The LOI gets you a seat at the table. Due diligence tells you what you're buying. But the purchase agreement is the only document that actually protects you after the wire clears. Most buyers download a template, change the names, and sign away the exact protections they need most.
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 — pulling Google Analytics exports, verifying Stripe payouts, interviewing suppliers — and then sign a purchase agreement they downloaded from a legal template site in about forty minutes. It's backwards. Due diligence tells you whether the business is real. The purchase agreement is what you hold in your hand if it turns out it wasn't.
An asset purchase agreement (APA) is the legally binding contract that transfers the business from seller to buyer. It comes after the letter of intent is accepted and after due diligence is substantially complete. Unlike the LOI, which is mostly non-binding, the APA is enforceable. Every word matters. Every omission costs you.
The problem with generic templates is that they were written for physical businesses — companies with inventory, leases, equipment, and employees. Online businesses transfer differently. Your assets are domain registrations, Amazon Seller Central accounts, Google Ads accounts, Facebook Business Manager access, Shopify stores, email lists sitting in Klaviyo, and code repositories. A template written for a car wash acquisition does not have a clause about who controls the two-factor authentication on the domain registrar. And that's exactly the clause you'll wish you had.
This guide walks through the six sections that actually determine whether your APA protects you, what to insist on in each, and how to work with an attorney who understands digital assets. If you're actively evaluating deals, Deal Alert AI surfaces the deal intelligence that tells you which of these protections to fight hardest for on a specific listing.
This section looks simple and almost never is. The exact dollar amount goes here, but so does the entire structure of how that money moves. If you're paying $340,000 for a content site, the agreement needs to specify how much is wired at closing, how much sits in escrow, how long it sits there, and under what conditions it releases.
Escrow holdbacks are standard and you should push for one. A typical structure on a $340,000 deal might be $290,000 at closing with $50,000 held in escrow for 90 days, released after the buyer confirms traffic and revenue have not collapsed and all asset transfers completed cleanly. Sellers resist this. Reasonable sellers accept it. If a seller refuses any holdback whatsoever on a six-figure deal, that tells you something about their confidence in the business.
If there's seller financing, the terms need to be spelled out to the decimal. Interest rate, amortization schedule, payment dates, what constitutes default, and what happens on default. I've seen agreements that say "seller financing of $80,000 over 24 months" and nothing else. That's not a term — that's a wish. Does interest accrue? Is it interest-only for six months? Can the buyer prepay without penalty? If the buyer misses a payment, does the seller get the business back, or just a claim in court? Answer all of it in writing.
Key insight: Earn-outs are the single most litigated clause in small online business acquisitions. If you're using one, define the metric with surgical precision. "Revenue" is not a metric. "Net revenue recognized in Stripe and Amazon Seller Central, measured on a cash basis, excluding refunds and chargebacks, for the trailing twelve months beginning on the closing date" is a metric. The difference between those two definitions has cost buyers five figures in disputes.
Earn-outs also need a measurement methodology and a payment schedule. Who calculates the number? Does the seller get audit rights? What happens if the buyer changes the business model in a way that suppresses the earn-out metric — say, shifting from Amazon to a direct-to-consumer store? Good earn-out clauses include a covenant that the buyer will operate the business in good faith during the earn-out period. Sellers ask for this and it's fair.
We scan Empire Flippers, Flippa, Acquire.com and Quiet Light daily — scoring every listing. Start free.
This is where template agreements fail online business buyers most badly. The asset schedule is the explicit list of what transfers. If it's not on the list, you may not own it, and arguing about it after closing is a losing position.
Your schedule should name, individually: every domain name including redirects and defensive registrations, all website files and databases, hosting accounts, the source code and any private repositories, all social media accounts with handles listed, email lists and the ESP account they live in, all software licenses and SaaS subscriptions used to run the business, supplier agreements and any preferential pricing terms, customer contracts, trademarks and copyrights, product photography and design files, standard operating procedures and documentation, and any Amazon brand registry or trademark registrations.
I once reviewed a deal where the buyer acquired a $600,000 e-commerce brand and discovered post-closing that the product photography — thousands of dollars of professional shoots — had been done by a freelancer who retained copyright and had never assigned it. The APA said "all intellectual property owned by the Seller." The seller didn't own the photos. Technically true, technically compliant, and the buyer had to re-shoot the entire catalog.
Also be explicit about what does not transfer. Sellers often keep a personal LinkedIn, an unrelated side project hosted on the same server, or a shared email address. Naming the exclusions prevents ugly conversations at handover. And specify the mechanics of transfer for each account type — domain unlock and auth code delivery, Google Analytics ownership transfer, Amazon account transfer procedure, and who covers any transfer fees.
Representations and warranties are the seller's formal, legally binding statements about the business. They're not marketing claims from the listing page. They're sworn assertions that carry consequences if false. This is where your due diligence findings get converted into legal protection.
At minimum, you want reps covering: financial statements accurately reflect revenue and expenses for the trailing 24 months; traffic figures in the Analytics account are genuine and not artificially generated; there are no undisclosed liabilities, debts, or obligations; there is no pending or threatened litigation; the seller owns or has valid license to all IP being transferred; the business is in compliance with the terms of service of every platform it operates on; no affiliate, advertising, or supplier relationship has been terminated or threatened with termination; and no material adverse change has occurred between the LOI date and closing.
That platform compliance rep matters more than buyers realize. If you're buying an Amazon FBA business, an affiliate site dependent on a single program, or anything running on a major ad network, a hidden policy warning or suspension history can vaporize the asset. Make the seller state in writing that there have been no policy violations, warnings, or account suspensions in the preceding 24 months. If the seller hesitates on that specific rep, you've found something worth investigating.
Watch for knowledge qualifiers. Sellers' attorneys love to insert "to the Seller's knowledge" before every representation. This drastically weakens your protection — it means the seller only breaches if they knew the statement was false. For core financial and traffic representations, push back hard and demand unqualified reps. A seller who won't state unqualified that their revenue figures are accurate is telling you they aren't confident in their own numbers.
Deals sourced through vetted marketplaces come with some verification already done. Listings on Empire Flippers go through financial vetting before they're published, which gives you a stronger baseline. Marketplaces like Flippa have a wider spectrum of listing quality, which makes robust representations even more important. Either way, the marketplace's verification doesn't substitute for reps in your APA — vetting is a snapshot, representations are a contract.
Representations without indemnification are decorative. Indemnification is the enforcement mechanism: if a representation turns out to be false and you suffer a loss, the seller compensates you. This section defines the scope, the time window, and the caps.
The survival period is your first negotiation point. This is how long after closing the representations remain enforceable. For online businesses, 18 to 24 months is a reasonable ask. Sellers will push for 6 to 12. The logic for a longer window is straightforward: some problems in online businesses don't surface for a full seasonal cycle. A Google core update might not hit until month nine. A supplier relationship might not sour until the next annual contract negotiation. Fundamental reps — IP ownership, authority to sell, tax liabilities — should survive longer or indefinitely.
Next is the cap. Many APAs cap the seller's total indemnification liability at some percentage of the purchase price — 10%, 20%, sometimes the full amount. On a $340,000 deal, a 15% cap means your maximum recovery is $51,000 regardless of how bad the misrepresentation was. Push for a higher cap on core representations. Fraud should always be carved out of any cap entirely.
Then there's the basket, or threshold. This says the seller isn't liable until claims exceed a minimum amount — commonly 0.5% to 1% of purchase price. It exists to prevent nuisance claims over trivial discrepancies. That's fine, but make sure it's a "tipping basket" (once you cross the threshold, you recover from dollar one) rather than a "deductible basket" (you only recover the excess above the threshold). That distinction is worth real money.
Finally, connect indemnification to your escrow. The cleanest structure ties the escrow holdback directly to indemnification claims, so if a breach surfaces in the first 90 days you have funds to draw against rather than a lawsuit to file against a seller who may have already moved the money offshore.
You're not just buying assets. You're buying the absence of the seller as a competitor. Someone who spent four years building a supplement brand knows the suppliers, the ad angles, the keyword gaps, and the customer psychology. If they can relaunch a near-identical brand ninety days after closing, you've bought considerably less than you paid for.
A workable non-compete for an online business typically runs 24 to 36 months, defines the restricted activity by niche and business model rather than by vague industry labels, and covers the geographies where the business actually operates. Be specific. "Seller will not compete in the health and wellness industry" is so broad a court may refuse to enforce it. "Seller will not, for 24 months, own, operate, advise, or hold more than a 5% interest in any business selling magnesium or sleep supplements direct-to-consumer in the United States, Canada, or the United Kingdom" is enforceable and clear.
Non-solicitation is the companion clause and it's often forgotten. It stops the seller from poaching the customer list, the email subscribers, the freelance team, the VAs, and — critically for e-commerce — the supplier relationships. If your business depends on one manufacturer in Shenzhen with favorable pricing, a seller who circles back and launches a competing SKU through that same factory can wreck your margins.
Key insight: Enforceability of non-competes varies significantly by jurisdiction, and some places are hostile to them entirely. When the non-compete is a material part of what you're buying, allocate an explicit portion of the purchase price to it in the agreement — say $15,000 of a $340,000 deal. This strengthens the argument that it was bargained-for consideration and makes it far harder to challenge later.
The handover period is where most acquisitions actually succeed or fail operationally. Your APA needs to specify exactly what the seller owes you after the money moves, in hours and deliverables, not in vibes.
Define the support period — 30, 60, or 90 days is typical depending on complexity. Define the format: how many hours per week, availability windows, response time expectations, and preferred channel. "Seller will provide reasonable assistance" means nothing. "Seller will provide up to 10 hours per week for the first 30 days and 5 hours per week for the following 30 days, responding to buyer inquiries within 24 business hours" means something.
Specify the deliverables too. Full documented SOPs for every recurring operational task. A written introduction to every supplier, contractor, and key partner. Credential transfer for every account with two-factor authentication moved to buyer-controlled devices. A walkthrough of the ad accounts including historical campaign structure and what's been tested. Access to any private communities, wholesale portals, or affiliate manager relationships.
For businesses that can't be immediately transferred — some Amazon accounts, certain payment processors, apps in an app store — you may need a transition services arrangement where the seller continues operating the account on your behalf for a defined period. These are genuinely risky and need careful drafting: who holds the funds, how often they remit, what happens if the seller becomes unresponsive, and what indemnity protects you if the seller does something that gets the account suspended while it's still in their name.
Run through this before you sign anything. If you can't answer every item with a specific clause reference in your draft, the agreement isn't finished.
Not every business attorney is the right attorney for this. General M&A lawyers are excellent at corporate structure, stock purchases, and employee matters — and largely unfamiliar with the specific failure modes of online businesses. You want someone who has closed website, e-commerce, SaaS, or app acquisitions before and can tell you about the deals they've worked on.
Ask direct questions when you interview. How many online business acquisitions have you closed in the past two years? What's your standard approach to platform compliance representations? Have you drafted an Amazon Seller Central transition arrangement? How do you handle escrow on a deal where a domain transfer might take five days? An attorney who answers these fluently is worth 2x the hourly rate of one who has to research each one.
On cost: for a deal in the $100,000 to $500,000 range, expect roughly $2,500 to $7,500 for APA drafting and negotiation depending on complexity and how much back-and-forth the other side generates. On a $300,000 acquisition, that's around 1–2% of purchase price. Buyers routinely skip it to save money and then discover the actual cost of a bad agreement measured in five figures. The math is not close.
Bring your attorney your due diligence findings, not just the LOI. If you found that 62% of traffic comes from one keyword, or that a single supplier accounts for 80% of COGS, or that the Facebook ad account had a policy strike eighteen months ago — those specific facts should become specific representations. Generic agreements protect against generic risks. Your deal has particular risks, and those are the ones that will actually bite you.
The best purchase agreements are built backwards from diligence. Every concentration risk, every anomaly in the numbers, every dependency you uncovered should map to a clause. That mapping is the whole game — it's how you convert what you learned into what you can enforce.
This is where Deal Alert AI fits into the process. We track listings across the major marketplaces and surface the structural characteristics of each deal — revenue concentration, traffic source mix, platform dependency, pricing relative to comparable transactions, and how a listing's fundamentals stack against the multiple being asked. That intelligence tells you which representations to prioritize before you're paying an attorney by the hour to figure it out.
A practical example: if the data shows a content site derives most of its revenue from a single affiliate program, you already know your APA needs a specific rep confirming that program is in good standing with no notice of termination or commission change, plus an indemnity if it turns out otherwise. If an e-commerce brand is concentrated in one SKU, you need supplier agreement transferability spelled out explicitly. The pattern of the deal dictates the shape of the contract.
Whether you're browsing vetted listings on Empire Flippers or sifting through the broader inventory on Flippa, the discipline is the same. Verify, then document what you verified, then get it into the agreement in language a court would enforce. Start with the deal flow at Deal Alert AI, run your diligence properly, and treat the purchase agreement as the last and most important step rather than a formality on the way to the wire.
The LOI gets you exclusivity. Due diligence gets you knowledge. The purchase agreement is the only one of the three that's still working for you at 2 a.m. eight months after closing when something goes sideways. Spend the money. Read every line. Make them name every asset.
By Sophal Lanh, Founder of Deal Alert AI
This article is for informational purposes only and does not constitute legal advice. Consult a qualified attorney licensed in your jurisdiction before entering into any acquisition agreement.
We scan Empire Flippers, Acquire, Flippa, and Quiet Light daily. The best sub-$500K businesses are gone within 48 hours.