Stop Valuing Every Online Business the Same Way. SaaS Math Will Wreck You on an Ecommerce Deal.

The standard pitch goes like this: find the SDE, apply a multiple, you've got a value. Clean. Universal. Works for any online business. I believed this for longer than I should have, and it cost me on my second content site acquisition. I was using the same mental model I'd built buying an ecommerce brand. I looked at the earnings, I looked at the multiple, I signed. What I didn't look at was the traffic breakdown. Eighty-one percent organic search, nearly all from three informational clusters, no Search Console history provided. Six months post-close, a core update hit. Revenue dropped 40% in eight weeks. The SDE math was fine. The model-specific risk analysis was nonexistent.

The multiple is not a universal number that floats above business models. It's a compression of model-specific risk into a single figure. When you don't understand the model, you can't read the multiple. You're just guessing at whether 4x is cheap or expensive without knowing what you're actually buying.

The Consensus Position and Why It's Incomplete

Every valuation guide you'll find online treats SDE times a multiple as the universal formula. SDE is the right earnings metric for most sub-$10M deals. That part is correct. The problem is that most guides stop there and imply the same checklist applies regardless of what you're buying. Content site, FBA brand, B2B SaaS, newsletter. Same formula, different number. If that's how you're thinking about it, you're missing the part that actually matters. ¹

SaaS businesses routinely trade at 6x to 10x annual earnings or higher. Content and affiliate sites typically fall in the 2x to 4x range. Ecommerce profit multiples had stabilized at roughly 3.98x on average by late 2024. The gap isn't arbitrary. It reflects what the market understands about model-specific durability. But knowing the typical multiple for each category is not the same as knowing how to evaluate whether a specific deal deserves that multiple. Or whether it deserves a haircut. ², ³

SaaS: The Earnings Metric Is Almost Secondary

When I looked at a small B2B SaaS a few years ago , $210K SDE, clean books, 90 paying customers. The broker walked me through the earnings multiple the same way he'd have walked me through an FBA deal. SDE times a reasonable multiple. I almost fell for it. What actually mattered was the monthly churn rate (it was 4.2%, which is brutal), the CAC relative to LTV, and whether the MRR was growing or contracting.

A 4.2% monthly churn means you're losing roughly half your customer base in a year. The SDE looked fine in the TTM because the owner had been aggressively acquiring customers to paper over the exits. Strip out the acquisition spend, normalize to what a new owner with less hustle would actually run, and the real SDE was closer to $140K. Same business. Different number once you understand what drives SaaS value.

A CLV-to-CAC ratio of 3:1 to 5:1 is considered ideal and signals marketing efficiency. A 1.5:1 ratio, which sounds survivable, means you're spending two-thirds of what a customer will ever pay you just to get them. That's not a business with a recurring revenue premium. That's a business with a structural acquisition problem dressed up in MRR clothing. The multiple should reflect that. Most first-time buyers don't know to look. ⁴

Contractually recurring revenue is the single most important value driver in any software business. Not growth. Not margins. Recurring revenue. When that revenue is churning at rates above 2% monthly, the premium the market assigns to SaaS models evaporates fast. ⁵

Ecommerce: The Risk Is in the Supply Chain and the SKU Concentration

Ecommerce looks simple on paper. Revenue minus COGS minus expenses equals profit. The risks that kill ecommerce deals are almost never in the income statement. They're in the supplier relationships, the return rates, the ASIN concentration, and the platform dependency.

On the returns side: U.S. online retailers are looking at roughly 19.3% of online sales being returned. That number is an average. For certain categories. Apparel, consumer electronics, anything size-dependent. It runs higher. If you're looking at an ecommerce deal and the seller hasn't provided return rate data broken out by SKU, ask for it. A business with a 28% return rate on its top three SKUs is a fundamentally different asset than one with 9%, even if TTM SDE is identical. ⁶

ASIN concentration is the FBA equivalent of customer concentration. If a business generates 70% or more of revenue from one marketplace listing, you're one suppression, one hijacking, or one policy change away from a bad quarter. That concentration risk knocks the multiple down materially. One framework discounts the multiple by approximately 0.8x for heavy single-platform dependency. That's $80K off the price on a $100K SDE business at 4x. Real money. ⁷

Then there's supply chain transferability. When the owner is the personal relationship with the factory in Shenzhen, that's not a business asset. That's a personal contact list. You need supplier agreements in writing, minimum order terms documented, lead times that don't depend on a phone call to someone who doesn't know you exist yet.

Content Sites: The Entire Story Is in the Traffic

Content sites are the model where I've seen the widest spread between what buyers pay and what they actually get. Partially because the business looks passive. Partially because the financials are simple. Partially because most buyers don't look hard enough at where the traffic comes from.

Traffic source diversification is not a nice-to-have. It is the entire risk profile. A content site generating $180K SDE with 85% organic search traffic from a handful of keyword clusters is priced like a stable asset. It isn't. It's a bet on Google's continued goodwill. I've owned that bet. It doesn't always pay.

Request Search Console access on every content deal before you sign an LOI. Specifically look for manual action history, traffic drops correlated with known algorithm update dates, and page-level traffic to understand whether growth is concentrated in a few posts or distributed across the site. A site with 400 ranking pages and distributed traffic is a different asset than one with 12 pages that generate 80% of clicks. The SDE can look identical. The durability is not.

Backlink profile quality matters too, though it's harder to get at quickly. Lots of sellers have bought links at some point. Some link profiles are ticking penalties. You can pull this in Ahrefs or SEMrush in an hour. A backlink profile full of low-DR exact-match anchors on irrelevant sites is not a neutral data point. That's a future manual action waiting.

The Objection: SDE Normalizes Across Models, So Why Does This Matter?

The strongest pushback on this whole argument is: SDE already accounts for model differences because it normalizes to what a new owner would actually earn. A content site with algorithm risk will have lower SDE if it's already declining. An FBA business with supplier concentration will show up in the margins. So why do model-specific failure modes matter if the earnings metric captures them?

Because TTM SDE is backward-looking and the failure modes I'm describing are forward-looking risks that haven't materialized yet. The content site I mentioned earlier was growing at the time of purchase. Positive TTM trend, clean financials, justifiable multiple. The algorithm risk was latent. It didn't show up in historical SDE. It showed up six months post-close.

The multiple is supposed to price future risk. But you can only apply a risk-adjusted multiple if you know which risks to adjust for. That requires model-specific knowledge. A buyer who learned acquisition math exclusively on SaaS will look at a content site and think about churn. Churn is irrelevant. A buyer trained on FBA will look at supplier terms when evaluating a newsletter. Supplier terms don't exist. They're asking the wrong questions because they're applying the wrong mental model.

Newsletters Are Their Own Thing Entirely

I'll add newsletters because I bought one and almost every valuation framework I tried to apply from my previous deals was wrong. A newsletter's value is almost entirely in subscriber engagement and list health. Open rates, click rates, churn on paid subscribers if it's a paid product, and the relationship the audience has with the author's voice.

The hard problem: if the newsletter's voice is the founder's personal brand, you're not buying a transferable asset. You're buying a leased audience that may or may not follow you once the name on the byline changes. I'd want at minimum 60 days of post-transition performance data before I'd put a full multiple on a personality-driven newsletter. Without that, you're applying a 4x multiple to a subscriber list that might open a future newsletter at 30% the current rate.

What to Do Before You Look at Another Listing

Before you open the next CIM from a broker, identify the model. Not 'online business.' The specific model. Is it subscription SaaS? FBA private label? Content site with affiliate revenue? DTC ecommerce with its own Shopify storefront? Paid newsletter? Each one has three or four failure modes that are specific to the model and almost invisible in the financials.

For SaaS: monthly churn rate, CAC-to-LTV ratio, MRR growth trend, and whether the recurring revenue is contractual or habitual. Habitual recurring revenue. Customers who just haven't cancelled yet. Is not the same thing as contracted MRR. Don't let a broker tell you otherwise.

For ecommerce: return rates by SKU, supplier agreement transferability, single-ASIN or single-channel concentration, and whether the ad spend in the TTM is sustainable or was juiced for the sale. A seller who doubled their Google Shopping budget in the six months before listing is not showing you a repeatable SDE. That's a common move and it's worth reading twice in the financials.

For content sites: Search Console access is non-negotiable before an LOI. Traffic source breakdown, algorithm update correlation, page-level concentration, and backlink profile age and quality. If a seller won't provide Search Console before LOI, that's the answer.

Build your diligence list around the model's specific failure modes, not a generic acquisition checklist. If you want a starting point for that, DealScorer has a free acquisition checklist organized by deal type. Use it as a floor, not a ceiling. The generic version will get you to second base. The model-specific version is how you avoid paying $800K for a business that's about to have a very bad quarter.

The SDE times multiple formula is correct as a structure. It fails when the buyer doesn't understand what the multiple is actually pricing. Figure out the model first. Then decide if the multiple is telling the truth.