The Tech Press Covers 7% of Software. AI Eats Half the Coverage.

Software buyers treat press coverage as a quality signal. If a tool is in TechRadar and on Hacker News, it must be worth a look. We tracked 11,506 press mentions across 387 sources between February and August 2026 and mapped every one to a catalog of 9,960 tools. The data says coverage is a signal of exactly two things, hype and incumbency, and almost nothing else. It is one of the worst proxies for whether a tool fits your job.

The dataset

Toolradar runs a media-monitoring network that ingests articles from 387 tracked sources (tech press, trade publications, and newsletters) and matches each to tools in the catalog. This report covers the 11,506 mentions logged between February 10 and August 4, 2026, mapped to 701 distinct tools. Each mention is a named reference in a published article, not a passing keyword. This is first-party data collected by Toolradar and released under CC BY 4.0.

The 7% problem

Start with the denominator. There are 9,960 published tools in the catalog. Over six months, 701 of them got any press coverage at all. That is 7%. The other 9,259 tools, many of them profitable, widely used, and genuinely good, generated zero mentions in the sources we track.

This matters because buyers read absence as a verdict. It is not. Press coverage is driven by news hooks: a funding round, a launch, a controversy, a benchmark. A dependable invoicing tool that has shipped steadily for eight years has no news hook, so it never appears, while a three-week-old AI wrapper with a splashy demo gets written up ten times. If you filter your shortlist by "have I seen this in the press," you are filtering for novelty, not for tools that will still be maintained next year.

Ten tools, half the coverage

Coverage does not just skew toward the famous. It collapses onto them.

These ten tools account for 5,792 mentions, 50.3% of the entire corpus. ChatGPT by itself is 22.3%. The gap between first and second place (2,564 versus 612) is larger than the total coverage of every tool ranked below the top 20 combined. When a single product owns a fifth of all software press, "as seen in the press" stops being a useful filter.

AI absorbs the attention

Break coverage down by category and the concentration gets starker. A few AI categories, some with only a handful of tools, pull more mentions than large mature categories with a hundred products each.

Chatbots generated 2,598 mentions across just 8 tools, roughly 325 mentions per tool. Developer Tools spread 1,970 mentions across 116 tools, about 17 each. The press is not covering software, it is covering the AI story, and attaching a tool name to it. That is great if you sell an AI chatbot and terrible if you are trying to use coverage to evaluate a CRM.

Who does the covering

The sources doing the work are the general tech-review outlets and the security trade press, not a broad, balanced press corps.

No single outlet dominates, which is healthy, but the mix is tilted toward review sites and security news. That shapes which tools surface. A category that outlets like to review (VPNs, AI tools, developer utilities) gets a steady drip of coverage. A category they rarely write about (procurement, medical billing, field service) can be enormous and still invisible.

What this means when you evaluate software

• Do not read zero coverage as a red flag. 93% of tools have none, including many category leaders. Absence from the press is the default, not a warning.
• Do not read heavy coverage as a green flag. It correlates with news value and category, not with whether the tool fits your team. A tool can be the most-covered in its space and the wrong choice for you.
• Use signals that scale with quality, not hype. Review volume weighted by rating, feature fit, and pricing transparency separate tools far better than press mentions do. Toolradar's category rankings deliberately weight external review strength and recency over raw coverage for this reason. The method is on the how we rate page.

Press coverage answers one question well: what is the industry talking about right now. It answers the question buyers actually have, what should I use, almost not at all.

Methodology

Source: 11,506 media mentions ingested by Toolradar's media-monitoring network from 387 tracked sources between February 10 and August 4, 2026, each matched to a named tool in the 9,960-tool published catalog (701 tools received at least one mention). A mention is a named reference in a published article. Category figures count distinct mentions and distinct covered tools. Released under CC BY 4.0. Reuse it with a link back to this page.

Frequently asked questions

Does press coverage mean a tool is good?

No. In this data, coverage tracks news value and category (mostly AI) far more than quality. 93% of tools, including many category leaders, get no coverage at all, and the most-covered tool is not the highest-rated in its space. Treat coverage as a measure of attention, not merit.

Why does ChatGPT get so much coverage?

ChatGPT is both the reference point for the entire AI story and a product that ships news constantly (model updates, features, controversies, benchmarks). It accounts for 22.3% of every software mention we tracked, which reflects its role as the anchor of tech coverage, not a claim that it is 22% of useful software.

How should I evaluate a tool with little or no press coverage?

Ignore the coverage gap and evaluate the tool directly: read the external review corpus (volume and recency matter more than the star), check pricing transparency, and confirm it fits your specific workflow. Most of the best tools in unglamorous categories are quiet in the press.

Where does this data come from?

From Toolradar's own media-monitoring network, which tracks 387 sources and matches articles to catalog tools. It is first-party data, released under CC BY 4.0 so anyone can cite or reuse it with attribution.