How to Structure the Purchase Agreement When Buying an Online Business

Deal Alert AI is reader-supported. We earn commissions from affiliate links at no cost to you.

This post is based on a video from our Deal Alert AI YouTube channel. Watch the original or read the full breakdown below.

I've watched buyers spend six weeks on due diligence, pull apart every line of a Profit and Loss statement, verify traffic in Google Analytics, and interview three suppliers — then sign a five-page purchase agreement the seller's cousin drafted from a template. That's like inspecting a house down to the foundation bolts and then accepting a handshake for the deed.

The purchase agreement — sometimes called the Asset Purchase Agreement (APA) or Share Purchase Agreement (SPA) — is the only document that actually governs your acquisition. The Letter of Intent is largely non-binding. The seller's email promising "I'll help you for three months, no problem" is not enforceable. The broker's listing page claiming the business has no pending legal disputes is marketing copy, not a warranty.

If it's not in the purchase agreement, it doesn't exist. Full stop.

This post walks through the six core sections of an online business purchase agreement, what buyer-friendly language looks like in each, and where deals typically go sideways. I'm not an attorney and this isn't legal advice — but after years of looking at deal structures and building Deal Alert AI to help buyers evaluate listings, I've seen enough agreements to know which clauses cost people money.

Why You Need an Attorney — And What "Experienced" Actually Means

Let me get the obvious out of the way first: hire a lawyer. Not your family lawyer who does wills and real estate closings. Not a generalist small-business attorney who's never handled a domain transfer. You want someone who has actually closed digital asset acquisitions and understands how these businesses work mechanically.

Here's why that specificity matters. A traditional M&A attorney will know how to structure indemnification caps and survival periods. But will they know to ask whether the Amazon Seller Central account can legally transfer, or whether Amazon's Terms of Service require a full account transfer versus a listing migration? Will they know that a Google AdSense account cannot be transferred at all and the buyer must open their own? Will they understand that Shopify apps, Stripe accounts, Meta Business Manager assets, and Facebook ad pixels each have their own transfer mechanics — and that some of them can't be transferred, only rebuilt?

Cost expectations: for a deal under $250,000, expect to pay $2,500 to $6,000 for a competent attorney to review and revise a purchase agreement. For deals in the $500K to $2M range, budget $7,500 to $20,000 depending on complexity and how much back-and-forth negotiation happens. On a $1.2M acquisition, $12,000 in legal fees is one percent of purchase price. That is cheap insurance against a seller who "forgot" to mention a pending trademark opposition.

Section One: Asset Identification and Transfer

This section defines exactly what you are buying. It sounds trivial. It is not. This is where the largest number of post-closing disputes originate, and the reason is simple: buyers assume things transfer, and sellers assume things don't.

The governing principle is that anything not explicitly listed may not transfer. So the schedule of assets attached to the agreement needs to be exhaustive. At minimum, it should enumerate: all domain names including parked and defensive registrations, all website files and source code, all content including articles, images, videos, and product descriptions, all trademarks and copyrights registered or unregistered, all social media accounts with handles specified, all email lists with subscriber counts, customer databases and CRM records, supplier and vendor contracts, active affiliate agreements, hosting accounts, third-party tool subscriptions, standard operating procedures and documentation, and any physical inventory with agreed valuation methodology.

I've seen a real case where a buyer acquired a content site for $340,000 and discovered after closing that the seller had built the site's illustrations using a personal Canva Pro account and a stock photo license issued in the seller's individual name. The images were not transferable. The buyer had to either relicense roughly 800 images or replace them. That's a five-figure problem created by one missing line in an asset schedule.

Also pay attention to excluded assets. Sellers frequently want to keep an email list they use for other projects, or a related domain, or a Facebook group they built. That's often negotiable and sometimes reasonable — but it has to be explicit. Ambiguity always favors whoever has more money to spend on litigation, and that's usually not the first-time buyer.

Section Two: Purchase Price and Payment Structure

This section states the total consideration and precisely how it gets paid. In a simple all-cash deal this is short. In a structured deal — which is most deals above $500K — it gets complicated fast, and every ambiguity is a future fight.

A typical structured deal in the $800K to $3M range might look like: 70% cash at closing, 20% seller note over 24 months at 6-8% interest, and 10% tied to an earn-out based on trailing twelve-month revenue at the 12-month mark. Each of those three components needs its own precise language. For the seller note, specify the principal amount, interest rate, amortization schedule, first payment date, prepayment rights, default definitions, and whether the note is secured by the assets. For the earn-out, specify the exact metric, who calculates it, what accounting method applies, what happens if the buyer changes the business materially, and what audit rights the seller has.

Earn-outs are where I see the most disputes. A seller agrees to an earn-out based on "net profit" — then the buyer hires a full-time operator, adds $80,000 of salary expense, and net profit collapses. The seller screams that the buyer sabotaged the earn-out. The buyer says they ran the business normally. Neither party is obviously wrong because the agreement didn't define the metric tightly enough. If you use an earn-out, tie it to revenue or gross profit rather than net profit whenever possible — those metrics are harder to manipulate on either side.

Section Three: Representations and Warranties

Reps and warranties are factual statements the seller makes about the business, which they become legally liable for if false. This is the section that converts due diligence findings into enforceable protection. If your diligence revealed the business has 14,300 email subscribers and $47,000 in trailing twelve-month net profit, those numbers should appear as representations.

The core reps you want from any online business seller: that the financial statements provided are accurate and prepared consistently; that there are no undisclosed liabilities, debts, or obligations; that there is no pending or threatened litigation, arbitration, or governmental investigation; that the seller has valid and exclusive ownership of all intellectual property being transferred and that it doesn't infringe on any third party's rights; that the business is in compliance with all applicable laws including tax, consumer protection, privacy (GDPR, CCPA), and advertising disclosure rules; that all material contracts have been disclosed and are in good standing; that no supplier, customer, or affiliate representing more than 10% of revenue has indicated an intent to terminate; and that the business has not received any notice of Terms of Service violations from Google, Amazon, Meta, or other critical platforms.

That last one is underrated. A content site that has received a manual action notice from Google, or an Amazon FBA business with a suppressed listing history, carries meaningfully more risk than the financials suggest. Ask for the rep in writing. If a seller resists warranting that they've received no platform violation notices, that hesitation is itself information.

Pay attention to knowledge qualifiers. Sellers' attorneys love to insert "to the Seller's knowledge" in front of every representation, which transforms a hard warranty into a soft one. If the seller says "to my knowledge there is no pending litigation," and litigation was filed but they claim they hadn't seen the notice, you may have no recourse. Push back on knowledge qualifiers for facts the seller absolutely should know — their own financials, their own IP ownership, their own contracts.

Section Four: Indemnification and Survival Periods

Indemnification is the enforcement mechanism. It answers: if a representation turns out to be false and it costs me money, who pays? Without a real indemnification clause, your reps and warranties are decorative.

Three numbers matter here. First, the survival period — how long after closing the seller remains liable. Standard for online business deals is 12 to 24 months for general reps, and longer or indefinite for "fundamental" reps like title to assets, authority to sell, and tax obligations. A 6-month survival period is aggressive in the seller's favor; many issues (a Google algorithm update revealing manipulated traffic, a tax notice, a supplier dispute) don't surface within six months.

Second, the cap — the maximum the seller can be liable for. Sellers usually push for 10-20% of purchase price. Buyers want 100%, or at minimum 30-50%. On a $900,000 deal, a 15% cap means $135,000 of protection. If the seller misrepresented revenue by 30%, that cap won't make you whole. Fundamental reps and fraud should always be uncapped.

Third, the basket or threshold — the minimum aggregate damages before indemnification kicks in, typically 0.5% to 1% of purchase price. This exists so sellers aren't chased over $400 problems. Reasonable. But make sure it's a "tipping basket" (once you exceed the threshold, you recover from dollar one) rather than a "deductible basket" (you only recover the amount above the threshold) if you can negotiate it.

The most practical protection is a holdback: 10-15% of purchase price held in escrow for 12 months, released if no claims arise. It converts an indemnification promise from a lawsuit into a bookkeeping entry. Sellers dislike it. Push anyway — especially on deals where you're buying from an individual rather than an entity with assets.

Section Five: Non-Compete and Non-Solicitation

You are buying a business, but you're also buying the seller's agreement not to immediately rebuild it. Without a non-compete, nothing stops a seller from launching a near-identical site 60 days after closing, emailing their personal network, and cannibalizing the asset you just paid a 40x monthly multiple for.

A well-drafted non-compete has four dimensions: duration (typically 2 to 5 years — 3 years is the common landing spot), geographic scope (usually worldwide for an online business, since the internet has no geography), business scope (defined by niche, product category, or customer type — not so broad it becomes unenforceable), and prohibited activities (owning, operating, advising, investing in, or being employed by a competing business).

Non-solicitation is separate and equally important. It prevents the seller from poaching customers, email subscribers, affiliates, suppliers, contractors, and employees. If the business runs on three key freelance writers and a virtual assistant, the seller shouldn't be able to hire them all away on day 31. Name them in a schedule if they're that critical.

Two practical notes. First, enforceability varies dramatically by jurisdiction. California is famously hostile to non-competes; some countries require paid consideration for a non-compete to hold. Your attorney needs to know where the seller is domiciled and draft accordingly — sometimes structuring part of the purchase price as explicit consideration for the non-compete. Second, if you're buying from a seller who has a portfolio of similar sites, expect a carve-out. That's often fine, but the carve-out should list specific existing properties by URL, not describe a category the seller can expand into freely.

Section Six: Transition, Training, and the Closing Checklist

The transition clause is where a lot of buyers under-negotiate because it feels like housekeeping. It isn't. The difference between a smooth 30-day handover and a six-month scramble is usually whether the obligations were written down with specificity.

Vague language like "Seller will provide reasonable training and support" is worthless. Specify: number of hours (e.g., 40 hours over the first 60 days), response time for questions (e.g., within 48 business hours), format (video calls, documented SOPs, screen recordings), duration (30, 60, or 90 days is typical — longer for complex operations like FBA or agency businesses), and consequences for non-performance. On larger deals, tie a portion of the purchase price or holdback release to completion of transition obligations. That's the only leverage that reliably works once the seller has your money.

Also build a closing checklist directly into the agreement or as a schedule. Here's the operational sequence that should be documented and completed:

1. Escrow agent confirms receipt of buyer funds and all closing documents are executed by both parties.
2. Domain names initiated for transfer at the registrar, with auth codes provided and transfer lock removed.
3. Hosting account access transferred or a full site migration completed to buyer-controlled hosting.
4. All platform accounts transferred: Google Analytics, Google Search Console, Ahrefs/Semrush, email service provider, payment processors, ad accounts.
5. Social media accounts transferred with primary admin rights, recovery emails, and 2FA reassigned to the buyer.
6. Supplier, vendor, and affiliate relationships formally introduced and contracts assigned in writing where assignment is required.
7. Customer and subscriber data transferred in a documented format, with a signed data processing acknowledgment for GDPR/CCPA compliance.
8. All SOPs, passwords, and documentation delivered via a shared password manager vault rather than email or spreadsheet.
9. Buyer confirms operational control by independently logging into every account and verifying revenue is flowing to buyer-controlled accounts.
10. Escrow releases funds to seller only after the buyer signs off on the verification step above.

That last item is the one people skip. Never release escrow before you've personally verified you control every asset and revenue is landing in your bank account. On Flippa and other marketplaces, the standard inspection period after transfer exists precisely for this. Use the full window.

Asset Purchase vs. Stock Purchase — And Why It Matters for Tax

Most online business acquisitions under $5M are structured as asset purchases rather than stock or membership-interest purchases. There's a good reason for that from the buyer's side: in an asset purchase, you're buying specific assets and explicitly assuming only specified liabilities. You don't inherit the entity's history — unknown tax obligations, old contracts, prior lawsuits, undisclosed debts.

In a stock or equity purchase, you buy the company itself, and the company brings everything with it, known and unknown. Sellers often prefer equity deals because the tax treatment can be more favorable (capital gains on the whole thing rather than partial ordinary income allocation) and because it's cleaner for them — contracts, licenses, and accounts often stay in place without needing assignment.

There are cases where an equity purchase makes sense: when critical contracts have anti-assignment clauses, when the business holds licenses or platform accounts that can't be transferred, or when the entity has an operating history that's valuable (an established Amazon seller account, for instance). If you go that route, your diligence needs to go deeper and your indemnification protections need to be significantly stronger — bigger caps, longer survival periods, and a larger holdback.

Also negotiate the purchase price allocation. In an asset deal, the total price gets allocated across asset classes — goodwill, intangibles, non-compete value, inventory, equipment — and that allocation drives both parties' tax treatment. Buyer and seller often have opposing interests here. Get your accountant involved before signing, not after. On a $1M deal, allocation differences can move your effective tax position by tens of thousands of dollars.

Negotiating From a Position of Market Knowledge

Here's the part most buyer guides skip: your leverage in purchase agreement negotiations is directly proportional to how well you understand the market. A seller's attorney will push for a 12-month survival period, a 10% cap, and a 24-month non-compete. Whether you accept that depends on whether you know what's standard — and whether you have alternatives.

If this is the only listing you're seriously considering, you'll fold on terms because you're emotionally committed. If you're tracking 15 comparable businesses across multiple marketplaces and you know that a similar SaaS at a 3.4x multiple listed last week, you negotiate differently. Deal flow is leverage. That's the entire reason I built Deal Alert AI — to give buyers continuous visibility into what's actually listed, at what multiples, across the marketplaces that matter, so no single deal ever feels like the last train out.

Practically, that means before you sign an LOI, you should know: the typical multiple range for this business model and size, what comparable listings sold for, whether the asking price already assumes seller financing, and what deal terms similar buyers accepted. That context turns "the seller says a 15% cap is standard" into "I've seen three comparable deals at 30% caps with 18-month holdbacks, and here's why this business's traffic concentration justifies more protection, not less."

Use the marketplaces to build that context. Empire Flippers publishes multiple data and has standardized deal processes that give you a baseline for what "normal" looks like. Flippa gives you volume and visibility into a broader range of deal sizes and structures. Then use Deal Alert AI to keep that pipeline running so you're never negotiating a purchase agreement from a place of scarcity.

The purchase agreement is the last document you sign and the first one you'll reach for if something goes wrong. Spend the money on the attorney. Read every schedule. Verify before escrow releases. And never accept a term because you're tired of negotiating — that's precisely when the expensive mistakes get signed.

By Sophal Lanh, Founder of Deal Alert AI