Crypto Fundamental Analysis: What to Actually Look For

A coin passing our Shariah screen tells you one thing: you are allowed to own it.

It tells you nothing about whether you should.

Those are two different questions and most people asking the first one never get around to the second. They find a permissible coin, treat the green badge as an endorsement, and buy. Then the thing goes to zero and they conclude that halal investing does not work. Halal investing worked fine. They just never checked whether the project was any good.

Fundamental analysis is that second check. Here is what it measures, what each part actually means, and how to read the score sitting on every asset page we publish.

Halal and good are not the same axis

Start with the confusion, because it is the whole reason this article exists.

Look at what our own data says. Tether is a serious operation by any commercial measure and scores 87 out of 100 on fundamentals. Its Shariah verdict is Doubtful. Hyperliquid scores 84, which is a genuinely strong project, and its verdict is Haram, because the thing it is good at is perpetual futures. Meanwhile plenty of coins pass the Shariah screen without difficulty and score in the thirties, because nothing about them is prohibited and nothing about them is real either.

Permissible and well-built are independent. A coin can be both, neither, or one without the other.

So when you look at an asset page and see a green Halal chip next to a number, do not read them as one judgement with two parts. Read them as two judgements that happen to be about the same coin. The chip says the fiqh is clear. The number says how much of a business is underneath it.

The four things that are being measured

Our fundamental score is 100 points split across four dimensions, and the weights tell you what we think matters:

• Utility and adoption, 30 points. Is anyone using this, and have they been using it for long enough that it means something? Years of real usage with genuine network effects sits at the top. Plausible utility with thin adoption sits in the middle. A whitepaper and a Telegram group sits at the bottom.
• Economic sustainability, 30 points. Where does the money come from? A protocol earning real fees from real usage is durable. A protocol paying its users in freshly minted tokens is running a subsidy, and subsidies end. This is the dimension that catches the most projects, and it is the one retail investors check least.
• Technology and security, 20 points. Audited, battle-tested, years in production with no unresolved exploit. Or not.
• Trust and transparency, 20 points. Named team, clean history, disclosures you can actually read. An anonymous team is not automatically disqualifying, but it costs points, and a fraud history costs most of them.

Notice that utility and economics carry sixty of the hundred points between them. That is deliberate. Technology is usually fine and rarely differentiating. What separates a project that exists in five years from one that does not is almost always whether people use it and whether the money adds up.

You asked what to look for. That is what to look for.

The things people look at instead

Market cap is not a fundamental. It is a price multiplied by a supply, and both halves are set by the market's current mood. A large cap tells you other people have already bought, which is information about them, not about the project.

Liquidity is not a fundamental either, though it is worth knowing before you enter a position, because thin liquidity is how you discover that your exit costs fifteen percent. Check it. Just do not confuse it with quality.

Community size is the one I would push back on hardest. A loud community is frequently the strongest signal that a project has nothing else. Real infrastructure tends to have boring holders. If the most impressive thing about a coin is how much its owners post about it, that is a finding, and not the good kind.

Competitors matter, and they matter in a way people get backwards. The question is not whether a project has competitors. Everything worth doing has competitors. The question is what happens to this project if the largest competitor ships the same feature next quarter. If the answer is "nothing much, it dies," you have learned something the market cap did not tell you.

How to read the score we publish

Every asset on the screener carries a fundamental score, and as of this week that is 226 of them. The column sits next to the Shariah verdict, and the bands are fixed:

Bitcoin scores 96 and Ethereum scores 96. Solana scores 85, at the floor of the top band. XRP scores 78, which puts it one band below, and that gap is the point of publishing a number instead of a label. Both are real projects. They are not the same size of real.

At the other end, PEPE scores 20. Nothing about that number is a surprise to anyone who has looked at what a memecoin is, but it is worth saying plainly that the score is not being unkind. There is no utility to measure, no revenue to model, and no team to assess. Twenty is what is left when you score honestly and there is nothing there.

The band is calculated from the score, not written by hand. That matters more than it sounds like it should, and it is a change we made this week: previously the label was produced alongside the number rather than from it, and the two drifted. XRP at 78 was carrying the same wording as Bitcoin at 96. If you saw that and thought it looked wrong, you were right, and it is fixed.

Use the filter, then throw it away

On the screener there is a one-click filter for the top band. Turn it on with the Halal filter and you get the intersection: assets that pass the Shariah screen and score 85 or above. That is 27 coins out of 188.

That list is a starting point, not an answer. Twenty-seven is still far too many to hold, and the score cannot tell you whether something is expensive today. A 96 bought at the top of a cycle loses money exactly as efficiently as a 20 does.

What the score does is stop you wasting attention. Most people research in the wrong order: they hear a ticker, get interested, and then go looking for reasons. Filtering first inverts that. You start from the small set of things that are permissible and built, and then you do the slow work of reading the actual report on each one.

What the AI does and where you still have to think

The report behind each score is produced by our research model, and I would rather be specific about what that means than let you assume either too much or too little. It reads documentation, disclosures, audits, and public history at a volume no person is going to match, extracts the facts, and scores them against a fixed rubric so the same project earns the same score twice rather than a fresh guess each time. The full reasoning is published on every asset page, line by line, with the evidence attached.

What it cannot do is tell you what a thing is worth. It has no view on price, no view on timing, and no idea what you already hold.

It also has a failure mode worth naming, because it is the funniest one in our database and it will catch somebody eventually. There is an asset in the screener whose ticker is SOLANA. It is not Solana. It is a memecoin actually named BarbieCrashBandicootRFK88, it scores 0, and it is flagged Haram and fraudulent. It sits a few rows from the real thing.

Read the name. Not the ticker. Every time.

If you want to screen something we have not covered, the AI screener will run the same analysis on any coin you name and return the same report, including the fundamental section. We wrote separately about why we use a machine for this at all and where it breaks down, which is worth reading before you trust any of it too far.

The honest limit

I do not know how to score a project that is eighteen months old and genuinely novel, and neither does anyone else. The rubric rewards proven usage and durable revenue, which means it systematically underrates things that are too new to have either. Some of those will turn out to be the best assets of the decade and they are sitting in the Speculative band right now with a 62.

That is a real weakness and I am not going to pretend the number is smarter than it is.

What the score is good at is the opposite direction. It is very hard for something worthless to score highly, because scoring highly requires evidence that does not exist for worthless things. Use it to rule things out, which it does reliably, rather than to rule things in, which it does not.

Check the verdict to know if you may. Check the score to know what you are actually buying. Then read the report, because the number is a summary and the reasoning is the thing.