How our star ratings work

What a rating means, and what it does not

Most people glance at stars first and read second. That is natural, and it is also the reason we want to be plain about what our stars are. A rating on this site is a short summary of a longer argument. It tells you how a product or a product family came out when we weighed the things that matter to most buyers in that category. The argument is on the page, and the stars are only the headline.

A rating is not a lab result. We do not claim that our team ran a controlled test unless a page says so and shows what was done. Our comparisons are built on published specifications, manufacturer documentation, the update and warranty policies that companies state in writing, independent measurements and reviews where they exist, and the common experience that owners describe over time. When a number on a page comes from someone else's test, we say it comes from a published source.

A rating is also not a universal score. A phone with five stars for a student on a tight budget might get fewer for a professional photographer, because the two readers care about different things. This is why every comparison on the site explains who each option suits. If you only take one habit from this page, let it be this one: read the sentence that explains the score before you look at the score.

Finally, a rating is a snapshot. Products get new versions, software updates change behaviour, prices move and stores change what they sell. A rating written last year may not describe the product on the shelf today. We say more about this further down.

The scale

We use a five-star scale with half-star steps. The words below describe how we think about each level. They are the same for every category, which keeps a four-star washing machine and a four-star laptop comparable in spirit, even though the products have nothing in common.

Two things about this scale. First, a high score does not mean "best in the world". It means the product does its own job very well for what it costs. A modest, well-priced product can beat an expensive one on this scale, because we always ask what you get for your money. Second, because we compare products that people actually consider buying, most of what you see on the site lands in the middle and upper part of the scale. A rating of three stars is not an insult. It usually means a product with a particular catch that you need to understand.

The five things every rating looks at

Every rating is built from the same five questions. Some products are strong on one and weak on another, and that is exactly what the written comparison is there to explain.

1. Value for the price

This asks what you get compared with what you pay and with what rival products cost. A cheap product that does the basics well can score high here. An expensive one has to give you something real for the extra money, such as longer support, better materials or features you will actually use. Price here means the price you are likely to pay, including things the box does not include, like a charger, a case or a subscription, and not only the headline figure. Because prices change by country and by month, we treat this as the most perishable of the five.

2. Everyday usability

This is about how the product feels to live with. Is it easy to set up? Is the interface clear? Is it comfortable, quiet, or awkward in ordinary use? A product with excellent specifications can still be tiring to use, and a plain product can be a pleasure. We look for features that help with the main job and for annoyances that appear every day. Usability is partly personal, so we try to describe what a thing is like and who will enjoy it, and avoid pretending that one answer fits everybody.

3. Features that matter

Here we count what the product can do, but only the parts that affect real decisions. A long list of features does not raise a score if you will never touch most of them. We separate headline numbers from useful ones. For example, a bigger number on a spec sheet is only a plus when it translates into something you notice. A feature that works well counts for more than three that work badly. When a feature is limited by country, language or model version, we say so, because a feature you cannot use is not a feature for you.

4. Build quality and reliability record

We look at materials, repairability, durability claims and, most of all, what is known about long-term reliability. That includes the pattern of problems reported by owners and independent surveys, known recalls and common faults in a model or a generation. Where little is known, for example with a very new product, we say that the record is short and treat it with care, and we do not guess. Reliability has a long memory, and it can take a year or more for the real picture to show up.

5. Support and availability

This covers the warranty, the software or update promise, the availability of spare parts and accessories, and how easy it is to get the product repaired where you live. It also covers whether the product is actually sold and supported in your market. A very good product with no local service can be a poor buy, and a plain one with a strong service network can be a safe one. This criterion matters more outside the biggest markets, and we try to say when it does.

Why the weights change by category

The five questions stay the same, but their importance does not. When you buy a car, long-term reliability and the service network matter far more than a fancy screen. When you buy headphones, comfort and sound fit matter more than the length of the warranty. A rating that treated all five questions as equal for every product would be neat and also wrong.

So each category has its own emphasis. The table below shows what usually counts most. It is a guide to our thinking, not a formula that we apply mechanically, because each comparison also has its own special points.

Where the evidence comes from

We try to be clear about the kind of evidence behind each claim, because they are not equally strong.

• Manufacturer specifications. The starting point. They tell us what a product is designed to do. They are claims made by the seller, so we treat them as useful but not final.
• Written policies. Warranty terms, update promises, return rules and safety ratings from independent test bodies. These are specific and checkable, so they carry a lot of weight.
• Independent measurements and reviews. Where a trusted outlet has measured something such as brightness, battery life or noise using a stated method, we use it and say it is a published result.
• Owner experience. Patterns reported by many owners over time, such as a common fault or a feature people love. A single story is not evidence, but a repeated pattern is.
• Our own analysis. How the pieces fit together for a given kind of buyer. This is the part that turns facts into advice, and it is judgment, which is why we explain it.

Where the evidence is thin or conflicting, we say so. We would rather tell you that something is unclear than cover the gap with a confident-sounding sentence. For more on how we gather and check material, read our methodology page, the test process and the notes on equipment and evidence standards.

A worked example

Here is how the same two products can end up with different scores for different readers. The products and numbers below are invented to show the method. They do not describe real phones.

Imagine Phone A, a modest model at a low price, and Phone B, a premium model that costs much more. We score each on the five questions from 1 to 5:

Now give each question a weight from 1 to 5 that shows how much it matters to a particular reader. A student on a budget might weigh them 5, 3, 2, 4 and 3. Multiply each score by its weight, add the results and divide by the total of the weights. For Phone A that is 68 divided by 17, which is 4.0. For Phone B it is 67 divided by 17, which is about 3.9. The cheap phone edges ahead.

A keen photographer might weigh the same questions 2, 3, 5, 3 and 2. Phone A comes to 55 divided by 15, about 3.7. Phone B comes to 67 divided by 15, about 4.5. The premium phone wins clearly. Nothing about the phones changed. Only the reader's priorities did.

This is why our star rating on a page reflects the typical reader of that comparison, and why the explanation underneath matters. If your priorities are different from the typical ones, the tool below lets you do the same sum yourself.

Make your own score

Enter two products, say how much each of the five questions matters to you, and give each product a score from 1 to 5 on each question. You can take the scores from our comparison text or use your own judgment. The tool does the arithmetic. Nothing you type is sent anywhere.

A tip: set "matters to me" to 0 for anything you do not care about. A tied result usually means you should look at the written comparison for the deciding detail.

How to read small differences

People like to rank things in a neat order, and a half-star difference looks like a clear gap. It rarely is. Ratings are made from judgments, and judgments have a margin. If two products are within half a star of each other, treat them as roughly level and let your own priorities, the price you can get today and the service near you decide.

A full star or more is a more meaningful gap, but even then you should check the reason. A product that is a full star ahead overall might still be worse for the one thing you care about most. The most useful question is not "which has more stars" but "what is each product bad at, and can I live with that".

We also avoid false precision. You will not see scores like 8.37 out of 10 on this site. A number with decimals suggests a measurement that we did not make, and it would give a feeling of accuracy that the evidence does not support.

When ratings change

A rating is revisited when something material changes. That can be a new version of the product, a major software update that changes how it behaves, a big change in price, a new warranty or update policy, a recall, or a better understanding of long-term reliability after owners have used it for a while. It can also change when readers point out a factual mistake. The way to report one is on our contact page, and our corrections policy explains what happens next.

We do not change a rating just because a retailer runs a sale or because a company asks us to. A price drop can change the value question, and when it does, we say that the change is about price. Check the date at the top of any page. If it is old, treat the details as a starting point and confirm the exact model and current offer before you buy.

Money and independence

Ranking Hours is a website that has to be paid for, and some pages contain affiliate links. If you buy through one, the site may earn a commission at no extra cost to you. Our affiliate disclosure explains this in detail. The important point for ratings is that these links do not decide which product is recommended or how many stars it gets. Editorial decisions, including which products are compared and how they are scored, are made before any commercial relationship is considered, and a page with no links is not more or less trustworthy than one with links.

If you ever think a rating looks shaped by commercial interest, please tell us. Our editorial independence statement and code of ethics set out the standard we hold ourselves to, and they are only worth something if readers feel free to challenge us when we fall short.

Limits you should know about

• Judgment is involved. Two careful reviewers could give the same product a half-star difference. We explain our reasoning so you can disagree with it intelligently.
• Versions and regions differ. The same product name can hide different parts, features and warranty terms in different countries. Always check the exact model sold near you.
• Long-term data takes time. New products have a short track record. We say so when it applies, and we do not pretend to know how a product will age.
• Personal taste is real. Comfort, style, sound and screen preferences are partly personal. A rating cannot capture them for you.
• We cannot see your situation. Your budget, your electricity price, your other devices and your local service network all affect what is best for you.
• We do not claim lab testing. Where a number comes from a published test, we identify it as such. Where we describe a general behaviour, we do not present it as a measurement of our own.

Common mistakes when reading ratings

Treating the number as the answer. The number is a summary. If it disagrees with the text, trust the text and tell us.

Comparing across categories. A four-star laptop and a four-star blender share a scale, not a standard. The stars only compare things that do the same job.

Ignoring the date. A rating from two years ago may be about a product that has since been replaced or changed.

Chasing the highest score at any price. The top-rated product is often not the right one for you if it costs more than you need to spend. Look at the product that fits your priorities and your budget.

Reading only the good news. Every product has a catch. The "who should skip it" part of a comparison is often the most useful paragraph on the page.

Trusting any single source. Including us. Use our ratings next to owner reports, the seller's written warranty and, if you can, a hands-on look in a shop.

Questions people ask

Questions about this page or a specific rating? Write to support@ranking-hours.uz or use the contact page.