# DeFi for RIAs: an advisor’s guide to the asset, protocol, and chain > DeFi is not one asset class and “the protocol” is not the whole risk. An advisor has to evaluate the asset held, the protocol holding it, and the chain settling it. - URL: https://riadefi.com/defi-for-rias/ - Section: Foundations - Author: Jon Ragsdale - Reviewed: 2026-08-24 Reader objective: Understand DeFi from a fiduciary and due-diligence perspective. ## What DeFi actually is Decentralized finance is lending, borrowing, trading, and asset management performed by software deployed to a blockchain. A lending protocol can match suppliers of capital with overcollateralized borrowers without a bank operating the ledger. The rules are visible in code and transactions settle on-chain. That description does not make the arrangement safe, decentralized in every respect, or appropriate for a client. It only identifies the mechanism. A USDC lending position, for example, combines an issuer-controlled dollar token, an upgradeable lending protocol, an oracle, and a settlement chain. Each dependency can fail differently. The mechanics of the flagship case, overcollateralized lending, fit in a paragraph. A supplier deposits USDC into a pooled market contract and receives a receipt token that grows with interest. A borrower posts collateral worth more than the loan and pays a floating rate set by utilization: the more of the pool that is lent out, the higher the rate climbs, which attracts new suppliers and pushes borrowers to repay. If the collateral value falls toward the loan value, any third party may repay the debt and seize the collateral at a discount. That liquidation auction, not a credit officer, is what protects suppliers. Every step is a contract call recorded on the chain, which is why the diligence questions are answerable at all. ## Why advisors are looking at this now The clients arrived before the industry did. One in four U.S. adults owns crypto, about 67 million people, per the National Cryptocurrency Association’s 2026 survey. The Bitwise/VettaFi 2026 Benchmark Survey found that 77% of advisors’ crypto-owning clients hold it outside the advisory relationship, up from 71% the year before, and that 32% of advisors placed crypto in client accounts in 2025, an all-time high. A held-away position is unadvised risk inside a household the advisor already serves: unknown wallets, unknown protocols, no diligence file, no position limit. The rest of this guide is how a fiduciary closes that gap without pretending the asset class is simpler than it is. ## The three-layer test Evaluate every position at three layers: 1. Asset: What does the client legally and economically own? Who can freeze, redeem, dilute, or change it? 2. Protocol: What contracts hold the asset? Who can upgrade or pause them? Which oracles, bridges, curators, and governance processes matter? 3. Chain: Who orders transactions? Can the chain halt? Who controls upgrades and bridges? **Working rule** A position is only as strong as its weakest layer. “Non-custodial protocol” does not make an issuer-controlled asset sovereign. ## The test, applied to one position Take a client supplying USDC to a lending protocol. Layer one, the asset: USDC is issuer-controlled. Circle administers the token contracts, can block addresses, and stands behind the redemption framework the client’s dollar claim runs through. That is not a defect; it is a fact for the file, recorded in the [USDC control profile](/atlas/assets/usdc/). Layer two, the protocol: the lending market is a set of upgradeable contracts. Someone holds the upgrade key, someone sets collateral parameters, and an oracle feeds it prices. Each is a named dependency. Layer three, the chain: on [Ethereum](/atlas/chains/ethereum/), no single operator can halt settlement or reorder transactions by decision. The position’s effective control is the weakest of the three layers, and here that is the asset layer. The same test disposes of harder cases at the first layer. Binance Staked ETH appears in DeFi dashboards next to non-custodial staking. Apply the asset question and the answer is immediate: the token is a claim on a centralized exchange, wrapped so that it moves on-chain. [Ketju rejected it on that finding alone.](/rejections/binance-staked-eth/) ## Where the yield comes from Yield is compensation paid by someone or created by some mechanism. Lending yield comes from borrowers. Staking yield comes from protocol issuance and transaction fees. Tokenized Treasury yield comes from the underlying government securities, less fees. Private-credit yield compensates the lender for borrower and recovery risk. Reward-token yield is often a temporary subsidy. Classifying the source matters more than comparing the percentage. Two positions displaying 5% may contain entirely different duration, liquidity, credit, smart-contract, and control risks. Ketju’s clearest published case: Maple Finance screened near 5%, a modest premium over collateralized lending, and the premium was compensation for undercollateralized institutional credit, a different risk entirely. [The published rejection walks the whole case.](/rejections/maple/) ## Limits, review conditions, and the monitoring duty A decision file must state which facts reopen each layer. Ketju’s memo format, published for every current research record in [the research files](/rejections/), pairs each assessment with observable review conditions. The useful triggers are the ones a monitoring system can actually watch. An admin key changes hands. A timelock is shortened. Withdrawable liquidity stays under a floor for a set number of hours. A depeg passes a set depth and duration. An exploit crosses a dollar threshold. A governance vote alters the collateral rules. Monitoring is a duty because the position does not hold still after research or selection. Code upgrades, oracle changes, and new collateral classes happen without asking the holder. Kelp’s rsETH was a live instrument with audits behind it when a forged bridge message minted about 116,500 unbacked tokens in April 2026; diligence written a quarter earlier described a system that no longer existed. [The memo records the event.](/rejections/kelp/) A written re-review date and a named owner are the difference between a decision and an opinion. ## The workflow, start to finish 1. Inventory what the household already holds, including held-away wallets, and identify every position by exact token, contract, and chain. 2. Classify each position at the three layers and pull the control facts from primary sources. The [Atlas](/atlas/) records the asset and chain layers for the instruments advisors meet first. 3. Write the research assessment, firm-shelf decision, client-specific constraints, advisor rationale, and review conditions before money moves, and file the evidence. 4. Set the review date and the data source for each trigger. 5. State the risks in client language and record the suitability rationale. The [due-diligence checklist](/due-diligence/defi-checklist/) expands each step, and the [file format](/due-diligence/defensible-file/) shows the finished record. ## What belongs in the file A defensible process records the instrument, dependencies, thesis, disqualifiers, position limit, evidence, reviewer, review date, and observable events that revoke the decision. Rejections deserve the same documentation as approvals. They show that the universe was examined rather than merely ranked. Continue with the [DeFi due-diligence checklist](/due-diligence/defi-checklist/) or compare [DeFi and tokenized securities](/on-chain-finance/defi-vs-tokenized-securities/). ## Sources - [Crypto Assets](https://www.investor.gov/additional-resources/spotlight/crypto-assets) — Investor.gov --- Published by Ketju Research (https://ketjuresearch.com). Educational analysis only; not legal, tax, compliance, or investment advice. Machine-readable index: https://riadefi.com/llms.txt · Content API: https://riadefi.com/content.json