6 Types of Online Funnels – And When to Use Each

Since starting Closeable, I’ve explored hundreds of digital funnels, from optimizing partner flows to dissecting top DTC brands. Patterns emerged. But when brainstorming funnel experiments with a friend recently, I realized we didn’t have a common language for the different models at hand, so I decided to take my own inventory.

I’ll do so here, with my own takes and mixed metaphors along the way. We’ll cover

• 6 common funnel archetypes
• Their variations and strengths
• Who they’re best for

The Funnel Types

Funnels range from self-serve (retail-like) to brand-driven (you’re on rails). The right one depends on your product’s complexity, regulation, and how much assistance your buyers need.

1. Off-the-Rack

Use when: Your product is simple, low-friction, and familiar.

Examples: Apparel, gadgets, low-cost supplements

No quiz, no fluff. Customers know what they want. Let them buy.

2. Personal Shopper

Use when: Your product has options, and a wrong choice kills conversion or retention.

Examples: Meal kits, pet food, haircare

The quiz adds friction  but also adds conversion power: It combats overwhelm, collects data for product development, and captures leads for retargeting.

I really like Huel’s “Which Huel is right for me” as an example of doing this well – it quickly simplifies the decisions for a product line most people wouldn’t research deeply

Two final quiz thoughts – 

1. You don’t need to have a landing page first. In fact, once you know this is the best route for your customers, your next best test could be just to land them straight on the form.
2. If friction is coming, don’t be afraid of it early. The higher the cost of your product, and the more likely customers will exit at checkout, the more upside there is to (a) collecting lead information and (b) making email and phone required. 

Okay! One (real one) down. Moving on…

3. Personal Shopper + Fastlane

Use when: Some users need guidance. Others just want to buy.

Now we’re really mixing retail and roadway metaphors! This works for the same types of products as above, but just lets power users skip the quiz. They know what they want, and your gimmicky quiz annoys them, so just let them choose and check out.

I think Thesis nails this, letting users jump ahead at the quiz’s start. Keeps things efficient without losing the assistance for folks who need it.

4. Checkpoint

Use when: Not everyone qualifies, so vet them early.

Examples: Telehealth, banking, insurance

If you’re asking questions like age, medical history, or employment status before a purchase is even possible – this is your lane, and you’re probably doing this already by necessity.

The key question becomes when to collect account information, and what account information to collect. 

In reality, the results may surprise you,  so test both if you’ve got the traffic to do so. But typically, if you have a short qualification form (a handful of easy questions) you’ll collect account info at the end, or in the purchase itself. If your form is longer, or customers won’t have all the info upfront, then you’ll benefit from getting account info early, as you’re going to need it anyway.

Take Hims hair-loss onboarding as an example. They’ll verify hard requirements for their products (State, Age), make sure this treatment will be a fit. But at the same time, they collect data that will make for more compelling personalization – e.g. social proof, sharing the science – and improve retargeting  if you bounce.

5. Checkpoint + More

Use when: You need more detail, but only after commitment.

Examples: Loans, insurance, legal

You get enough info to qualify someone, then you ask for the deeper stuff after getting some commitment, whether that's a purchase or just an account creation. This is what we did at my last company with Fundera’s loan application process. Users start with a short form and then move on to more detailed steps only after they’re qualified and their experience is customized accordingly

6. Roadblock

Use when: You need to intervene with human help before someone can purchase.

Examples: High-ticket services, business software, real estate

Where a checkpoint slows you down,  a roadblock stops you. But (if I can torture this metaphor even more) – its for your own safety!

The same logic applies here. If letting customers move forward on their own is not possible, or does more harm than good, put up a roadblock. This is usually because your product  requires bespoke customization, individual pricing, or human fulfillment.

Look at Upnest, a Real Estate Agent matchmaking service. Once you’re connected with a Realtor, working with them happens in the real world. But they share information – proposals, rates, a personalized greeting – to prime the customer to be ready for that next (offline) step.

Better.com exemplifies a variation where you can still let customers get a little further – customers can provide extra information to accelerate the approval process – even though a human will still step in. This variation is useful for products that require deep customization but where some details can speed things along. 

If you’re thinking “so this is just for lead generation or marketplace businesses” that's not necessarily true. We’ve also supported partners with this flow where the onboarding process is too complex/personal to be handled through an online form, or where they’re a traditionally offline business now marketing online, so haven’t yet built this capability. The key is that you need to have the infrastructure (sales agents or onboarding team) toa support that transition in the experience smoothly. 

Choosing the right funnel

Ask yourself two big questions

1. Can anyone buy this, or do you need to screen them first?

Yes? Start with Off the Rack or Personal Shopper

No? You’ll need a Checkpoint or Roadblock

2. Can your average customer confidently buy without help?

Yes? Keep it lean, simple Off the Rack storefront

No? Use a quiz as their Personal Shopper or a qualification form they need to complete

There’s no universal “best” funnel, but there’s a best fit for each product.

The more personalized, confusing, or regulated your product is, the more you’ll want to lean into qualification and guidance, whether that’s purely digital or with human assistance.

You can only guess what’s right upfront, so test overall models against each other, then tweak landing pages and step by step flows within each model as you monitor dropoff points.