How We Test Keyboards

Transforming Our Data Into Comparable Reviews

Product Selection

Hundreds of keyboards are released each year—and while we may want to—we can't test them all. We primarily decide which keyboards to test based on how popular a keyboard is at the moment or how popular we think it may be in the future. We analyze search result trends and closely monitor which products people are talking about online. We try to purchase major releases as soon as they're available to the public, but we also rely on input from people like you and everyone else in the wider enthusiast community for guidance.

We love hearing from you in our forums, but the most direct way to voice your opinion is by voting for which keyboard we should review next. We restart our polls every 60 days and buy the product that wins the most votes each cycle.

Philosophy

We purchase our own products and put them under the same test bench so you can easily compare the results. No cherry-picked units sent by brands. No ads. Only real tests.

That's our company slogan, but it's not just a tagline—we want it to reflect the core principles behind how and why we test the way we do.

How we buy products

We buy each product the same way you do and don't accept review samples from brands. We do this to ensure we aren't getting hand-picked units that could skew our results and give you a false impression of what to expect if you buy the same product. The trade-off is that we don't get products as early as reviewers who receive brand-supplied samples.

Standardized tests

We test each keyboard on our standardized test bench, which is key to making our reviews helpful when you're trying to make a buying decision. By testing products in the same way, you can use our side-by-side tool to compare keyboards on an even playing field and determine which one best meets your needs.

We also have a powerful custom table tool that lets you explore our entire database of tested keyboards using adjustable filters. You can create your table with only a few basic filters for price, keyboard size, switch, or switch type, or you can go more in-depth with options for connectivity, software compatibility, keycap material, and even spacebar stabilizer size, just to name a few.

We're constantly working to improve our test benches, regularly adapting existing tests as well as introducing new ones to keep up with technological changes and react to community feedback. You can track all the changes we introduce to our test bench in our changelogs.

Each time we update our methodology, we assess our inventory to decide which products to update and retest, prioritizing as many notable and relevant models as possible. At the same time, we designate some models to remain on previous test benches. This helps manage our workload while ensuring a natural cycle where keyboards stay in our inventory only as long as they remain relevant and widely available. We'll cover our retest process and how we decide which models to keep in inventory in more depth later on.

Testing

Our keyboard reviews have four sections: Design, Typing Experience, Performance, and Software and Operating System. The following is a quick overview of each section. In several cases, we have dedicated pages with more precise details about our methodology, particularly for some of our most sensitive tests for evaluating latency, typing noise, and typing experience. If you're interested, see a list of all our keyboard articles.

Design

This is our largest test section, covering a wide range of fundamental tests focused on a keyboard's dimensions, connectivity, and hardware customization. We use various tools, including electronic scales and calipers, to ensure precise measurements.

This section also covers several subjective tests, including ergonomics, where we document home row height, key alignment, and typing angles. We also evaluate backlighting features and clarity using a specialized lightbox rig to ensure we test each keyboard under the same controlled lighting conditions.

Typing Experience

In this section, we provide details on a keyboard's switches and produce a keystroke actuation graph using a Mecmesin force testing machine.

The keystroke actuation graphs demonstrate distinct differences between switches, so you can see at a glance how lightweight or heavy a switch feels, whether it has a tactile bump, and where the actuation and reset points occur along the length of travel.

This section features a typing noise test with video footage so you can hear what the typing noise sounds like. The video also incorporates a noise level meter to measure a keyboard's average loudness.

This is also where we feature our overall typing quality test. We evaluate a host of factors that define the typing experience, such as switch smoothness, stabilizer rattle, key stability, sound profile, and bottoming-out experience. We provide a written account of how these individual elements combine in the overall typing experience. For keyboards designed with hardware customization in mind, we also endeavor to comment on which elements are immutable and which you can easily modify by using different switches, mounting styles, or acoustic treatments.

Performance

The name of this section probably gives it away, but this is where our performance tests live. We break down latency testing into single-key and multi-key latency scenarios and use a solenoid as a stand-in for a human finger to produce hundreds of latency samples for each review. Additional tests in this section cover chord split performance, which evaluates how effectively a keyboard reports many simultaneous inputs as well as data transmission details, which uses the keyboard's raw USB output to confirm N-key rollover support and its maximum effective update rate.

We film our latency testing with an associated in-game video. This video only serves as a way for you to validate our tests, but we don't use this video to calculate measurements. This is important because we're not testing end-to-end latency. We intercept input information directly from the keyboard using a Beagle 480 USB analyzer.

We ensure each keyboard is tested under optimal latency conditions. This includes minimizing switch debounce, actuation, and rapid trigger settings, enabling low-latency or competition modes, disabling RGB lighting effects, and using the highest available scan and polling rates. This step is essential for evaluating each keyboard at peak performance and ensuring fair comparisons between models.

Software and Operating System

This final test section is relatively simple. We assess keyboard compatibility across different computer and mobile operating systems and evaluate any configuration software options. We'll add details about how intuitive the software is to use and document any new or unusual features that could benefit from an explanation. In some cases, we'll even comment on how well the software is received within the wider community to give you advance warning of issues or complaints people are having.

Writing

Once testers complete the testing phase, we begin the first of two peer review processes between writers and testers to validate the results. Though this step is invisible in the final review, it's a critical quality check. At this stage, we cross-reference results with similar models, apply our expert knowledge, and consider community expectations and feedback. Any discrepancies or anomalies prompt further questions and often lead to retesting. Depending on the product's complexity and any unexpected findings, this process can take anywhere from 20 minutes to several days. Once writers and testers align on the results, we publish early access findings for our insiders and begin writing the review.

Our reviews cover a wide range of details. Each includes a brief introduction highlighting standout aspects and advertised features, along with targeted insights on a keyboard's performance for specific use cases. Later sections provide free-form comparisons, examining size and model variants within the same lineup and contrasting them to other keyboards in the wider market.

While our test results are designed to stand on their own for a knowledgeable audience, many include supporting text to provide additional context if you're unsure how to interpret them. This text often highlights comparisons to other products and, when necessary, explores unusual results or exceptions in more technical detail.

Once the writing is complete, a second round of peer review begins. A different writer reviews the text, posing questions, making corrections, and suggesting changes for clarity and readability. Lastly, the original tester verifies all writing for technical accuracy.

Once validation is complete, the review is sent to our editing team, who work behind the scenes to ensure proper formatting, consistency, and accuracy. Editors also enforce our internal style guidelines and uphold the high-quality standards our readers expect across the entire website.

Recommendations

Our keyboard recommendations are outside the scope of individual reviews but remain closely related. These curated lists help users decide which keyboard to buy, and we update them frequently. When preparing and maintaining these recommendations, we consider more than just test scores—we also factor in pricing, availability, and often more subjective elements like the buying experience and customer support.

At the end of the day, our recommendations are just that—recommendations. They aren't intended as definitive tier rankings for longtime hobbyists. Instead, our goal is to make them most useful for non-experts or those who haven't purchased a keyboard in a while and need an approachable starting point.

Retests and Updates

We keep the products we test for as long as they remain relevant and widely available—and often even longer. At regular intervals, our writing, testing, and test development teams collaborate to determine which products to retain and which can be safely resold. We typically hold on to products featured in our recommendation articles, those expected to remain popular despite their age, and noteworthy models we may need as references when updating our test bench.

Keeping a large number of keyboards in-house allows us to retest them whenever needed. As mentioned earlier, we retest many keyboards when updating our methodology, but that's not the only reason. The most common is major firmware updates, which often promise improved performance or new software features. We also frequently retest products in response to community requests about potential build quality concerns or specific performance questions.

This process mirrors our review testing but on a smaller scale. Testers conduct the retests and then collaborate with writers to validate the results. Writers update the text with context on the new findings, and after our editorial team reviews it, we publish an updated version of the review. For full transparency, we include a public message citing the changes made, the reasons behind them, and which tests were affected.

Videos and Further Reading

If you're curious about watching our review pipeline in action, check out this video below.

We also produce in-depth video reviews and recommendations using a similar pipeline of writers, editors, and videographers with multiple rounds of validating for accuracy at every turn. For keyboard, mouse, and monitor reviews in video form, see our dedicated RTINGS.com Computer YouTube channel.

For all other pages on specific tests, test bench version changelogs, or R&D articles, you can browse all our keyboard articles.

How to Contact Us

Constant improvement is key to our continued success, and we rely on feedback to help us. We'd encourage you to send us your questions, criticisms, or suggestions anytime. You can reach us in the comments section of this article, anywhere on our forums, on Discord, or by emailing feedback@rtings.com.