Acquisition channels are places where you source customers from. Think Facebook Ads, SEO, cold email, referrals, App Store and podcasts. And many more. In fact, any place where potential customers can notice you for the very first time qualifies as a potential channel.
Strategically, acquisition channels fall into three main categories.
Paid channels include influencer marketing, sponsorships and all advertising platforms. They're "paid" because your returns scale as a linear function of your spend. For example, you pay per click, per impression, per conversion, per referred sale. This makes paid channels predictable: if $1000 gets you 20 customers, $2000 should get you about 40. The flip side is that paid growth can only ever be linear. What you get out is proportional to what you put in:
- # Customers = Total Spend / Customer Acquisition Cost (CAC)
Customer Acquisition Cost
CAC is the price you pay a channel per customer it brings you. For example, if $1000 gets you 20 customers, CAC is $50. CAC is a function of funnel performance from channel impression to on-site purchase. For example:
- You pay a channel $2.5 every time a prospect clicks from ad to site.
- 2% of website visitors buys your product.
- You acquire a customer per $125 spent. That’s your CAC for the channel.
A paid channel is viable if the customers it brings you pay you (significantly) more than you pay the channel. How much more? Enough to cover costs and still earn a profit.
What can you afford?
Consider Affordable CAC: how much you can afford to spend to profitably acquire a customer.
- You sell coffee subscription boxes at $20 per month.
- On average, customers keep their subscription for 10 months. Average Revenue Per Customer (ARPC) is $200.
- It costs you $7 to source and deliver one box.
- Fixed employee and office costs add another $3 per box.
- 10 * ($20 - $7 - $3) = $100
- → You can spend up to $100 CAC without making a loss.
Put differently, the $125 CAC sample channel is too expensive to acquire customers for our coffee subscription boxes.
Discounting channel cost
To make a channel (more) affordable, you can (a) drive down CAC by optimising your funnel, and/or b) increase Affordable CAC by generating more revenue per customer.
- Funnel optimisation — Cost-per-click goes down as your engagement rate goes up. Say better ads and targeting bring down CPC from $2.5 to $1.5. At 2% site conversion, you now pay $75 CAC instead of $125.
- Funnel optimisation — If you improve landing page conversion from 2 to 3% conversion, channel CAC - at $2.5 CPC - drops from $125 to ~$83.
- Increase revenue — You increase the variety of coffee so customers stay subscribed for 12 months instead of 10. You now earn $240 ARPC, which grows Affordable CAC from $100 to $120.
- Increase revenue — You introduce a new subscription with rare coffee blends at $30 per month. As 25% of customers picks it over the $20 box, ARPC grows to $225 (10 months retention). Affordable CAC grows to $125.
Unfortunately, it also works the other way around.
Rising click costs caused by saturation (people getting tired of your ads) or increased competition can turn a channel unprofitable seemingly overnight. Still, paid channels come with more control than others — and they're easy to get going. Give or take, it takes ~$2000 ad spend for CAC to stabilise. From there, you can scale with confidence by reinvesting the set amount of revenue into more ads. In contrast, it’s common to not see any results from organic channels like content marketing in the first six months.
Paid channels mainly differ in how you target potential customers:
- Behaviour-targeting — Your ads reach people through actions that signal interest in your product. This how you show ads on Google and Amazon searches. If you sell pizza ovens, you're going to want to reach people that search for "pizza oven". Other examples of behavioural targeting are Pinterest (searches and pins) and Quora (searches and questions).
- Profile-targeting — Your ads reach people based on characteristics your ideal customers supposedly share. Facebook and just about all other ad channels work this way. For example, if you sell pizza ovens you'd target people aged 30-55 interested in italian food, home cooking, and maybe a barbecue brand like Weber. If you sponsor a YouTube channel, you'll assume the majority of its audience is a fit for your product.
Profile targeting comes with saturation risks. At some point your ads will start tiring audiences. In contrast, behaviour targeting matches your ads to real-time market demand. Timing is always on point.
When site visitors don't convert, you can re-target them with new ads. This works well because these people already know you. Their engagement validates interest.
Retargeting completes the trinity of paid acquisition:
- Behaviour-targeting connects you to ready buyers.
- Profile-targeting gets you on the radar of potential buyers.
- Re-targeting keeps pitching prospects you know are on the edge.
For example, I was unaware wood-fired pizza ovens sold for as little as $299 and so never Googled it. When the ad hit me on Instagram, I bought one on the spot. Had I clicked without buying, a retargeted ad would have closed my deal later.
High-level, paid acquisition success hinges on:
- Profit margin — The more you make per sale, the easier to make ads work. And the faster you can scale.
- Payback cycle — The quicker customers pay you cash, the quicker you can reinvest.
E-commerce is a natural fit: near-to-immediate purchases generate the cash to quickly turn paid loops self-sustaining. Subscription services typically demand more strategic diligence as it takes longer for revenue to pay off acquisition costs - especially when you’re a startup and customer lifetime revenue is still unclear. In fact, surprisingly few companies can rely on paid channels alone to turn profit.
There’s a bigger picture. Because they’re so easy to turn on and off, paid channels are ideal for:
- Fuelling funnel experiments — Buy sample audiences to test landing pages, checkout sequences, onboarding flows and other conversion rate optimisations.
- Seeding owned channels — Content runs on readers. Organic social runs on fans. Virality runs on users. All can be bought with cash at Facebook, Google & friends.
This makes an argument for burning cash on paid channels to accelerate owned channels (discussed next). These are naturally slow to get going, but don’t cost you per sale - making them much cheaper long-term. All the while supplying the volume necessary for statistically significant A/B tests. You temporarily overpay so to quicker graduate from paid to owned.
Paid channels come in many shades and colours. Subscribe to get notified of forthcoming material on paid channel strategy and success.
Organic channels revolve around content and community. Examples include blogging (SEO), newsletters, closed communities, shareable videos, and organic social media. They're unpaid: you invest in assets you own rather than rent assets owned by a third party, like Facebook. Compared to paid channels, organic channels grow super-linearly (exponentially): you don’t pay more as returns scale.
- A blog post costs you today, but can still attract customers via Google Search three years from now. In fact, if it rises to rank on page 1, it can get you more than ten times as many.
- The more followers on Instagram and Twitter, the bigger the reach of your posts.
- The more members in a Discord community, the bigger the potential for value-sharing, engagement and impromptu recruiting.
- Your owned audience grows every time your newsletter is forwarded.
Organic channels come with barriers: it takes time, quality and persistent hard work for returns to start compounding. Consider how about 85% of all Google searchers click on results 1 to 8. Instead of paying Facebook for access to its 2.5 billion audience, you build your own from scratch. Once across the tipping point however, the same barriers work in your favour by protecting you from competition. Organic gains can’t be copied.
Content ⇄ Community
Content is organic's atomic particle. It connects your brand to customers on their terms, subtly educating them on why they should buy your product through something that matters to them.
Over time, content cheapens marketing costs as a function of quality. Quality disarms visitors on their guard against being sold to. It convinces to read (watch, listen) closely, cultivating casual attention into long-term connection. As prospects become subscribers, cost per impression slopes down to zero. Your brand morphes into a friend they know, like, trust — and would buy from.
Outwardly, the channel gains gravitational pull in two ways, mutually reinforcing:
- Word-of-mouth — Fans tell their friends.
- Algorithmic distribution —The higher ratios of quality to quantity, the more preference content gains with algorithms. You rank higher in search and get more real estate in news feeds.
Note that content impacts growth beyond acquisition. Even if it fails to acquire new ones, content can still educate and engage existing customers - driving retention and word-of-mouth.
Community is the organic end-game. It happens when content extends your brand into a culture, an implicit context for people to naturally engage. Distribution becomes decentralised: you get talked about (almost) regardless of what you put out. Network effects kick in: the bigger the community, the harder for outsiders to ignore and the harder for insiders to leave. Every node attracts and retains others. Growth marketing on steroids.
For acquisition, content strategies intersect two dimensions:
1. Sourcing — How content is created.
- Company — Content is created in-house, as if you were a publication. Scale/volume is a function of the internal resources allocated to it, e.g. employees and contractors.
- Community — Content created by users also goes by the name of user-generated content (UGC). This expands the potential for scale (communities are bigger than company teams), but comes in different leagues. For most, UGC is about hashtags and reviews. UGC champs have products designed for users to create content with. Examples:
- ~Eventbrite — Users create events.
- ~Reddit — Users create posts.
- ~Kickstarter — Users create crowdfunding projects.
- ~Typeform — Users create surveys.
- ~GoPro — Users create videos.
2. Distribution — How content reaches audiences.
- Search — Content reaches users as search results (e.g. Google, YouTube). Content reach is a function of keyword search volume and quality. Optimising content to rank higher in search is known as Search Engine Optimisation (SEO).
- Share — Content is shared with users via open (Twitter) and dark (WhatsApp) social networks. Content reach is a function of viral coefficient: # shares * (# views / share) - discussed in detail below.
Sourcing and distribution combine into four prototypical strategies:
1. Company + Search (SEO)
Company team creates content to rank highly for keywords that topically spill over into product value propositions. The end game is to be trusted as the ultimate authority in the space your product creates value in. Examples of high performers include Headspace (mental health), Runners' World (running) Square (small business owners), Ahrefs (SEO) and Lattice (people management).
2. Community + Search (SEO)
Rather than in-house, content is created by the company’s users - then ranked by search engines as search results. Typically high, because of larger scale. This is why sites like Quora, Pinterest, Deliveroo and Airbnb dominate Google page one for searches in their space.
3. Company + Sharing
While virality is hard to engineer, content is more likely to be shared when it presents a fresh take, stirs emotion and/or makes the user feel understood. This typically requires originality and opinion as well as an authentic voice - none of which fit well with ticking SEO boxes. Some best practices:
- Be the primary source. If you don’t have the resources for original research, leverage company data nobody else has. Spotify Wrapped and Pornhub's viewership reports provide wonderful examples.
- Nail the first impression. Sharing is preceded by the battle for eyeballs in newsfeeds. Your headline/intro/image gets about 5 seconds to lock in.
- Novelty hooks. Subvert common topics with fresh takes. Controversy gets clicks but might boomerang back if unfounded. Mr. Beast is the gold standard of surprising optimism.
- Leverage visuals. Our brains process images about 15x faster than text. Infographics are extraordinarily shareable because they minimise time-to-value (Digit and CourseHero show how it's done).
4. Community + Sharing
Users create content to share with others, who then discover your brand in the process. Note, for example, how often cat food subscription service Cat Person gets tagged on Instagram. Real UGC God Mode is when viral content creation is the product’s core function. TikTok, Twitch, Eventbrite, Typeform and Instagram itself are well-known examples. MyHeritage’s side-project Deep Nostalgia proves non-tech unicorns can pull this off too.
Content is a long-term game. While quality sets the pace, it inevitably takes time for sharing and SEO to compound. That said, don't sit on your hands waiting for network effects to switch on. Zero multiplied by whatever is still zero. To accelerate returns, build distribution around your content from the get-go:
- Get existing customers to read. Link from your site, share via email and structure in knowledge bases. It helps them get more value out of the product and trains them as evangelists.
- Partner with blogs, newsletters and influencers on the watchlist of your target audience. Arrange links, features, shares and guest blogs. Get it done by returning the favour or creating the content for them to share. Pay if it's worth it.
- Leverage your best content to build organic social channels. Resist the reflex to just share links and, instead, refashion the content for in-feed engagement. Turn blog posts into threads (Twitter) and slide decks (LinkedIn, Instagram). Rework videos and podcasts for native upload. True, this won't get you site traffic, but both platforms and users hate clicking away. Optimise for maximum value on the spot.
- Seed with ads. Paid channels are shortcuts to scale. Use them to test content on audiences and, once you validate content-audience fit, buy readers/viewers/listeners in droves.
- Leakproof your audience funnel to maximise long-term ROI. Iteratively design a content journey from discovery to subscription (e.g. newsletter) and sharing that performs well before opening the flood gates. Just another way to hammer home the importance of quality. Low quality multiplied by high volume turn network effects against you.
What about SEO and social media?
This post only scratches the strategic surface of organic channels. Stay tuned for tactical content on topics like SEO, newsletters, and organic social media.
We all trust friends over ads and brands. Getting word-of-mouth is straightforward: build a product people can't stop talking about.
Once up and running, you can structurally accelerate word-of-mouth by rewarding customers for referrals. Referral programs are sort of paid virality: you pay per referral.
This kind of artificial virality thrives when:
- The product already has word-of-mouth. Turns out you can't just pay people into referring. People only tell friends about products they think will put them in good light.
- Referrals are effortless. Inviting should feel as easy as talking. Time-to-value for the newcomer near immediate.
- The new customer also gets rewarded. This makes the referrer feel altruistic and the newcomer privileged.
- Rewards are not (purely) financial, but encourage further product use.
Some examples that make the last point:
- Dropbox pioneered the modern referral program. It originally rewarded both sides 500MB of extra storage.
- Get a friend to order Uber Eats through your link and you'll both get some free food.
- Tesla grants both sides with free charging miles.
Software products have potential for organic virality, also referred to as product-led growth. This is when the product's user experience naturally invites new users into it. For example:
- You can't do a Zoom call al by yourself. You invite someone to it.
- To receive money somebody sent you with Paypal, you need a Paypal account.
- Calendly lets you embed available meeting slots in emails. Recipients use Calendly when they book.
- The visual prototyping tool Figma is most useful when you collaborate in real-time with team members.
- When you write an answer on Quora, it can reach new users as a Google Search result.
- Intercom's live chat widget signals Powered by Intercom to its users' site visitors.
In summary, you can engineer organic virality by making users collaborate and create shareable content. User-generated content (mentioned earlier) is an organic channel that scales virally.
Virality is measured as a coefficient: the ratio of referred customers per customer. For example, if you get 1 new customer per every 4 existing ones, your viral coefficient is 0.25. You can break this number further down into:
- Number of invites per existing customer. Example: The average Notion user shares pages with 5 non-users.
- Conversion rate per invite. Example: 1 out of 25 non-users becomes a user upon discovering Notion through a shared page.
- Viral coefficient = 5 * 0.04 = 0.2 → Every 5 customers bring in 1 extra.
Like organic channels, virality scales super-linearly so acquisition becomes cheaper over time. Every user acquired through virality in turn can bring in new users.
What about sales?
You need sales people if your product isn't something customers easily buy on the Internet.
This generally excludes:
- E-commerce — Customers browse your store, add products to cart, fill out shipping details and pay — all on their own. Examples: coffee, barbecues, supplements.
- B2C software — Customers find, use and pay for your app autonomously. Ads, content or referral gets them, landing page excites them, onboarding hooks them and the product experience keeps them. It's all self-service. Examples: Netflix, Airbnb, Headspace, Uber.
Sales logically associates with B2B. Of course, plenty of people pay for Zoom, Slack and Google Docs without ever talking to a sales person. Sales comes into the picture when 1-1 communication between buyer and seller significantly helps the purchase process.
- The product is complex/expensive to a point where the buyer needs a personalised explanation to properly evaluate the investment.
- The product needs input from the buyer to materialise. Think agencies, architects, kitchens, cars and custom enterprise software. Anything that is tailor-made.
Not every business can afford sales. The manual cycle of lead generation, qualification, pitching, negotiating and closing is costly. It takes serious profit margins for it to make sense. The bigger the companies you sell to, the more you'll rely on sales for acquisition.
That doesn't mean you get to skip ads, content and referrals. Rather, you work channels to generate leads for sales people to close.
- You draw the attention of an ideal customer (prospect) on a channel.
- The prospect clicks to your site and learns about your product.
- The prospect signals interest to buy (e.g. request an offer or demo) and becomes a sales lead.
Leads coming your way are inbound. Outbound means you take the initiative with cold calls, emails, LinkedIn DMs or at networking events.
You don't pick the channel, the channel picks you
This post laid out the playing field. So, where do you play?
Ignore the mantra that says to "relentlessly test every channel." It's bad advice.
Whatever your product and business model, by design it will fit some channels and rule out others. This is fine because, contrary to popular belief, you usually only need one core channel to do the heavy lifting.
Strategically, the more a customer pays you, the more you can spend on acquisition.
From this it logically follows that social apps with minimal revenue per user (e.g. Facebook and WhatsApp) rely on organic virality to cheaply scale growth, whereas companies like Palantir pay sales people to negotiate big contracts.
In-between, companies with shorter pay-back cycles — how quickly a new customer returns the acquisition cost — use paid channels to fast-track revenue so they can reinvest in more ads, kick-starting a self-perpetuating loop. This naturally fits e-commerce. Subscription-based software (aka SaaS) products typically leverage paid channels to accelerate growth of unpaid channels, acquiring readers (SEO), followers (community) and/or users (virality). As we've seen, paid buys time to build one's own channels. Which are more expensive starting out but get cheaper over time.
A few examples help sharpen intuition.
- Clubhouse — Organic Virality.
- Discord — Organic Virality.
- Revolut — Organic Virality amplified with Paid.
- Tripadvisor — User-Generated Content (SEO).
- Kickstarter — User-Generated Content (Viral).
- Booking.com — User-Generated Content (SEO) amplified with Paid.
- Headspace — Paid.
- Adidas — Paid.
- Warby Parker — Paid.
- Dollar Shave Club — Paid multiplied by Artificial Virality.
- Patagonia — Word-Of-Mouth amplified with Paid.
- Eight Sleep — Paid multiplied by Word-Of-Mouth.
- Tesla — Virality-led Sales.
- Canva — Paid.
- Shopify — Organic Virality amplified with Paid.
- Eventbrite — User-Generated Content (Viral).
- Ahrefs — Content (SEO).
- Slack — Organic Virality + Sales.
- Notion — Organic Virality amplified with Paid.
- Intercom — Organic Virality + Sales.
- Hubspot — Content (SEO) + Sales.
- Salesforce — Sales.
Setting up new channels is costly: you need copy, creative, tracking and systems. If you take a shotgun approach to testing channels, most work will end up as sunk cost. Skip trial and error by matching how customers buy from you with patterns proven by success of others like you. There's no sense in testing what you can work out on paper.
Work out your channel mix
Forthcoming is a deep-dive on acquisition strategy. It'll help you narrow the field so you can exclude ill-timed bets and run high-impact experiments. Subscribe to get notified on release.