as a live JSON API.

Instead of scrolling through portals and guessing which tenders fit, you get a shortlist with a MATCH / POTENTIAL_GAP / UNKNOWN verdict per requirement and a bid-readiness checklist. Every extracted requirement carries the exact text and source URL it came from, so you can verify it yourself.

The Standby API is live and read-only — the Actor stays ready so you can call it:

The Actor exposes a live, read-only API through Apify Standby mode. Direct requests to an Apify Standby endpoint may require Apify authentication, so do not assume the URL is unauthenticated; see the authentication note in the API section below.

https://bitbychelle--tenderher-procurement-intelligence.apify.actor/api/health

It is a working, read-only frontend for the data — useful for prototypes and wiring up your own dashboard.

Table of Contents

• What does TenderHer do?
• Why use TenderHer?
• How to use TenderHer
• Input
• Live API reference
• Output
• Data table
• Evidence-First Intelligence
• Requirement Assessment
• Bid-readiness Checklist
• How it works
• Technology
• Procurement Sources
• Pricing and cost
• Tips and advanced options
• FAQ, disclaimers, and support
• Source Code
• For developers
• Limitations

What does TenderHer do?

TenderHer discovers procurement opportunities from three real sources: NOCOPO (the federal portal), Ebonyi State e-Procurement, and Jigawa State Due Process. For each tender it downloads the tender document when one exists, extracts the text from PDFs, identifies the requirements (eligibility, documents, financial, technical, experience, submission), and produces an optional assessment against your business profile plus a bid-readiness checklist.

Because everything runs on the Apify platform, you get API access, scheduling, integrations, proxy rotation, and run monitoring out of the box — and Apify's own dataset export gives you the results as JSON, HTML, CSV, or Excel.

Why use TenderHer?

• Stop reading 200-page PDFs by hand. Requirements are extracted automatically and each one is linked back to its source document and text.
• Know your bid fit before you invest time. A business profile turns requirements into MATCH / POTENTIAL_GAP / UNKNOWN so weak spots surface early.
• One feed instead of three portals. NOCOPO, Ebonyi, and Jigawa in a single, consistent schema.
• Evidence-backed, not guessed. No requirement appears without the text and URL that supports it.
• Front-end friendly. The standby API means a dashboard can load tender data without scraping anything itself.

How to use TenderHer

Option A — browse the live API (fastest)

1. Open the live API in a browser.
2. Confirm it is alive with GET /api/health, which returns {"status":"ok"}.
3. Pull a page of tenders with GET /api/tenders?limit=25.
4. Filter with ?source=, ?status=, or page with ?offset=.
5. Open one tender with GET /api/tenders/:id to read its requirements, assessments, and checklist.

Option B — run a fresh crawl

1. Open the Actor in the Apify Console and go to Input.
2. Pick your sources and, optionally, fill in businessProfile to get assessments.
3. Click Start run.
4. Watch results stream into the Output tab, then export the dataset or call the API.

Input

All fields are optional except where noted; the defaults produce a working run.

Field	Default	Description
sources	["nocopo-bpp-ng"]	Which portals to crawl: nocopo-bpp-ng, ebonyi-egp, jigawa-dp.
sourceUrl	NOCOPO listing URL	Override the NOCOPO listing page.
maxTenders	15	Maximum tenders to collect per selected source.
maxRequestsPerCrawl	50	Global cap on requests for the run.
ebonyiOnlyOpenTenders	true	For Ebonyi, keep only open tenders.
businessProfile	none	Optional. industry, location, services, businessSize, experienceYears, certifications, annualTurnover. Omit it and every assessment is UNKNOWN — nothing is invented.

{
    "sources": ["jigawa-dp", "ebonyi-egp"],
    "maxTenders": 10,
    "businessProfile": {
        "industry": "Technology",
        "location": "Lagos, Nigeria",
        "services": ["software development", "cloud services"],
        "businessSize": "SME",
        "experienceYears": 4,
        "certifications": [],
        "annualTurnover": 20000000
    }
}

Live API reference

Base URL:

https://bitbychelle--tenderher-procurement-intelligence.apify.actor

Method	Path	Returns
GET	/api/health	{ "status": "ok" }
GET	/api/tenders	{ "data": [...], "total": number }
GET	/api/tenders/:id	{ "data": {...} }, or 404 with { "error": "Tender not found" }

Query parameters for /api/tenders: limit (default 25, max 100), offset, status, source.

:id accepts a bare sourceTenderId or a composite source:sourceTenderId when ids could overlap across sources. URL-encode the / in ids.

# Health check
curl "https://bitbychelle--tenderher-procurement-intelligence.apify.actor/api/health"

# First 5 tenders
curl "https://bitbychelle--tenderher-procurement-intelligence.apify.actor/api/tenders?limit=5"

# Only Jigawa tenders that are still open
curl "https://bitbychelle--tenderher-procurement-intelligence.apify.actor/api/tenders?source=jigawa-dp&status=open"

# One tender with requirements, assessments, and checklist
curl "https://bitbychelle--tenderher-procurement-intelligence.apify.actor/api/tenders/jigawa-dp%3AJG%2FMOC%2F2026%2F002"

CORS is enabled for all origins, so a browser frontend can call the API directly. Authentication: direct requests to an Apify Standby endpoint may require Apify authentication, so the API is not guaranteed to be publicly unauthenticated. When you call it from your own code, send your token in an Authorization: Bearer YOUR_TOKEN header and keep the token in a server-side environment variable — never put it in frontend code or in a URL query string. To try the endpoints without handling a token at all, open the Endpoints tab of the Actor in the Apify Console.

Output

Each item is one normalized tender. Simplified:

{
    "source": "jigawa-dp",
    "country": "Nigeria",
    "sourceTenderId": "JG/MOC/2026/002",
    "title": "CONSULTANCY SERVICES FOR THE CONSTRUCTION OF MODERN MARKET",
    "buyer": "Jigawa State Government",
    "status": "Open",
    "deadline": "2026-08-14T00:00:00.000Z",
    "tenderUrl": "https://dueprocess.jigawastate.gov.ng/tenders/2",
    "documentUrls": ["https://dueprocess.jigawastate.gov.ng/storage/tenders/Y9gEA8SjnBItDsQOLZqaInVzr10ck5ix9KLIe0UZ.pdf"],
    "requirementsStatus": "EXTRACTED",
    "requirements": [
        {
            "requirement": "The consultant must have at least 5 years of relevant experience",
            "type": "experience",
            "evidence": {
                "text": "The consultant must have at least 5 years of relevant experience",
                "sourceUrl": "https://dueprocess.jigawastate.gov.ng/storage/tenders/Y9gEA8SjnBItDsQOLZqaInVzr10ck5ix9KLIe0UZ.pdf"
            }
        }
    ],
    "businessProfileProvided": true,
    "assessments": [
        {
            "requirement": "The consultant must have at least 5 years of relevant experience",
            "assessment": "POTENTIAL_GAP",
            "reason": "the business profile lists 4 years of experience, which is below the 5-year requirement"
        }
    ],
    "readinessChecklist": [
        {
            "item": "The consultant must have at least 5 years of relevant experience",
            "status": "NEEDS_REVIEW"
        }
    ]
}

Each tender is stored as one record in an Apify Dataset. Apify's built-in dataset export lets you download that dataset in various formats such as JSON, HTML, CSV, or Excel, from the Console or the Apify API — TenderHer itself does not implement separate exporters.

Data table

Field	Description
source / sourceTenderId	Portal and the tender's own identifier.
country	Always Nigeria for the current sources.
title	Tender title as published.
buyer / procurementType / category	Buying entity, procurement method, and category.
referenceNumber	Published reference or lot number.
publicationDate / deadline / openingDate	Key dates.
location	Bid location or place of execution.
status	Tender status, e.g. Open.
estimatedValue / currency / bidSecurity	Financial terms.
tenderUrl / documentUrls / documentText	Source page, documents found, and extracted PDF text.
requirements / requirementsStatus	Extracted requirements with evidence; EXTRACTED or NONE.
businessProfileProvided	Whether a profile was supplied for this run.
assessments	MATCH, POTENTIAL_GAP, or UNKNOWN per requirement, with a reason.
readinessChecklist	Actionable checklist for preparing a bid.
evidence	Text and source URL backing each requirement.

How it works

1. Read the input — the run picks the selected sources, the crawl limits, and an optional businessProfile; unknown source ids are ignored with a warning, and the run stops early if none of the selected ids is supported.
2. Crawl the portals — a Cheerio crawler reads the listing pages over plain HTTP. No headless browser is used.
3. Normalize each tender — every listing becomes one record with the portal id, title, buyer, reference, dates, budget terms, status, document URLs, and the source page URL.
4. Read the documents — when a tender has a PDF, it is downloaded and its text is extracted inside the Actor.
5. Extract requirements — eligibility, documentary, financial, technical, experience, and submission requirements are extracted, each with the exact source text and document URL.
6. Assess and build a checklist — if a businessProfile was supplied, every requirement gets a MATCH, POTENTIAL_GAP, or UNKNOWN verdict with a reason, plus a bid-readiness checklist.
7. Publish the data — each record is pushed to the run's Apify Dataset. In Standby mode the same records are served over the read-only HTTP API below.

Technology

• Runtime: Node.js 24 on the apify/actor-node:24 Docker image (Apify Actor runtime), running JavaScript with ES modules. No web framework.
• Scraping: Crawlee @crawlee/cheerio (CheerioCrawler) — plain HTTP requests with concurrency limits and retries, no headless browser.
• Apify Actor SDK: the apify package for the Actor lifecycle, Datasets, and apify/log (credential-censored logging). Runs in Standby mode with an HTTP readiness probe.
• Storage: Apify Datasets hold one normalized tender record per item; the output schema links to the run dataset.
• API: Node's built-in node:http server — the read-only endpoints add zero runtime dependencies, enable CORS, and read the dataset per request.
• Documents: custom, in-house PDF text extraction.
• Requirement extraction: deterministic, rule-based logic written in this repository (src/requirements.js), operating on the extracted document text.
• Business-profile matching: deterministic rule-based logic (src/matching.js) producing MATCH / POTENTIAL_GAP / UNKNOWN — the same tender and profile always yield the same assessment.
• Schemas: JSON Schemas for the Actor definition, input, output, and dataset views.
• Quality: Vitest 5 (134 tests, including real captured HTML/PDF fixtures), ESLint with @apify/eslint-config, Prettier, and a GitHub Actions workflow that runs the tests on every push.
• Not used: no OCR, no machine learning, no LLM, and no external AI provider. TenderHer has exactly two runtime dependencies (apify and @crawlee/cheerio); all extraction, assessment, and matching logic is local, deterministic code, and no tender document is sent to a third-party model or service.

Source Code

TenderHer is open source and the Actor source code is available on GitHub:

GitHub: https://github.com/Ifeoluwayemisi/tenderherr

The repository contains the Actor implementation, procurement source adapters, document extraction, requirement extraction, business-profile matching, API, tests, and Apify configuration.

Pricing and cost

Pricing for this Actor is shown on its Pricing tab. What you spend depends on which side you use:

• Browsing the live API costs only the platform compute for standby runs that serve your requests.
• Running a crawl costs compute proportional to the work: maxRequestsPerCrawl and maxTenders are the main levers, and businessProfile adds no extra scraping cost.

Keep runs small while testing, then raise the limits for a scheduled production run. The Apify Console shows estimated cost before you start.

Tips and advanced options

• Raise maxTenders for breadth, not maxRequestsPerCrawl for speed. More requests per tender means deeper detail but higher cost.
• Always send a businessProfile if you care about assessments — without one every verdict is UNKNOWN by design.
• Combine sources in one run (["nocopo-bpp-ng", "jigawa-dp"]) for a single unified feed.
• Filter server-side in your frontend. Use source and status rather than downloading everything and filtering in the browser.
• Schedule runs in the Console to keep the dataset fresh automatically.

FAQ, disclaimers, and support

Does this replace reading the tender document? No, and it should not. TenderHer tells you what it found and where it found it so you can decide what matters. Always verify against the official document before bidding.

Where does the data come from? Publicly accessible procurement portals: NOCOPO, Ebonyi State e-Procurement, and Jigawa State Due Process. TenderHer does not bypass paywalls, logins, or access controls.

How accurate is the extraction? Requirements are extracted from the real tender document text with the source text preserved, but PDF layout quirks can affect extraction. Treat results as a fast first pass, not legal or procurement advice.

Is my business profile stored? It is used only for that run's assessments and is not written into the output records beyond the businessProfileProvided flag.

Something looks wrong or missing. Use the Issues tab on the Actor page to report it. Custom scrapers, extra sources, and tailored output schemas are also possible on request.

---

For developers