How VEYDICT scores a product

The short answer

This article explains each part of the score, how the verdict is decided, how we judge sources, how confidence works and what we refuse to do. A worked example at the end shows the arithmetic on a real looking case. The full reference is on the methodology page, and this post adds the reasoning behind the choices.

The goal is simple. You should be able to see why a product received its verdict, and to disagree with it if your situation is different.

The six parts of the score

Each part answers a different question about the purchase: how well it fits you, how good the product is, what you get for the money, whether today's price is attractive, whether something else suits you better and whether there is a reason to wait. Fit for you carries the most weight and timing the least, because the same product can be perfect for one person and wrong for another, while the date matters only some of the time.

The exact weights are listed on the methodology page. They may vary by category as we learn what matters most in each one. For example, battery life counts for more in a laptop than in a desktop monitor.

Fit for you

Fit carries the largest weight because the same product can be perfect for one person and wrong for another. When you start a check, we ask what you will use the product for, your maximum budget, up to three priorities from a list that depends on the category and how long you plan to keep it.

Your priorities are matched against the product's measured strengths. A laptop that scores highly on battery, portability and build quality is a strong fit for someone who chose those three, and a weaker fit for someone who chose gaming and performance. We then adjust for budget. A product priced above your maximum loses fit points in proportion to how far over it is, and one priced far above your budget becomes a blocker.

Product quality

Quality is an overall view of the product, independent of you. It draws on manufacturer specifications, independent reviews with measured results, long term owner reports and recurring complaints. A product with a serious problem that appears across many sources scores lower, even if its headline specifications are strong.

We look for patterns, not single opinions. One angry review is noise, while the same issue reported by many owners over many months is information. That is also why products with few reliable sources are marked with limited confidence.

Value and market price

Value asks what you get for the price. It weighs the product's quality against what you would pay and against how far the current price sits above the lowest credible price we can verify. A great product at a high price has weaker value than the same product at a fair price.

Market price is the narrower question of whether today's price is attractive. We compare it with the lowest credible street price, and we derive a buy price from it, set slightly above that low so it is realistic to reach. If we cannot verify a price history, we say so, and we do not guess.

Alternatives

A purchase is a choice among options, so we look at a few. For each product we search for credible alternatives in the same category that are available in your market, then compare their fit for you, quality, value and price. We show up to three: one that saves money, one that is better overall and one that suits your priorities better.

If an alternative fits you much better than the product you chose, the alternatives part of the score falls, and the verdict can change. We never rank alternatives by who pays us. Affiliate commission never changes a score or the order of options, as our affiliate disclosure states.

Timing

Timing asks whether there is a reason to wait. A confirmed successor, a price well above its recent low or a sale that is likely soon all lower the timing score. When no reliable information about a replacement exists, we say that no confirmed replacement timeline was found, and we do not suggest a new model is coming.

Timing has the smallest weight, but it matters for the verdict, because a good product at a poor moment becomes a WAIT. See should you buy last year's model for how successors affect value.

How the verdict is decided

The score is an input, not the verdict. After computing it we apply a short list of rules in order, and the first rule that applies decides the outcome. Serious problems come first: a known defect, a price far above your budget, a deal breaker the product fails, poor fit, poor value or an alternative that fits you much better each lead to a SKIP. Next come the reasons to wait, such as a price well above our buy price or a confirmed successor. Only then does a combined score of 70 or more give a GO.

That ordering is why a 90 out of 100 product can still be a SKIP for you, and why a product with a modest score can be a GO at the right price. The verdict answers the question you actually asked, which is whether you should buy this one. The full list of rules is on the methodology page.

Confidence is separate

Score and confidence answer different questions. The score says how good the purchase looks. Confidence says how strong the evidence is. A high score with limited evidence is a different thing from a high score with many reliable sources that agree.

Confidence depends on how many reliable sources we found, how closely they agree, how fresh the data is, whether we matched the exact model and whether we found a current price. It is shown as High, Medium or Limited, with a short reason. When evidence is thin, we say so and show what we could and could not verify.

How we weigh sources

Sources are not equal. A manufacturer is a very reliable source for specifications, and a less reliable one for opinions about quality. Independent testers with measurements carry more weight than casual impressions. Retailers are strong for price and availability, and customer reviews are useful for patterns. Community discussions are helpful for recurring issues but weak on a single comment.

We look for the same point across several sources before we treat it as a finding. A single post that says a laptop overheats is not evidence. The same complaint across many owner reviews and independent tests is.

What we never do

• We never invent a price history. When we cannot find one, the report says Price history unavailable.
• We never announce a replacement model unless it is reliably confirmed.
• We never let a commission change a verdict, a score or the order of alternatives.
• We never give a GO or SKIP when the evidence is too thin. The report says what we could not verify instead.
• We never hide the sources. Each verdict lists them and how reliable we think they are.

Rules decide, AI explains

Scores and verdicts come from fixed rules applied to the evidence, which makes them consistent and explainable. Artificial intelligence helps us read and summarise sources and write the explanation in plain language, but it does not set the number and cannot change the verdict. If the writer produces anything we cannot support, we fall back to a template and show the facts.

A worked example

Consider a pair of noise cancelling headphones priced at 449 dollars. For a buyer who cares about noise cancelling, battery and comfort, the scores are fit 92, quality 92, value 80, market price 62, alternatives 83 and timing 70. The arithmetic is 92 times 30 percent, plus 92 times 25 percent, plus 80 times 20 percent, plus 62 times 10 percent, plus 83 times 10 percent, plus 70 times 5 percent, which is about 85.

A score of 85 would normally suggest a GO. But our buy price for these headphones is 394 dollars, and 449 is more than 12 percent above it. The price rule applies, and the verdict is WAIT, with a note that we would switch it to GO at about 394. The Catch is that the microphone is average in noisy places, and the alternatives section offers a model that does better for calls. The numbers here are illustrative, and your own verdict uses your budget and priorities.

Where to go next

Try a check on a product you are considering, and read the report next to this explanation. Browse the categories we cover to see what each criterion looks like in practice, and read why a great product can be a bad purchase for the thinking behind the fit score. If you ever disagree with a verdict, the sources and the reasons are there so you can judge for yourself.