Readers get stalled. Publishers say it protects the newsroom.
Across major UK sites, human verification screens now appear more often, even for loyal readers. Newsrooms say automated scraping has surged, driving costs up and squeezing revenue. That pressure has pushed companies to tighten rules and scan for behaviour that looks robotic.
Why you are seeing more checks
Publishers face waves of automated tools that harvest headlines, extracts and live scores in seconds. Those tools feed aggregators, AI training sets and low‑quality websites that undercut the original reporting. The bill lands with the publisher through higher bandwidth, degraded performance and distorted audience data.
That is why many groups now draw a hard line. The News Broadcasting and Onic group of companies, which includes talkSPORT Limited, states that no access, collection, text or data mining of any content is allowed by automated means, whether direct or via an intermediary, under its terms and conditions. It also provides a route to ask for commercial licences via [email protected].
Automated access, scraping, text or data mining are banned by contract across several news brands. Commercial users must seek permission first.
These measures come alongside bot‑mitigation tools that score each visit. They examine patterns such as speed, sequence, device signals and network origin. If a visit looks risky, a challenge appears.
The signals that can make you look like a bot
Many real readers trigger checks by accident. The systems do not “know” your intent; they read the signals you emit.
- Using a VPN, corporate proxy or shared campus network that masks or rotates your IP address.
- Aggressive content blockers that strip scripts, fonts or captcha widgets needed to validate a visit.
- Rapid‑fire clicks, lightning scrolls or instant page switching that mimic automation.
- JavaScript disabled, third‑party cookies off or privacy extensions that hide a browser fingerprint entirely.
- Opening dozens of tabs at once during breaking news, then refreshing all of them together.
- Outdated browsers or headless settings commonly used by scraping tools.
- Preloading by some apps and browsers that fetch pages in the background before you tap.
Small changes help: slow down scrolls, allow the captcha to load, keep your browser up to date, and avoid mass refreshes.
The rules behind the wall
Terms and conditions shape how content may be accessed. In the UK, contracts can restrict scraping even of publicly reachable pages. Database rights and copyright also apply, especially where selection and arrangement involve skill and labour. Text and data mining exceptions exist in limited contexts, but publishers can opt out, and many have.
Commercial use and permission
Where a company wants to ingest articles for products or analytics, licences are the norm. Groups set conditions: what can be taken, how often, and where it appears. The News Broadcasting and Onic group points readers to [email protected] for such requests. Without a licence, automated collection risks blocks, takedown notices, and potential legal action. Human readers still get access, but they may face more checks when traffic spikes or tools detect risky patterns.
The numbers behind the blocks
Bot vendors and publishers rarely disclose exact scores, yet industry benchmarks provide a sense of scale. During peak events, a notable share of traffic gets flagged as “suspicious”, and a chunk of that turns out to be people using privacy tools or congested networks.
| Metric | Typical range |
|---|---|
| Traffic screened as high risk on news sites during spikes | 25%–45% |
| Legitimate users within those high‑risk buckets | 10%–20% |
| Average time to pass a verification | 2–7 seconds |
| Initial challenge failure rate for humans | 1 in 6 to 1 in 8 |
| Reduction in bot traffic after strict enforcement | 30%–60% |
These figures vary by publisher, protection vendor and news cycle. The headline point remains: the systems aim to deter non‑human collectors, but they will catch a slice of genuine readers. That is why most challenges resolve quickly once you interact like a person—pause, read, click deliberately, and complete the prompt.
How to pass checks without trading away privacy
There is a balance to strike. You can keep privacy tools while signalling you are a person.
- Allow the verification widget to run on trusted news domains, while keeping other protections on.
- Use a stable connection rather than rotating VPN endpoints for everyday reading.
- Update your browser and enable JavaScript for the site; both improve compatibility with human checks.
- Avoid opening twenty tabs at once. Queue stories to read, then open a few at a time.
- If you use reader or accessibility modes, wait for the page to finish loading before activating them.
- When challenged, complete it once. Repeated refreshes often raise the risk score further.
Think “steady and visible”: consistent network, modern browser, and clear interaction signals often get you through first time.
What publishers say they will do
Newsrooms know friction costs them readers. Many now promise shorter checks, clearer wording, and accessible alternatives for those who cannot solve standard captchas. Several groups have central contact points for commercial use, research access, or whitelisting under a signed agreement, with defined limits and audit trails.
Contact paths and commitments
For businesses that need structured access, a written licence remains the clean way forward. The News Broadcasting and Onic group invites enquiries at [email protected]. Expect to outline your purpose, data fields, refresh rates, retention, and security. In return, publishers can offer stable feeds or staged exports that meet both compliance and technical needs.
For researchers and startups
Unauthorised scraping may feel faster, yet it exposes teams to blocks and unpredictable gaps. A short protocol helps:
- State your scope: titles only, metadata, or full text. Limit to what you genuinely need.
- Set rate limits that mirror polite human use. Spikes invite throttling.
- Respect robots controls and any formal opt‑outs for text and data mining.
- Propose time windows for collection to avoid pressure during live events.
- Keep logs to demonstrate compliance if asked.
Extra context that matters to readers
Human verification is not just about bots. It protects ad budgets, shields live blogs from scraping during sensitive incidents, and reduces server strain that can crash sites at key moments. The drawback is friction: a lost second here, a failed challenge there. If you read on a phone over mobile data, those seconds can stretch.
Risk models will evolve. Expect more silent checks that rely on device signals and fewer picture puzzles. Privacy‑preserving attestation, passkeys and local risk scoring promise quicker decisions with less data sharing. Until then, your best bet is simple: use a steady setup, interact like a person, and when in doubt, complete the prompt once and carry on. For companies that need more, ask for a licence first and avoid the block altogether.
41% risk of bans… source?