There are an overwhelming number of channels out there. To systemize and simplify the process, we’ve developed an evaluation framework based on the following criteria:

1. Scale
2. Target-ability
3. Intent
4. Context
5. Effort
6. Cost
7. Time to results

In this lesson, we’ll walk through this evaluation framework in more depth by looking at each of the seven criteria. Channels vary in effectiveness based on a company’s audience. Channels may even vary in effectiveness for a single company targeting multiple audiences. We’ll provide examples so you have all of the context you need to pursue your channels.

Since it’s difficult to understand these criteria from a purely conceptual perspective, we’ll look at how three companies with different business models might evaluate acquisition channels:

• Otter.ai, a B2B company with software that records and transcribes online meetings
• Rareform, an ecom shop selling purses, bags, and accessories from recycled billboards
• DoorDash, a food delivery app

Scalability (Reach × Ramp)

When evaluating channels, it’s tempting to assume that a channel with millions of users should be a high-scale channel. What really matters is whether you can reach a high volume of your target market. But volume is only half the battle. Here’s what we look at when evaluating channels for scalability:

• Reach: how much of your target market you can actually reach on the channel
• Ramp: how far and how fast you can increase investment before the channel plateaus (due to unit economics breaking, operational limits kicking in, or audience saturation)

The higher a channel scores in both categories, the better.

Reach x Ramp examples

Amazon ads.

Amazon has massive reach. And there’s no shortage of brands that have built serious businesses off of Amazon as their sole acquisition channel. But that doesn’t mean it’s a high-scale channel by default. What if you’re a high-end luxury brand and your target market doesn’t shop on Amazon (or, at least not for your product category)? For you, Amazon is a low-reach channel.

Or, let’s say literally every single person in your market does shop on Amazon for your product category. Fantastic reach in this scenario. But your brand just launched, and you’re competing for attention against household names. To make matters worse, you only carry one product. Your average order value and annual customer value will be a fraction of that of your competitors with big product catalogues. Meaning they can afford to pay significantly more to acquire customers. i.e. their CAC is higher, and they have stronger model-channel fit. Suddenly, Amazon isn’t looking like the high-scale channel we initially thought because Ramp will be limited. We may only be able to capture a tiny percentage of the total available Reach until our CAC becomes unsustainable.

Snapchat.

Most users are 18–29. That tells you something about Reach by segment.

• For Otter.ai (remote work software), Reach is moderate at best. The overlap with decision-making professionals is small, so even perfect targeting would not unlock the audience they need.
• For Rareform (fashion) or DoorDash (food delivery), reach is significantly better because Gen Z and young millennials are in the target. Ramp then depends on whether they can capture more and more Reach without their unit economics breaking.

SEO.

Search looks big in the abstract, but it is only scalable if there are enough relevant queries.

• DoorDash benefits from approximately 100 million food-related searches. Reach is massive. If they can develop an SEO strategy that enables them to efficiently and economically produce a massive amount of content to capitalize on the Reach, then their Ramp would also be very high.
• Rareform faces a niche query set for recycled-material bags. Reach is lower even if a high volume of broad “bags” queries exists, right? Not necessarily. Search volume for their product may be relatively low compared to other bag-related terms, but the existing volume makes up a significant portion of Rareform’s target market. Reach is actually quite high. And if they’re able to rapidly climb the search ranks due to (presumably) lower competition, Ramp would also be high.

How to use this in your selection

• Start by estimating Reach for your ICP, not the channel’s total user base.
• Then pressure test Ramp. This is more complex. Paid channels can be scaled extremely quickly. In theory, you could go from spending $100/day to $100,000/day with the click of a few buttons. But at what CAC, payback, etc. could you afford to do so? Virality is highly scalable in the sense that your users are generating more users for you for free. The ceiling is almost nonexistent, but what about velocity? In order for users to generate a lot of other users, you need a lot of users to begin with. Where does that initial user base come from?

Targetability

Targetability describes the degree to which you can target a specific audience on a channel. Millions of your target users may use a channel, but if that channel doesn’t enable you to hone in on that specific segment (low targetability), it would underperform compared to a channel that offers such targeting.

Some channels enable you to target users by selecting a specific data point. For example, with Facebook and Instagram ads, you can make your ads only show to people of one gender, in a specific age range, and so on.

However, not all channels have this level of targetability. And some have a range of targetability within them. Think about all the ways you can reach people with Google:

• Google PMax ad campaigns are pretty much fully automated when it comes to targeting.
• Google search ads rely mostly on keyword targeting; your ads appear for keywords that you specify in Google Ads.
• Creating high-ranking SEO content also relies on keyword targeting, but without the precision available to us with Google Ads. Instead, you create content around these keywords to appeal to specific audiences (like parenting articles for new moms and dads).

Sales and digital paid acquisition channels tend to have better targetability than other acquisition lanes. Sales teams can focus on reaching out to people that match their exact customer criteria, and ad networks like Facebook let you pinpoint audience segments based on user data. By comparison, content and viral acquisition channels like content marketing and virality give you very little control over who your customers share your product with, or who will read your content.

In general, channels with high targetability make it easier to reach very specific audiences—a priority for companies selling niche products. For companies with larger markets, like DoorDash, targetability may not be as big of a priority when evaluating different acquisition channels.

That said, even with high targetability, channel efficacy varies based on who your audience is and the specific channel being used.

Consider Facebook ads and LinkedIn ads. As paid acquisition channels, they offer greater targetability than content and viral acquisition methods, but their scope of targeting differs.

• With Facebook ads, you can target users based on job title. However, since people use Facebook casually, many don’t actually add their employer or company info to their profile, making this targeting inaccurate.
• Because of LinkedIn’s purpose as a job platform, people reliably update work details on their profile. As a result, businesses that want to target based on profession (e.g., B2B companies like Otter.ai) will find LinkedIn ads more effective than Facebook ads at reaching their ideal audience.

The point here: Since targeting options can vary even across high-targetability channels, you must evaluate each channel’s targetability in relation to your specific market.

Intent

Your target audience may be active on a particular channel (high scale) that also allows very granular targeting (high targetability). But if their intent in using a channel doesn’t align with your product, they’re less likely to become a customer.

People browse channels with different goals—we call this intent. For instance, people tend to use Facebook for updates about their friends but use LinkedIn for professional networking.

Some channels aren’t as effective for acquiring customers because users aren’t actively interested in shopping on it. This ultimately depends on who your target audience is and what you’re selling.

Consider Pinterest. According to Pinterest Trends, some of the most popular keywords searched revolve around outfit ideas, recipes, and home DIY projects. Users often browse Pinterest with the goal of sourcing style ideas and inspiration.

Given this intent, running Pinterest ads doesn’t make sense for Otter.ai or DoorDash. But for Rareform, the intent behind using Pinterest aligns. Why?

• Otter.ai: Users don’t usually browse Pinterest looking for ways to improve their remote work communication. They’d more likely search on Google for answers to specific questions like “how to record Google Meet.”
• DoorDash: Even though food is one of Pinterest’s top-searched categories, users are generally looking for specific recipes or cooking tips. They don’t scroll on Pinterest when they’re hungry and want to order some food.
• Rareform: Since users explore Pinterest for fashion and style inspiration, Rareform could create ads targeting keywords about eco-friendly fashion.

It’s worth noting, however, that intent can also vary within a single channel. SEO and search ads are a good example, as not all search queries have high purchase intent.

Someone searching for “best eco-friendly purses” probably has a higher chance of becoming a Rareform customer (greater purchase intent) than someone Googling “why is sustainable apparel important?” In this case, searching for information about sustainable apparel reflects an intent to learn, rather than to buy.

Context fit

We use context to describe what users are doing on a particular channel or what they expect to get out of it. When a channel is a high-context fit, it means that your brand aligns with the channel’s culture and/or your product solves a problem relevant to the reason people spend time on the channel in the first place.

Context is important not just for judging different channels, but also for evaluating different strategies within a single channel. It includes factors like:

• Device: Some channels are more commonly viewed on mobile than desktop and vice versa.
  • Example: Since people typically browse Facebook and Instagram on their phones, it’s a lot more natural to advertise a mobile app like DoorDash on these platforms. Using a phone, you can get to the app store and download DoorDash more quickly and easily than if you saw an ad on desktop.
• Culture: Some channels have communities with very defined values and expectations, like Reddit. Or their users lean in a certain direction religiously, politically, etc.
  • Example: Rareform’s brand focuses on sustainability. With an influencer marketing strategy, fashion bloggers that thrift clothing and care about the environment would be a better contextual fit than influencers that clearly don’t value sustainability, e.g., luxury brand fashion bloggers.
• Relevance: Some channels cover specific topics, so if your product or industry doesn’t fit, audiences may find your content random or out of place.
  • Example: You can pay for a sponsored post on websites like BuzzFeed and Fast Company. Since BuzzFeed publishes pop culture and entertainment content, a sponsored post there wouldn’t be a great contextual fit for a B2B company like Otter.ai. Fast Company is more relevant since it publishes content about technology and productivity for business leaders.

Note: Some components of the eval framework are largely baked into the channel (like Cost). Context fit, however, is unique to every company due to its specific Brand and Product.

Effort

You can think of effort as the amount of time and expertise needed for a channel to succeed. There are two dimensions to this: setup and maintenance.

Some channels require more initial effort in terms of setup, with compounding results that require less maintenance later on.

For example, user-generated content strategies often involve more effort in creating social media campaigns and other marketing collateral to encourage user participation. But as these strategies gain momentum, companies can repurpose user-created content rather than producing their own.

Other channels require consistent effort all around, like viral content marketing strategies. Since creating viral content can’t be automated, companies must continually work on ideating, creating, and sharing new pieces of content to achieve virality.

Cost

As we’ve discussed, Cost varies significantly across the four major acquisition motions. Virality at the lowest end of the spectrum, followed by paid, content, and then sales at the high end.

But like effort, the cost of a channel is shaped by two factors: setup and maintenance. Some channels are costly upfront but after setup, require little (or at least less) in the way of maintenance.

Take SEO content. It typically requires a major upfront investment. It’s not cheap to create hundreds, if not thousands, of pieces of content and build domain authority. But eventually (note that this is not universal across all companies) the focus may shift from new content production (more costly) to optimization and maintenance of the content library (less costly). Virality is similar. Significant upfront investment to build the initial user base to get the flywheel turning, followed by reduced costs once the engine is working at full speed.

Other channels have ongoing maintenance costs in addition to setup. Paid acquisition requires constant monetary investment in addition to more creative production, operational support, etc. as you scale.

Time to results

In an ideal world, whatever growth channel you pick would convert users into customers immediately upon activating it. It would have a short time-to-results (TTR).

It’s worth noting that there’s a difference between time to any results vs time to meaningful results. Paid acquisition can have a short TTR across the board. All forms of content typically take a long time to produce meaningful results. SEO-based strategies will also take a moderate to long time to produce any form of results. Whereas founder-led social media could at least produce some result in a matter of hours, but has a much longer ramp-up towards producing meaningful results.

But we also need to ensure we’re not just factoring in the TTR once a channel is built and activated. The time it takes to stand up a channel is a key factor as well. Most paid channels can be activated in a matter of days. Motions that are fueled by human capital (content and sales) can take months to activate if hiring is required.

Now, putting all of this together, the primary metric for most startups will be total time to meaningful results (accounting for the time to activate a channel as well as the time it takes to cross a meaningful results threshold post-activation). With “meaningful” of course being relative and informed by your unique goals and constraints.