ConstructionWeekOnline Full-Text Article & Byline Dataset
ConstructionWeekOnline covers construction, infrastructure and the built environment across the Gulf, and this file carries the whole article — 14 columns with the full body text, the standfirst, the byline, the publication timestamp, the country tag, the permalink and the images.
The sampled article ran to 358 words of body text on Dubai utility costs, timestamped to the minute and tagged AE, with its own stable Article ID.
The file already exists, because this pipeline runs once a month whether you buy it or not. Most vendors start collecting after you order — which is why they quote a lead time. Here the most recent file is in your account within minutes of payment, with an API key issued at the same time.
What you get, in plain terms
Four things. A trade publication captured as text, not as a headline list.
Full body text, not a summary
Message carries the complete article — 358 words in the sampled piece, including the subheadings and the tag list — which is what makes topic modelling and entity extraction possible.
Headline and standfirst separately
Title is the headline and Media Title the standfirst or summary line, so you can index the claim and the framing as two different fields.
Timestamped to the minute
Created Time is a full ISO timestamp with a timezone, so publication order can be reconstructed exactly rather than by date alone.
However you want it
Direct download, REST API, Amazon S3, Google Cloud, Snowflake or SFTP. The API key comes with the dataset.
Fields included in this dataset
All 14 columns, in the exact order they appear in the file — taken straight from the sample, not from a brochure. The free sample ships with a data dictionary giving an example value for each one.
Sample rows from the real file
Real rows from the sample file, not an illustration. The supplied sample has 1 article with all 14 columns.
Coverage
Captured monthly across constructionweekonline.com. One row is one published article, identified by its Article ID and permalink.
| Metric | In the supplied sample | What it tells you |
|---|---|---|
| Articles | 1 — see the note below | Ask for a fuller sample |
| Body text | 358 words | Complete article, not a summary |
| Byline | “Staff Writer” | Named writers appear where credited |
| Timestamp | ISO, to the minute | 2026-08-09T07:04:00Z |
| Country tag | AE | Gulf-focused coverage |
Reach · Web Shares |
Empty | Not published — see the note |
Only tracking a topic, a project or a set of keywords? Send the list and we will price a narrower file.
Historical data
The last 30 days come with the dataset. Beyond that we hold ConstructionWeekOnline records from February 2025 onwards.
Ask about historical dataNeed more data points?
We can extend this dataset beyond the standard 14 columns — project, company and location entities extracted from the body text are the most requested additions, along with the article tags as their own field.
Request custom fields“Headlines are easy to get. Full body text with a stable article ID is what our monitoring actually runs on.”
What people use this dataset for
Contractors and developers
Track project announcements, tenders and regulation across Gulf construction coverage.
Communications teams
Monitor how your projects and executives are written about, with the full text to quote from.
Suppliers and manufacturers
Follow the schemes and specifications being reported before they reach tender.
Data and AI teams
Train classification and entity extraction on clean full-text trade journalism.
About ConstructionWeekOnline article data
ConstructionWeekOnline is one of the best-read construction and infrastructure titles in the Gulf, covering projects, contractors, regulation and the utilities that sit behind them. For anyone selling into UAE construction, it is a primary source rather than a secondary one.
Full text changes what you can do
The sampled article carried 358 words of body copy including its subheadings and tag list. With the complete text you can extract projects, companies and locations, cluster topics and search phrases — none of which a headline feed supports.
The supplied sample holds one article
This sample is a single row. It shows the schema and the depth of the text faithfully, but it says nothing about volume. Ask us for a fuller sample before you buy — we will run a month of coverage so you can see article counts, how often bylines are named and which topics dominate.
Two things to plan for
Reach and Web Shares came through empty, so the file carries no audience metrics. Journalist read “Staff Writer” on the sampled article; named bylines appear where the publication credits them, and we can report the share for your date range.
Is collecting this data legal?
We collect only what the publisher makes publicly visible: headline, standfirst, body text, byline, publication time and the permalink. Nothing behind a paywall or a login is touched, and the file holds no personal data beyond the writer's byline as published. Article text and photographs remain the publisher's copyright — the licence covers internal analysis, monitoring and model training, not republication. If you intend to redistribute article text, tell us and we will point you at the publisher's own syndication terms.
Frequently asked questions
Message holds the full body, 358 words in the sampled piece, with Title and Media Title carrying the headline and standfirst separately.
