Buy The Dip? A Look At “Real Revenue” DeFi Coins

23 hours ago 4

TL;DR

We look at real-revenue DeFi names – Ethena (ENA), Pendle (PENDLE), and Hyperliquid (HYPE) – and ask: have fundamentals held up while tokens dumped, or are the revenues themselves now under pressure?
The answer is mixed:
- ENA throws off huge fees, but almost all of it is recycled into subsidies to defend TVL, so actual “earnings” to the protocol are tiny.
- PENDLE’s TVL and revenue have grown into the sell-off – a cleaner “price down, business up” divergence – but the business is still deeply tied to fat DeFi yields.
- HYPE is a monster cash engine, with >$1.2bn in annualised revenue, with effectively all of it routed to token buybacks – but it’s already priced like a winner and is now cutting fees to keep growth going.
The big picture: the market is offering better entry levels for quality names – but the fact that even the winners are cutting take-rates (HYPE) or over-subsidising TVL (ENA) suggests that it’s still too early to declare this as a clean “buy any real-revenue dip” regime.

The “Real Revenue” Framework: What to Actually Measure

When you screen for “real revenue coins”, it’s easy to oversimplify and look for:

“Fees up + token down = value buy.”

On-chain data lets you go deeper. For each protocol, we ask four critical questions:

Fees: Are users still paying, or has activity rolled over?
Revenue: What share of those fees actually accrue to the protocol?
Earnings vs. Incentives: How much is left after token incentives and subsidies?
Valuation: What multiple of the revenue/earnings are we paying at current prices?

DefiLlama conveniently splits most of this into Fees / Revenue / Holders Revenue / Incentives for each protocol.

With that in mind, we will evaluate Ethena (ENA), Pendle (PENDLE), and Hyperliquid (HYPE) – not to find “the healthiest” option, but to show where price–fundamental divergences are real, and where “revenue” is being massaged by fee cuts or incentives.

Ethena (ENA): Big Fees, Thin Earnings, Heavy Subsidies

Ethena is trading at ~$0.28–0.29 with a $2.1bn market cap, generating a massive ~$365m in annualised fees from its $7.3bn Total Value Locked (TVL). However, since most fees are recycled into incentives to maintain high yields, the protocol's actual annualised revenue is only ~$0.6m, leaving almost no net earnings for holders. Buying this dip is not a value play based on current Profit/Loss, but a structural bet that Ethena can eventually normalise subsidies without causing its user base to collapse.

Fee and Revenue Profile

Ethena’s combined USDe contracts on Ethereum now hold roughly $7.3B TVL.

On DefiLlama’s fee dashboard, Ethena looks like a machine:

Annualised fees: ≈ $365m
Cumulative fees: ≈ $616m

But the key line here is to look at “revenue”:

Annualised revenue: only ~$0.6m
Revenue 30d: ~$49k

As for incentives? That’s where the gap comes from: most of the fee flow is effectively recycled into user yield and incentives, leaving minimal net earnings to ENA holders relative to the fee headline.

Pendle (PENDLE): Growth Through the Sell-Off

PENDLE is trading at ~$2.7, down ~64% from its ATH of $7.50. Its circulating market cap is ~$450–460m with a Fully Diluted Value (FDV) of ~$770m.

Fee and Revenue profile

Pendle’s core business is tokenising yield and letting users trade PT/YT pairs. On DefiLlama today:

Annualised fees: ≈ $45.7m
Annualised revenue: ≈ $44.9m
Annualised holders revenue (vePENDLE): ≈ $35.9m
Annualised incentives: ≈ $10.8m

So:

Almost all fees drop through as protocol revenue (take-rate is strong).
A big share of that revenue is actually being paid out to locked tokenholders (vePENDLE).
Incentives exist, but are relatively modest vs. revenue – the earnings line is still clearly positive.

On the TVL side, Pendle has scaled into the down-cycle:

Total TVL recently passed $10bn, with recent data showing figures around $10.3bn and annualised fee revenue around $78m.

The catch for Pendle is DeFi yield’s cyclicality:

Pendle monetises on-chain yields. If LSD/LRT yields compress and stablecoin carry goes flat, the demand to lock in yields and trade them will shrink.

A lot of the current revenue is a function of elevated DeFi carry and speculative positioning on funding rates; that can change quickly if macro goes risk-off or if on-chain yields converge to TradFi.

Hyperliquid (HYPE): A Rate-Cutting Revenue Machine

Hyperliquid trades at ~$35–36 with a $9-10bn market cap and a massive engine generating ~$1.21bn in annualised revenue with zero incentive emissions. However, the investment thesis is shifting from "pure cashflow" to "aggressive growth," as the team cuts taker fees by up to 90% on new markets to capture long-tail dominance. Consequently, HYPE is priced as a winner (approx. 8–10x P/S) where future returns depend on these fee cuts successfully driving massive volume expansion.

Fee and Revenue profile

Hyperliquid is now the biggest perps venue by on-chain metrics:

Annualised fees: ≈ $1.34bn
Annualised revenue: ≈ $1.21bn
Annualised holders revenue: ≈ $1.20bn
Annualised incentives: $0 (unconfirmed airdrop)

We believe:

The revenue is real,
There are no clear incentive emissions eating the PnL, or being suspected by users whose main focus is to farm the airdrop other than to use the product
Almost all of it is being earmarked for HYPE buyback and burn via the Assistance Fund.

Against a roughly $9–10bn market cap, that’s a P/S near ~8–10x on current DefiLlama numbers – not absurd for a fast-growing exchange, but not bombed-out either.

New Growth Venues

The key nuance for this cycle: Hyperliquid is no longer simply “let the revenue rip and buy back”. It’s now actively:

Opening permissionless markets via HIP-3, where market deployers share in fee revenue; and
For new HIP-3 markets, slashing taker fees by up to ~90% to bootstrap volumes in long-tail perps (equities, niche assets, etc.). Public posts on HIP-3 and exchange docs outline this “growth mode” fee schedule.

Putting It Together: What’s Mispriced?

After a look at the facts, we arrived at a few working conclusions:

1. “Real revenue” alone is not enough

ENA proves that fees ≠ earnings. The protocol shows hundreds of millions of annualised fees and still has almost nothing left for token holders after paying for TVL and user yield.
HYPE shows that revenue is endogenous: When the team cuts fees to grow market share, revenue and its multiples move with that decision, not just with user demand.

Any “buy-the-dip” screen that stops at “fees up” will systematically mis-classify these.

2. PENDLE is the clearest “business up, token down” pattern in this set

TVL and revenue are both meaningfully higher than at prior highs.
Incentives are modest vs. revenue; holders’ share is material.
Yet the token is down more than half from the peak.

That doesn’t make it risk-free – it’s still structurally cyclical – but it’s the closest to the classic value shape in this cohort.

3. We’re seeing stress even in the winners

The most important takeaway for timing:

HYPE cutting fees to grow new markets
ENA maintaining very high subsidy levels to keep USDe attractive

These are both signals that even dominant protocols are feeling the current environment – either on the volume side (perps) or on the “yield competition” side (synthetic dollars).

If the leaders are tinkering with take-rates and incentives, we’re probably not in a regime where you can blindly buy anything with fees and wait for mean-reversion.

The Bottom Line

Yes, there are divergences: PENDLE especially looks like a business that quietly got stronger while its token re-rated lower. HYPE and ENA are both big, real businesses – but their own decisions (fee cuts, subsidies) reveal that this environment is still fragile.

For now, this looks less like a “blind buy-the-dip in anything with fees” moment, and more like a stock-picker’s market inside DeFi, where winning moves include: