Debunking the “DA is a commodity” Myth

A common critique I’ve seen regarding Celestia’s revenue model is that DA should be regarded as a commodity in the long run, meaning fees will inevitably race toward the bottom or marginal costs.

For this claim to hold, DA as a resource would have to be identical across all DA services. But it isn’t - and here’s why:

Celestia DA ≠ Any Other DA

The important thing to note here is that Celestia does not only provide DA; it also provides consensus for rollups. Obviously, some DA providers offer stronger security guarantees than others. Moreover, some will have stronger network effects. This point alone makes DA services differentiated, meaning it is not a commodity by definition.

Intuitively, I don’t think this should be that hard to grasp, but let’s look at it from the rollup’s POV.

The Rollup POV

Rollups are the consumers of DA and consensus. They do not just choose DA based on what’s the cheapest option available; they also have to consider what is secure enough for them. Moreover, they want to use something that everyone else uses. That way, if they switch DA providers, they know they won’t get screwed over because what they are using has been battle-tested.

Seeing other protocols switch builds confidence, further differentiating that DA provider from others. This is a network effect in itself - one that cannot be forked and would be very difficult to replicate.

Ok, so DA is not a pure commodity - but how should we value it?

Given that DA is not a commodity, it will be reasonable for it to trade at some premium to its costs, but not so high that rollups are discouraged from switching to it. After all, it should provide these rollups with a net-positive outcome to switch. Currently, that net-positive is a huge reduction in fees. Let’s dive into the numbers:

Fees Are Set Low on Purpose!

Yes, you read that right. You may have seen rants on Twitter arguing that Celestia’s current revenue is too low and its valuation is too high, even after a large drawdown from ATHs. But what they’re completely missing is that the current revenue is intentionally low. Here’s why:

Beating the Competition and Winning Market Share

Understand that this has been a strategic choice for Celestia to win market share and beat its competitors. The goal is to attract users with free DA, let them try the product, and learn to love it. Once you have tons of users (rollups), monetizing them is not a difficult problem. However, winning over users and beating the competition is, which is why you cannot set high fees out of the gate and expect high inflows of users.

Here’s a quote from @musalbas confirming this on the TL:

Estimating Annual Revenue for Celestia

Recently, @musalbas opened up a discussion on the Celestia forum, debating whether or not to gradually increase current fees. The first proposed increase is by 4x (potentially more). The table provided in the tweet ranges from a 2x to 128x multiple in fee increases.

There are also discussions on anchoring the fees to USD and preventing them from being subject to volatility in TIA terms. Keep in mind that this is just a starting point, and it’s still up for discussion.

Now that we have some numbers to work with, let’s attempt to value DA fee revenue. Keep in mind that this is a very difficult task, and we’ll have to make some assumptions. But my guess is that if you’re already reading this article, you’re bullish on the demand for DA.

Assumptions

At this point, it’s becoming very clear that the total data posted by Celestia will increase significantly (currently 1.5TB), as more modular projects go to mainnet. So far, we’ve only seen Eclipse stress-testing their chain, but there is much more to come. My personal take? We’ve seen nothing yet.

To provide some context, Initia, Movement, and Abstract are just a few of many projects laying the foundation to build killer apps. But what will all these apps have in common? They will be entirely dependent on having secure, fast, and cheap DA.

With this in mind, I’ll make the case that Celestia will eventually post 50TB of data annually - and it will continue to grow from there.

TIA Revenue Napkin Math

As of today, the total TIA paid in fees is 313k. With TIA at $3.20, that’s approximately $1M USD. But again, remember that these fees are intentionally set low.

Let’s allow ourselves to dream a little here:

If we increase fees by 15x, it would still be significantly cheaper than ETH DA. Without raising costs too much, this would remain a great offer for rollups - approximately 66x cheaper than EIP-4844. We could adjust it higher or lower, but for this scenario, we’ll use 15x as our baseline.

Now we’re at $15M USD in revenue.

Next, we add DA demand into the equation - this is where the fun begins:

$15M × 50TB = $750M USD in annual revenue.

If we assume that the total data posted will increase year over year (YoY) by a decent percentage, we could easily surpass $1 billion USD in annual revenue.

Final thoughts

There are many variables that factor into this calculation in reality. Obviously, it’s not that simple either. Increase or decrease certain variables, and the revenue will adjust accordingly. But since people love numbers, I thought I’d provide one - it’s simple intuition that most can follow.

We can continue adjusting the numbers, arriving at different revenue figures. But my final take? $1 billion in annual revenue is imminent.

I don’t want to make this article too long, so I’ll stop here. If you found it interesting, let me know, and I’ll continue making more. Also, if you disagree with my views or have anything to add, please share your thoughts!

In the near future, I plan to post more analyses on modeling revenue and DA demand more accurately. This is just the beginning.

Mammoths! 🦣

Disclaimer:

1. This article is reprinted from [DE Analytics]. All copyrights belong to the original author [DE Analytics]. If there are objections to this reprint, please contact the Gate Learn team, and they will handle it promptly.
2. Liability Disclaimer: The views and opinions expressed in this article are solely those of the author and do not constitute any investment advice.
3. The Gate Learn team does translations of the article into other languages. Copying, distributing, or plagiarizing the translated articles is prohibited unless mentioned.