All posts by Derek Pankaew

Lifetime Value: Understanding The Metric That Drives All Marketing

It’s the metric that drives all marketing. It determines how much you can spend, it determines how much you make. It decides whether any marketing channel is profitable. It determines your CPC, your ROI, and everything in between.

I’m talking about Lifetime Value, or LTV. It’s one of the most important metrics to understand in almost any type of marketing. It affects just about every area of your business – and yet, it’s one of the most frequently misunderstood metrics.

In this article, we’ll take a deeper look at lifetime value. We’ll start by defining LTV, then we’ll take a deeper look at how it can impact your business and your marketing.

The Basics of Lifetime Value

Lifetime value is, simply put, the amount of money a customer will spend with you over the course of their lifetime. Typically LTV is an educated guess. While companies can look back in time and analyze their historical LTV, but that doesn’t necessarily mean their LTV will the the same moving forward.

For newer companies – companies that are less than 3 years old – most LTV numbers are mostly approximations. Even so, these educated guesses can be very useful. Having a rough sense of your LTV will let you estimate how much you can spend to acquire a customer.

Understanding LTV allows you to compare the amount you’re spending to acquire a customer, against how much you expect to earn from them over the course their lifetimes. For planning marketing spend, this is far better than comparing cash out and cash in.

Cash In, Cash Out vs. Lifetime Value

Why is measuring LTV better than measuring cashflow?

Because generally when you acquire a customer, you’ll be cashflow negative. For example, let’s say you have a $100 LTV. Your revenue might come in like this:

  • Acquisition Cost: 45
  • Week 1: Free Trial
  • Month 1: $20
  • Month 2: $20
  • Month 3: $20
  • Month 4: $20
  • Month 5: $20

If you’re only looking at cash in and cash out, a $45 customer acquisition cost might look like a failure by month 2. But over the lifecycle of the customer, it’s actually a very successful campaign.

Measuring Lifetime Value

So how do you actually measure lifetime value? The formula for calculating lifetime value is fairly simple:

Transaction Value x Number of Transactions x Profit Margin

Or, for subscription businesses:

Monthly Revenue x Number of Months x Profit Margin

Although calculating LTV is relatively simple, actually measuring and gathering all the data can be rather complicated.

For one, actually linking purchases to a buyer can be tricky. People often have multiple email addresses, especially if they’re splitting orders between work and home. They might be purchasing items as gifts, or can move residences. You might also have the option to checkout as guest, in which case tying transactions back to individuals is even trickier.

There are tools you can use to help you her this data. Even Google Analytics can help provide some of this data. Unfortunately, part of the art of marketing is also working with incomplete data.

Instead of waiting until you have a perfect set of LTV data, you often have to work with what you have, discard certain segments, and make inferences. For example, you might exclude people who checkout as guest entirely from your LTV calculations. Using people who have longstanding accounts, you can baseline a rough LTV estimate, and apply that across your customer base.

What’s a “Good” Lifetime Value?

A typical rule of thumb is that you want your lifetime value to be 3 to 5 times your customer acquisition costs. In other words, your gross profit should be 3 to 5 times what it costs you to get a new customer.

This 3-5x rule of thumb gives you room to hire staff, pay for the office, phone bills, etc and still have enough margin left over to make a decent profit.

For example, a eCommerce site might calculate their LTV-to-CAC ratio like this:

  • Revenue: $50
  • Cost of Goods: $20
  • Gross Margin: $30
  • Customer Acquisition Cost: $10
  • Average # of orders per customer: 1
  • LTV to CAC Ratio: $30 to $10

There are a lot of rules of thumb with LTV. While these rules of thumbs are useful, they’ll vary significantly from business to business. A grocery store with 7% margins needs to think about LTV very differently than a software business with 70% margins.

Using Cohorts to Analyze Lifetime Value

Your customers’ lifetime value can change over time. As you improve your products, develop your brand, and test different marketing approaches, your customers’ behavior will change as well. Cohorts allows you to measure this change over time.

Let’s say you have a consistent marketing campaign running, from January to June. Although your marketing is the same, your product quality and your customer service have both improved. That means the number of transactions per customer should go up, while your returns should go down.

Instead of just tracking LTV based on customer IDs, or based on marketing campaigns, cohorts let you compare customers by groups of time. Each customer is grouped by the time they made their first purchase – January, February, etc. These groups are called cohorts.

In an ideal world, we’d be able to track the impact of every single variable – website changes, product changes, marketing changes, etc. In the real world, that’s usually not possible. Instead, often the best we can do is to measure our LTV over time, based on when customers first purchased, to make sure our numbers are going in the right direction.

Doing a cohort analysis will tell you a few things:

  • Are new customers likely to make a repeat purchase, or less likely to make a repeat purchase, than other customers in the past?
  • Are new customers spending more or less per transactions?
  • Are your products and your marketing getting better or worse over time?

How Much Can You Spend to Acquire a Customer?

The #1 reason for figuring out your LTV, is so you can figure out how much you can afford to spend to acquire a customer.

By taking the reverse of the 3-to-1 rule, we can say that typically you can spend up to 35% of your LTV to acquire a customer. If your customer LTV is $100, you can afford to spend $35 to acquire them. Again, these are based on averages, and will vary a lot by individual businesses and industries.

Payback Periods

One key thing is missing from the LTV-to-CAC calculation. The factor of time. Although a customer might spend $100 from you during their lifetime, we don’t have their whole lifetime to recuperate our investment.

A metric almost as important to LTV then is Payback Period. Payback Period is the amount of time necessary before you recuperate your customer acquisition costs. You can absolutely go out of business by spending $35 to acquire customers that earn you $1 a month for 100 months.

Venture backed or angel backed businesses tend to be comfortable operating with payback periods of up to 2 years. Most bootstrapped businesses tend to operate on 3 – 6 month payback periods.

Other Factors

Building a business isn’t always about hard dollars and cents. Especially in the early days of a company, or even in a new product launch, it can make sense to spend up to 100% of LTV to acquire customers.


Because you want to get as many people using your product as possible. You want word of mouth, brand, referrals, and network effects to start kicking in. You’ll start the new product off with more momentum, as well as gather a lot more data about what’s working and what’s not. By having more customers, you’ll also get more feedback on what’s working and what to improve.

You can’t continually run a business by spending all your profits on acquiring customers. But it can be a powerful way to get a new product off the ground.

Using LTV to Determine Your Cost Per Click

If you know how much you’re willing to spend on acquiring a customer, you easily figure out how much you can afford to pay per click (CPC) to your website. Based on that CPC, you can figure out which marketing channels you can test.

For example, let’s say you have a LTV of $200. You’re willing to spend up to $60 to acquire this customer.

Your website conversion rate is 2%. Which means it takes 50 clicks to get a conversion. Another way to look at it, is each click is worth $1.20 to you ($60 per sale / 50 clicks per sale.)

Now you know that as a rough rule of thumb you can afford to spend $1.20 per click. Marketing channels that cost substantially more than that should be avoided.

Expanding LTV = More Marketing Channels

Increasing your LTV is one of the best things you could do for your marketing. Increasing your LTV does not mean a linear increase in traffic. A 40% increase in LTV doesn’t mean a 40% increase in revenue – it could mean a 400% or even 4,000% increase in revenue.

This is because increasing your LTV lets you open up new marketing channels that you wouldn’t have been able to make profitable on the old LTV. For example, let’s say your business was profitable at $200 LTV. Then, by improving your product and marketing, you increase this to $300. Now, AdWords and podcast ads are working, doubling your revenue.

A year later your LTV is $600. Now you can afford to buy ads on TV, pre-roll video, and display ads, which drives your revenue another five times higher than before.

In other words, an increasing LTV has an exponential impact on the entire business. Not only does it increase profitability, it unlocks entirely new marketing channels.

Lifetime Value is Not Evenly Distributed

Most companies use an average LTV to calculate the amount they can spend on acquiring customers. Unfortunately, taking the “average” LTV is often not the best idea.

Using an average LTV to make marketing decisions gives the impression that LTV distribution looks something like this:

When in reality, LTV usually plays out something like this:

In truth, using an average LTV is deceptive. Customer values can vary widely, and often follow a power law distribution. 10% or 20% of your customer can account for 80% of your revenues.

The bottom line is, a small portion of your customers are often responsible for the vast majority of your revenues. This means that taking a simple average is often not the most useful number.

Using an Average vs Segmenting Customers

When you take an average of all customers and use that to calculate LTV, you’re essentially treating all customers as equal. Unfortunately, doing so often results in lower ROI and less effective marketing.

If some customers are worth $5 and others are worth $70, it doesn’t make sense to spend money evenly to acquire an “average customer” in hopes of getting more high value customers.

Instead, your marketing will be more effective if you separate customers into different buckets, and identify the behaviors and traits that highlight a high value customer.

So, how do you go about finding high value customers?

Finding the Patterns of High Value Buyers

Identifying your high value buyers, and then finding more of them, first starts with diving into the data. Ask yourself:

  • Is there a difference in demographics between our high value and low value buyers?
  • Do high value buyers tend to purchase different kinds of products?
  • Do certain traffic sources tend to bring in more high value buyers?
  • Do certain messages or marketing funnels tend to bring in higher value buyers?
  • Etc

Note: You can export a list of your high value buyers and upload it to Facebook’s Audience Insights. That’ll give you a lot more information about their demographics and interests.

Once you’ve identified the buyers that tend to spend the most money, adjust your marketing accordingly. For example:

  • Use an email list of only your highest value buyers to create a Facebook Lookalike Audience, and run Facebook Ads to them.
  • Identify the demographics of your high value buyers, and use those demographics when you’re doing your ad targeting.
  • Feature the products that tend to attract high value buyers in your ads.
  • Create referral programs to incentivize higher value buyers to refer more of their friends, who are also likely high value buyers.

Most businesses have an uneven distribution of LTV. That said,  there are exceptions. For example, gas stations and locksmiths both most likely have a fairly even distribution of customer values.

The first step in determining your approach to LTV is to figure out if your distribution of profits is even or uneven. Once you know that, you’ll know whether you should focus on tailoring your marketing to high LTV customers, or to focus on the average LTV-to-CAC ratio.

Dealing with Uncertainty

As much as marketing managers love to put definite numbers in powerpoint slides – me included – the truth is, LTV is a fuzzy number. Most companies don’t know their LTV, and the true LTV number is almost impossible to know.

Instead, we learn to use our best guesstimates to make decisions. We might not know if our LTV is $231.3 or $191, but we can estimate that it’s around $200. We can tell if certain groups are more valuable than others. And we can use that data to inform our marketing efforts.

How do you and your team think about LTV? Share your ideas in the comments below!

Derek Pankaew

When and How Much Should a Startup Invest in Branding?

The marketing strategies that startups should use are very different than those that big companies use. Startups often get in hot water by imitating big brands.

When startups see brands like Dove, Toyota, and American Express spending big bucks on fancy commercials, they often think that’s what they should do as well.

More specifically, even without copying the marketing budget, they try to copy the mindset. The underlying assumption is this: “I should spend money to get my brand out there, and to build ‘good feelings’ about my brand name.”

So they embark on a pared-down version of brand building. Instead of spending millions on TV ads, they hire a PR firm, build up their Instagram channel, and buy banner ads. A year later, they find that their marketing has resulted in few sales and many thousands of dollars in wasted cash.

Why do these strategies work for larger brands, but not for startups? The difference comes down to …


Mindshare vs. Marketshare

Most large brands invest in buying “mind share.” In other words, they’re buying brand awareness and brand affinity. When you think of a particular product category – tooth paste for example – Colgate is buying a small segment of your mind. When you think of tooth paste, you think Colgate.

But for this approach to be successful, there has to be a pre-existing method for you to turn “mindshare” into sales. Mindshare has to be turned into marketshare at some point in time.

For Colgate, that happens in supermarkets, when you’re standing in the toothpaste row considering which one to buy. The millions of dollars they invested in building that “mildly warm and trustworthy” feeling translates into someone picking up that toothpaste and making a purchase.

In other words, branding works best when there’s an existing point of purchase. Consumers are choosing which option to select. Investing in mindshare helps win that battle to be the option consumers choose.

Unfortunately, in the beginning, most startups don’t have any distribution channels. Which means that, as good as it would feel to have a stronger brand, a stronger brand won’t necessarily result in stronger sales. Startups can’t really afford to spend money on marketing that doesn’t boost sales.

Does that mean startups shouldn’t invest in branding at all? There is a certain kind of branding startups should invest in. To explain …


Let’s Deconstruct Branding into Two Parts: Identity and Media

People often mean very different things when they say the word “branding.” It can often be broken into two separate parts: brand identity and media spend.

Brand identity are things like:

  • Clarifying your core message, beliefs, and customer base.
  • Getting clear about the brand voice and style.
  • Coming up with the right name.
  • Getting the right domain name.
  • Having a stellar design, including logo and user experience.
  • Having a great website.
  • Figuring out your brand color and fonts.
  • Writing copy that makes people smile.

Brand identity is a foundational piece of a startup’s marketing. If a company doesn’t have a clear identity, it won’t “feel” consistent to customers. The message will be convoluted, and difficult to spread.

Startups should, absolutely, invest in brand identity.

Media spend, on the other hand, is what people often refer to when they think of brand. Media spend is when you’re actively pushing your brand out into the world. These include things like:

  • TV, radio, or billboard advertising
  • Developing a social media following,
  • Paying influencers to post on social media,
  • Sponsoring conferences or athletes,
  • Investing in SEO
  • Hiring a PR firm

These strategies work great for building mindshare. Unfortunately, it’s entirely possible to build mindshare and not have it result in market share. spent millions on a SuperBowl ad, and promptly went out of business a few months later.

A Startup’s #1 Marketing Goal

A new business’s first marketing goal should to find its first scalable, repeatable customer acquisition engine.

The engine might be something simple, such as a Facebook Ads to product page funnel:

Or, it could be something much more complex. For example, you might use paid spend to generate a lead, nurture the lead, then do a sales call.

You’ll often need a lot of testing before finding a funnel that works. You’ll know you’ve succeeded when you can predictably get a customer using a specific series of steps.

Those series of steps usually start off with a scalable traffic source at the top. That can be AdWords, Facebook Ads, cold email, or any number of other traffic sources. The key is that it’s predictable.

Once you have a predictable traffic source, the next challenge becomes to scale up. But getting that first one is the most challenging.

Bottom Line: New Companies Should Focus on Sales, Until …

New companies should focus on sales, and let branding be a byproduct. Remember that even as you’re generating sales, thousands of people are being exposed to your brand and product. If you’re running ads, thousands of people are already seeing your brand. Generating revenue is what will allow you to purchase more ads and drive more sales – as well as branding.

In the beginning, put most of your energy into marketing spend that generates sales. Eventually, you’ll want to invest in brand building media spend – just not in the beginning.

Focus on building a predictable and repeatable engines for generating sales. Once you have extra cashflow being generated from the business, then set aside a portion of it for brand building.

Derek Pankaew

The 18 Highest ROI Marketing Channels – and When to Use Them

There are hundreds of marketing channels to choose from. Which one’s the best for your business? Which one can bring you the highest return on your time, energy and money?

Unfortunately, there’s no one size fits all marketing plan. What brings in droves of customers for one company might be complete flop for another. The key to making marketing work is to pinpoint the best marketing channels for your product and your audience. Then, test quickly to find out what works, and scale up the channels that do the best.

With that in mind, let’s take a look at the 18 highest ROI marketing channels. For each channel, I’ll give a brief overview of how it works, as well as its pros and cons, and a description of which type(s) of businesses should or shouldn’t use it.

Cold Email Outreach

Believe it or not, cold outreach works. It’s actually extremely effective. For a detailed example, Predictable Revenue is a book that outlines how Salesforce added $100 million a year in recurring revenue using this strategy.

Here’s how it works. You use a database provider like to find prospects. You can get pretty detailed with this. For example, you can target retailers with at least $10m in revenue, or people with the title “VP of Product” in tech companies with at least 50 people.

Alternatively, you can pay someone to find email addresses for you. For example, you can scrape Yelp for all the Irish pubs in New York. One of the best tools is to use to find email addresses on a website (free tool.)

Once you have the email addresses, you can use an automated tool like to send emails out to the list of people who match your criteria. About 1% of your cold emails should turn into warm leads, contacts, and product demos.

Who Should Use This Channel?

Best For: This channel should only be used when selling to businesses. Typically this works best for businesses that are selling higher ticket products, usually with a $5,000+ lifetime customer value.

This channel only works when combined with a salesperson, so you need to take into account both the cost of both generating the lead, as well as hiring the sales team.

Not For: This channel shouldn’t be used for selling to consumers. For one, it’s illegal to email consumers without permission, because of CAN SPAM laws. It’s also not worth the time and effort to make a $30 sale.

Google AdWords

Google generates over $60 billion a year in advertising revenue for one simple reason: their ads work. They make businesses money, so more and more businesses spend money with them.

AdWords works by letting you bid on keywords. Your text ads then show up for the keywords you’re bidding on.

Who Should Use This Channel?

Best For: This channel works best for locating people with buying intent. If someone is searching for pearl jewelry, they’re probably looking to buy pearl jewelry. AdWords works for a very wide range of businesses, and most startups should at least experiment with AdWords to see if it works for them.

Not For: Products with low search volume, especially if there’s a large group of prospects that aren’t necessarily searching for that product. For example, let’s say you’re selling funny t-shirts for people who love their dogs. Every month, only about 3,000 search for “dog shirts” in Google. However, there are millions of dog lovers in the country. They might not know to search for a dog shirt, but if they saw a funny one, they might buy.

In this case, another channel that lets you target by interest (dogs) rather than by buying intent might work better. You’ll have to create buying intent rather than find buying intent, but instead of advertising to 3,000 people you can advertise to millions.

Facebook Advertising

Facebook has a better interest targeting platform than anyone else on the planet. Unlike banner ads, these ads are in locations people are trained to pay attention to: the newsfeed. This results in much higher click through and conversion rates.

Facebook’s targeting works very differently than AdWords. Instead of targeting based on keywords, you target based on interest. Instead of targeting based on what people are looking for, you target them based on what they’re interested in.

Who Should Use This Channel?

Best For: Facebook works best when you have a message or product that can appeal to an entire group of people. For example, anyone who loves cooking, or software business owners, or women who like to cycle.

Like AdWords, Facebook is a channel that most businesses should experiment with at some point.

Not For: Products that can only be purchased at a specific point in time. For example, let’s say you love to play ping pong (table tennis.) Even though you’re a diehard fan, the reality is you’re only going to buy a ping pong table once every few years. So even if a ping pong table company showed ads to you, you won’t buy – because you’d only purchase at a specific point in time. It would be better for them to bid on AdWords and wait for you to search for a ping pong table, rather than try and get you to buy one now.

Facebook Ads is also not good for brand awareness and brand lift campaigns. The ad rates are much higher than other channels like TV, and the impact on brand is traditionally lower per dollar spent than more traditional channels.

Search Engine Optimization

SEO is one of the go-to marketing channels for most business owners when they’re first getting started. But the truth is, SEO only works for a very small subset of companies.

While successful SEO can bring in a ton of free traffic, succeeding in SEO is difficult. There’s only one #1 position for any given keyword, so you’re competing with every company in the world for that particular keyword. While it’s possible to win this fight, it’s not a fight new companies should pick.

SEO should generally be tackled by established companies, once they’ve already made a few other marketing channels work.

Who Should Use This Channel?

Best For: Businesses that publish a lot of written content. Also best for businesses with other marketing channels. SEO tends to work best for businesses with substantial marketing budgets, and entrepreneurs with existing marketing experience.

Worst For: Businesses that don’t have content, or for whom writing content doesn’t make sense. For example, if you sell cheap iPhone cases, people don’t necessarily want to read about the iPhone case – they just want to buy one.

Don’t bet the company on SEO, especially in the beginning. SEO takes time and money to develop. Work on your SEO as you also build out other channels, so faster acting channels can bring in revenue while you slowly increase your rankings.

Google Display Network

Google Display Network, or GDN for short, is the world’s largest platform for purchasing display ads. Using GDN, you can place banner ads on any website that’s making money using Google AdSense.

Unlike AdWords, the targeting for GDN is based on the content of the web page. If you’re advertising home furniture, your ads will most likely show up on home furniture sites. You can use keyword targeting, interest targeting, or specific website targeting to run your ads.

Who Should Use This Channel?

Best For: This channel is best thought of as a scaling channel. Scale up to GDN once other channels are profitable, but typically Facebook Ads will convert better for display ads than GDN. Retargeting also tends to do well on GDN.

Not For: Not for testing new ad campaigns or offers. If you’re launching a new product or offer, and it doesn’t work on GDN, it’ll be difficult to say whether it’s an issue with your product or an issue with the channel.

Google Shopping Ads

Google Shopping is one of the most effective channels for selling physical products. If you’re selling a physical product, you should be on Google Shopping.

Google Shopping ads are the ad units with rows and rows of product images. These ads convert very well.

The only downside to Google Shopping is that it’s difficult to setup. You have to turn your products into a product feed, which can be quite technical. Most shopping carts – Shopify, Magento, etc. – have plugins that do most of the work. Even so, it’s still not as easy as AdWords or Facebook Ads.

Who Should Use This Channel?

Best For: If you sell physical products, especially low to mid priced, this channel should work well.

Not For: Digital products or services, or any product that’s not physical. It’s also difficult but not impossible to sell $5,000+ products via Google Shopping. Generally speaking, higher ticket items work better through other channels.

Private Lead Brokers

Lead and click brokers can be a very good source of clients, especially in established industries.

For lawyers, can be a great source of leads. For B2B software, is a popular lead gen site. For real estate agents, Trulia and Zillow are both well known lead sellers.

The downside to buying leads is that there are a lot of low quality lead sellers as well. Marketers should enter this market with skepticism.

In the beginning, it’s best to only do business with the Top 3 brokers in your industry. If you’re in real estate for example, buy leads from Trulia and Zillow first. This will allow you to benchmark their performance.

Then, if you venture outside of the Top 3, you can use your conversion and ROI metrics from the Top 3 to compare against less established lead brokers. Working with new brokers can be a great source of leads, as there’s often less competition for those leads. Just don’t start with them, as it’s entirely possible their leads won’t pan out.

Who Should Use This Channel?

Best For: industries where lead brokers consistently deliver results. A lot of industries don’t have lead brokers, so this strategy won’t work in every industry.

Worst For: Industries with no benchmarks. A real estate agent for example can use Google to figure out what percentage of their leads should turn into demos, and what percentage of demos should turn into sales. If you’re the first one in your industry using purchased leads, and you have to benchmark from scratch, it will be difficult to figure out if you’re on track.

Strategic Partnerships

One big strategic partnership can drive a massive amount of business.

A strategic partnership is a win-win deal between two companies that results in increased revenue for both parties. There’s no cookie-cutter type deal with these partnerships. Usually it involves leveraging the unique strengths behind both companies.

Here are a few examples of strategic partnerships:

  • Google lets people order Lyfts directly from Google Maps. Lyft gets to take advantage of Google Maps’ massive user base.
  • Delta gives away NatureBox snack bars for free on their flights. This gives NatureBox massive free exposure to Delta’s audience, some percentage of whom may become customers.
  • Spotify gives students free access to Hulu. Hulu, of course, is hoping that students will get hooked on Hulu, so when they leave school they’ll decide to become a paying member.
  • Postmates, a food delivery company in New York, partnered with a local pie shop to deliver free pies on Pi Day, driving exposure and press for both companies.
A NatureBox snack bar in Delta Airlines’ snack tray.

Who Should Use This Channel?

Best For: Companies with larger budgets. Often times arranging a partnership is a full time job, so you need to be able to afford a Head of Partnership role. In cities like New York, this is often a $60,000+ salary; so this strategy is often reserved for larger businesses. That said, a founder can definitely reach out and structure these kinds of win-win deals – just expect it to take a lot of time and attention.

Not For: “Unsexy” businesses. A janitorial service, for example, might have difficulty doing a strategic partnership.

Direct Mail

Yes, direct mail works, even in the 21st century. It works very well in fact – to the tune of $10 billion a year.

Direct Mail comes in many forms. Some companies send low quality prints, on newspaper-style paper, going for cheap scale. Others send high quality prints to carefully selected addresses. Both approaches work.

The key to direct mail is math. For example:

  • The cost for printing, including the paper, is $0.40
  • The price of the envelope and postage is another $0.40.
  • The cost of the address itself, from an address broker, is $0.05 each.
  • The total cost to send a letter is $0.85 each.
  • After doing a test mailing, you find that the response rate is 1 out of 100, or 1%. That means your customer acquisition cost is $85.
  • Your average customer lifetime value is $170, so you make a 50% ROI on the spend.

Who Should Use This Channel?

Best For: Products with higher lifetime value. It’s difficult to make a $3 product work via direct mail. This also works best in “major” industries, as you’ll have more address lists to work with. A good rule of thumb is the Magazine Rule: if there are at least 2 magazines in the bookstore catered to an audience you can sell to, then direct mail could work.

Not For: Niche industries that don’t have address lists you can buy. For example, life coaches probably shouldn’t use direct mail, as you can’t buy a list of people who’re looking for a life coach.

Press and PR

Press is one of the most common marketing channels businesses tend to go for. In addition to the traffic, there’s a lot of prestige that comes from being featured.

PR is a “feast or famine” channel. If you can get into one or two publications, getting into another 5-10 is quite easy. But getting the first few writeups can be challenging.

Generally I’d recommend most companies not to chase PR unless it comes to you first. Once you’re doing something that’s naturally attracting PR, then it makes sense to go out and solicit more. But spending money on PR agencies before you have a marketable story can be a big drain on cash, time, and energy.

Instead, focus on using other channels to generate your initial sales. Then, once you have enough traction that bloggers and journalists start writing about you on their own, put together a strategy to increase the flow of press.

Who Should Use This Channel?

Best For: Products that are very unique and head turning. It’s  difficult to get a newspaper to write about a new hair salon, but very easy to get them to write about the world’s first artificially intelligent teddy bear.

Not For: Businesses or products that are similar to a lot of other products on the market.

Affiliate Marketing

Affiliate marketing is like hiring an army of commissioned salespeople to sell your product.

Usually, you start by listing your product on one of the affiliate marketing exchanges. The largest ones are Pepperjam and Commission Junction.

Then, affiliate marketers who’re on the platform find your product and promote it to their audience. When they make a sale, you pay them a commission. The whole transaction is tracked by the affiliate management software.

Who Should Use This Channel?

Best For: Markets where there are a lot of existing websites with large audiences. For example, weight loss, dating, marketing, fashion. Also, this tends to work for existing businesses – affiliate marketing shouldn’t be used as your first channel.

Not for: Markets without existing audiences. For example, if you’re selling dance inspired keychains, it’s unlikely that there are many people with large dance audiences on affiliate networks. Instead, you’re better off targeting dancers on Facebook Ads.

Refer a Friend Campaigns

Refer a friend campaigns can be extremely powerful. In fact, Blue Apron, a $500 million food company, gets almost half their revenue via refer a friend campaigns.

The idea is simple: create an incentive for people to refer their friends to your business. Make sure it’s a win-win for both them and their friends, as nobody wants to feel like they’re “selling out” their friends for profit.

Who Should Use This Channel?

Best For: When you can give away something that “feels” highly valuable. For example, Dropbox lets you give your friends 16 GB of free storage space. Blue Apron lets you give friends a free meal delivered to their door.

Not For: Businesses that have a “boring” product, or have a hard time giving away free product. It’s hard to give away a free Xerox machine.

Content Marketing

Content marketing means using content to both generate trust and bring in more traffic. Content marketing can be used to increase conversion rates by building trust, as well as to increase traffic via SEO and social media. is a content marketing business.

Who Should Use This Channel?

Best For: Products or problems that people are passionate about. Home beer brewing, digital marketing, and dating tips are all things people can spend hours a day on.

Not For: Products people don’t want to think about. For example, if you sell ballpoint pens, people don’t want to think about ballpoint pens – they just want to pick up a pack of 20 at Staples.

Native Advertising

Native advertising has two definitions:

1 – Integrated Marketing

Paying for advertising on a website, in the same format that the website is in. The ad is “native” to the website experience.

For example, this article, “12 Foods You Had No Idea Have This Much Protein,” is actually an ad paid for by Boca. The article features Boca’s veggie burger as part of the article. Since the ad format fits into the site’s format, this is one style of native advertising.

Pros: Very good for branding.

Cons: Negotiating these deals is a very manual process. CPMs are typically much higher than other ad platforms as well. It’s much harder to evaluate, negotiate, and sign five $10,000 deals than to just spend $50,000 on Facebook or AdWords.

2 – Native Ad Networks

The ad carousels on the bottom of news articles are also called Native Ads. These are typically brokered through ad platforms. The three main ones are Outbrain, Taboola, and Yahoo Gemini.

Pros: Massive scale. If you can get a campaign to work, it can generate a lot of revenue.

Cons: It’s very difficult to get ROI positive. A lot of testing is required. Also, these types of ads need a multi-step funnel and generally can’t go direct to product. So it’s a lot of work to make it work. For most companies, it’s best to start on Facebook or Google, then scale to native ads once other channels are working already.

Podcast Advertising

Podcasts are like radio, but on your phone. And you can listen to it at any time.

Unlike radio, shows can be much more “laser targeted.” Instead of having a radio for talk shows or love songs, podcasts can be very niche. There are podcasts for digital entrepreneurs, for crossfit enthusiasts, and for people who love their dogs.

Podcast ads let you get in front of a very passionate group of people. And you often get to leverage the connection hosts have with the audience – because the hosts are actually reading your ads for you.

Who Should Use This Channel?

Best For: Products that have a higher lifetime value. Most products that are successful in podcast ads are relatively high LTV. These include subscriptions, such as SquareSpace, or high ticket items, like Casper.

Not For: Low priced items, and single purchase business models. For example, if you sell a $20 iPhone case, it’s very difficult to make the economics of podcast ads work. Also, products that hosts don’t want to associate with their brand can be difficult to sell, because everything has to be approved by the host first.

Social Media Marketing

Social is both a powerful marketing channel, and a very misunderstood one. Newer businesses especially often get in trouble by only relying on social.

Social is generally not a new lead generator. Instead, social is a way to build relationships with your existing audience. You’ll get some traffic from people in your audience sharing with their networks, but it’s usually not a large amount.

Most startups should initially focus on one of the other more scalable marketing channels – Facebook ads, AdWords, etc. Then use social to amplify their efforts later down the line. But “we’ll do social media” is not a viable marketing strategy by itself.

Who Should Use This Channel?

Best For: Companies that post “share worthy content”. This is content that generates emotional reactions: “wow!” “aww,” “wtf” “yum” and so on.

Not For: Companies with everyday products. It’s hard for a company selling signage to make share worthy content.

YouTube Video Ads

YouTube video ads are very cheap relatively to Facebook Ads. While Facebook’s CPMs rose by 171% in 2017, YouTube’s ad rates have stayed relatively flat. Also, YouTube doesn’t charge you for ads if people skip before 10 seconds, while Facebook charges you for every impression.

In other words, YouTube ads are cheaper than Facebook. Which means the ROI can often be higher, as long as the CTR and conversion rates are similar.

Who Should Use This Channel?

Best For: B2C products, or lower market B2B products. For most products below $500, YouTube can be a good channel. Especially if it’s visual. You can sell physical goods, software, and services through YouTube ads.

Not For: Higher ticket sales. It’s difficult to sell a $1,000+ product via YouTube ads.

YouTube ads are also not good if you need to target a very specific audience of people. YouTube’s ad targeting is much worse than Facebook’s. So if you need to target heavy machine operators in Iowa, use Facebook. If you need to people who love dogs, both platforms would work.

YouTube Sponsorships

Want a funny, interesting, and high converting video, done without you having to do any video work? Look no further than YouTube. Thousands of YouTube personalities have already built up a large audience, and are looking for additional ways to turn that audience into money.

Although YouTubers can make money from the ads Google shows on their videos, it’s a small paycheck for the amount of traffic they get. So, they’re often interested – excited even – to do sponsorships for brands.

Who Should Use This Channel?

Best For: Brands in fashion, food, health, or any other industry that has a lot of YouTube personalities.

Not For: “Dry” products, products that don’t have a strong “wow!” factor. Also, keep in mind YouTube tends to be a younger audience, so typically businesses with a younger demographic do better.

Selling on Marketplaces (Amazon, uDemy, etc.)

You don’t always have to build your own traffic from scratch. Instead, you can leverage existing marketplaces to jump start your sales.

If you’re selling a physical product, Amazon can bring in a lot of revenue. If you’re selling software, consider the Apple App store. If you’re selling a training course, listing your program with uDemy or Lynda can get you a lot of exposure.

There are marketplaces in almost every industry. These are sites with a high volume of existing traffic that you can take advantage of. Instead of having to get customers from scratch, you can instead focus on increasing your rankings within these marketplaces.

Usually this comes down to reviews, keywords and sales. Make sure to study up on how your specific marketplace works before listing. For example, Google “how to improve Etsy rankings” and study everything you can about how the ranking algorithm works before you start selling products there.

Who Should Use This Channel?

Pros: Allows you to tap into a large base of existing buyers.

Cons: Typically margins are much lower. You’ll be forced into price competition with other sellers on the marketplace. Also, you don’t own the direct customer relationship, which could cause problems for your business later down the line.

Marketplaces are usually a good jumpstart, but long term you’ll want to build your own relationships with customers.

Choosing Your Marketing Channel

Which marketing channel would work best for your business?Hopefully, this article gave you some ideas as to where to start looking.

It’s impossible to say without testing which would work best. A good approach is to pick 2-3 that you think make sense for your business, and start testing. Take the one that worked best and scale it up.

What’s your favorite marketing channel? Share your thoughts in the comments below!

Derek Pankaew

$100,000 a Year via AdWords: Marketing Lessons from the #1 Thai Language Course

Learn Thai from a White Guy has both an amazing name, and an amazing business. In just a few years, it’s gone from startup to the #1 Thai language learning course in the world. The course generates over $100,000 per year, using marketing strategies that need very little management.

I had the opportunity to sit down with Brett to learn a bit more about how he runs his business, and how he gets his customers.

The system and the funnel are fairly simple and straightforward, and extremely effective. Let’s take a look.

How to Market an Online Language Learning Class

Learn Thai from a White Guy uses several key marketing channels. We’ll explore:

  • Introduction, and SEO
  • Google AdWords
    • Google Display Network
  • Facebook Ads
  • Getting email addresses from visitors
  • Autoresponder series
  • Using a quiz to increase conversion rate
  • Marketing benchmarks and metrics

Getting Started

Before we go jump into the marketing, let’s take a quick look at the back story behind Learn Thai from a White Guy.

Brett first moved to Thailand in 2003. He started off teaching English, and quickly picked up Thai as well. By 2006, he was teaching private Thai lessons to westerners. Before long, he decided to package his classes into a digital version, and started selling them online.

He started by launching a blog, which got him to his first $500 a month. Then, once he started adding the marketing a few other marketing channels, his site quickly grew to six figures.

With that brief backstory covered, let’s take a look at how Brett was able to grow so quickly.

AdWords: The Most Effective Marketing Channel

The highest converting, highest volume, and highest ROI channel for Learn Thai From a White Guy is Google AdWords.

This makes sense: AdWords is the best way to find people with intent. In Thailand, the vast majority of foreigners aren’t interested in learning the local language. Most are just there to – well, travel or hang out.

His AdWords ads look something like this:

Improving ROI: Countries, Keywords, AdGroups

I asked Brett about the growth of his AdWords campaigns. What were they like when they were first launched started? Were they profitable right away? What did he test, and how did things improve over time?

It turns out, AdWords wasn’t profitable right out of the gate. Paid traffic seldom is. Instead, it was a process of iteration that gradually grew more and more profitable over time. From Day 1 it was clear that it would generate revenue; it just took some time before the revenue exceeded the ad spend.

Today their return on ad spend (ROAS) is over 60% – an extremely high return.

Here were a few of the major changes that helped him improve his CTR and Conversion Rate:

  • Single Keyword Ad Groups (SKAGs). Ad Groups with just one keyword and just one match type. So a phrase match and an exact match for the same keyword would be broken out into separate ad groups.
  • Starting with highly targeted keywords first. Avoiding “Learn Thai” and bidding on “Learn Thai Online” instead for example.
  • Bidding on fewer keywords, but the more targeted ones, initially. Then broadening out once the ad copy and funnel were converting well already.

A Journey into Google Display network

Once Brett had spent a few thousand dollars, a Google rep contacted him about adding Google Display Network (GDN) to his traffic mix. It wasn’t an expensive test, so Brett went for it.

GDN was very effective for him at first, and almost doubled his traffic, while giving him a strong ROI. Over the last few years this channel has started to return a worse return, and he’s since then closed it down.

Facebook Ads

Facebook Ads is the new, cool kid on the block, but AdWords is still king. In 2016, Facebook Ads revenue brought in $27 billion vs AdWords’ $79 billion. Don’t get me wrong – I’m a big fan of Facebook Ads. But, in this case it seems like it hasn’t worked all that well for Brett.

With Facebook Ads, the only targeting Learn That from a White Guy could get is:

  • Expats from a foreign country
  • AND live in Thailand (or are interested in traveling to Thailand)
  • AND are interested in language learning.

Unfortunately, you can’t target someone by their desire to learn a language. Facebook won’t know if you’ve already learned Thai, if you want to learn Thai, or if you have no interest at all. In this case, having the ability to target by intent rather than by demographics makes all the difference.

Facebook Ads did convert for Brett, but it was breakeven at best.

A Simple and Effective Sales Funnel

Here’s the journey their average user goes through:

First, users will land on the website through a paid traffic source, generally AdWords. They’ll land on an email capture page. Once they have the email, they’ll be entered into an email series with 20+ emails. If they buy, they’re removed from the series and added to the customer list. Otherwise, they keep getting emailed until they purchase.

Let’s take a look at each of these steps in turn.

Getting the Email Address

The AdWords landing page is a standard, single-page above the fold email capture box:

Landing page from AdWords

The email capture box has a simple offer: try 4 lessons for free.

I love the “Get Lessons” call to action box. Overall the page “feels” very inviting, like your friendly neighborhood professor. It doesn’t feel like a sales page.

Email capture

The box is both clear and simple, and converts well.

The Email Series

Most of Brett’s customers will buy within the first two emails, but about 25% will take weeks or months to buy. The email follow up sequence helps build trust with customers over time.

The email series starts by giving customers exactly what they were promised – 4 free lessons. Then, over the course of the next few days and weeks, he continues to give them free tips on learning Thai.

The call to action in the email is to click the link to start learning.

Using a Quiz to Sell

Originally, the emails Brett sent would direct visitors to a text lesson.

Over time, he instead developed an interactive quiz. After installing the quiz, conversion rates increased by about 10%.

That’s the nuts and bolts of the funnel. It’s a pretty simple and straightforward AdWords funnel, which gathers a lead, and then converts that lead into a sale. Once the keywords, landing pages, and emails were tested and optimized, the marketing became fairly automated and hands-off.

Marketing Benchmarks and Metrics

Here were a few of the metrics Brett was able to share with us.

How percent of your email subscribers would convert into buyers?

12% of email signups from AdWords end up purchasing.

2% of email signups from Facebook Ads end up purchasing.

How long after the initial click before someone makes a purchase?

75% of the sales come within the first two emails.

25% of the sales come in the long tail, from email #3 to #40.

What’s a typical return on your ad spend?

Average ROAS is 65% on AdWords. Facebook is about break even.


Brett has set up an ideal lifestyle business for himself. He’s got automated marketing systems that requires very little maintenance. Every once in a while he tests new things – such as new Facebook Ads – but, for the most part it runs itself.

What strategies have you found for “set it and forget it” marketing? Share in the comments below!

Derek Pankaew

An In-Depth Look at Blue Apron’s $100 Million Marketing Strategy

Blue Apron has over 1 million customers, and has sold well over 10 million meal kits. They’ve single handedly pioneered the entire meal delivery industry.

To create the kind of growth they created, they used several innovative marketing strategies. They were early adopters in podcast advertising, created a truly unique referral program, and manage a powerful blend of online and offline marketing channels.

So, what allowed Blue Apron to grow so quickly? How did they go from startup to revolutionizing the food industry? And, why is the company also struggling to turn a profit?

Let’s take a look at what we can learn from this behemoth’s marketing stack.

Introduction & Table of Contents

Blue Apron is a master at scaling up. They’re in dozens of different marketing channels, from the oldest of old media to the newest of new.

We won’t look into every channel Blue Apron’s advertising in. Instead, we’ll examine the most interesting things they’re doing in their marketing. We’ll skip paid search and social – although Blue Apron spends a lot in these channels, their strategy is fairly similar to other companies.

Here’s what we’ll cover (clickable):

The overall strategy looks something like this: use a wide range of different marketing channels to entice new customers. Measure the cost of acquiring customers, as well as the lifetime value of those customers, for each channel they’re advertising in. Reduce ad spend in the channels that don’t work, and scale up the ones that do.

Some of their channels, such as podcasts, can be directly measured using custom links (i.e. Others, like their TV ad buys, likely need to be measured with a mix of sales metrics and brand metrics.

The $35 Million Referral Program

Blue Apron spends a fortunate on referral marketing. In 2016, they spent $35 million on customer referrals.

Blue Apron's marketing spend

Source: Blue Apron S1 (IPO filing)

How does their referral program work?

The main idea is that their top customers most likely know people who are similar to themselves. If you’re an upper middle class family who likes to cook, you probably know other upper middle class families who like to cook.

Blue Apron is confident that if they can get those people to try their service, they’ll stick around and be a customer for a long time.

Well, what better way to get them to try it out than to send them a free box?
Blue Apron's referral program

That’s right – Blue Apron spends $35 million a year on giving away free food. And those free meals are some of the best marketing dollars Blue Apron spends.

Of our Customers for the first quarter of 2017, 34% were acquired through our customer referral program, in which certain existing customers may invite others to receive a complimentary meal delivery.  — Blue Apron S1 (IPO filings)

In Q1 of 2017, referrals was 14.75% of marketing spend, but was responsible for 34% of their customers. Presumably, these customers most likely spend more and stay customers for longer as well. It’s entirely possible that half of Blue Apron’s revenues comes from referrals.

A few ingredients go into the success of this strategy:

  • Have a great product that people want to share.
  • Give them a great offer that they’d feel good about inviting friends to.
  • Make sure the customer value created is larger than the spend.

Blue Apron isn’t alone in using this strategy, of course. Dropbox used a similar tactic to get to 4 million users in less than 2 years. Of course, it’s much easier to scale a referral program when you’re selling software rather than food. Blue Apron’s referral program is one of the most successful in the physical products industry.

Online Marketing Channels

Podcast Advertising

Podcast advertising today is like Facebook Ads 5 years ago. Response rates are high, costs are low, and few advertisers understand how the channel really works. It’s a big opportunity right now that only a few companies have taken advantage of.

Blue Apron was one of the first companies to really scale their podcast ads. They blanketed the airwaves (pod waves?) and have since scaled up to buying ads in just about every category.

Here are a couple examples of Blue Apron’s podcast ads:

Blue Apron’s Podcast Ad – on the “No Sleep” Podcast:


Younger Cop: Do you really think the valentino brothers will show?

Older Cop: They have to! If our intel is right, this is their base of operations.

Younger Cop: Yeah, but we’ve been sitting in this squad car for hours. My legs are cramping up.

Older Cop: This is your first stake out right? Eh, just sit there and be patient.

Younger Cop: (Grumble) ugh I’m so hungry, didn’t you bring snacks?

Older Cop: That was your job, kid.

Younger Cop: Ugh we’re gonna starve to death. Hey is it true what they say about the oldest Valentino? They call him the surgeon because he decapitates his rivals and keeps their heads?

Older Cop: Don’t believe everything you hear. Look alive kid, this might be them.

Younger Cop: Nope, its’ a delivery truck. Oh! Blue Apron!

Older Cop: Blue Apron?

Younger Cop: Yeah, they’re the #1 food delivery service in the country. I signed up a few months ago by going to Their food is incredible. Their culinary team likes to surprise us with new recipes that aren’t repeated for a year. Last night I gorged myself on seared steaks and peppercorn sauce, with roasted potatoes and green beans. Mmm, so good.

Older Cop: (Grumble)

Younger Cop: heh, looks like I’m not the only one who’s hungry.

Older Cop: Shut it, I didn’t have breakfast.

Younger Cop: Look at that box, sitting all alone on the door step. EAT ME it’s saying, EAAT MEEE.

Older Cop: Don’t tempt me.

Younger Cop: You know, not all ingredients are created equal. Their seafood is sourced sustainably, their beef, chicken, and pork come from responsibly raised animals, so you know it’s going to be delicious. It’d be a shame to waste all that good will on the Valentino brothers. We could run over there real quickly and make supper.

Older Cop: Even if I entertained the notion, we wouldn’t have time to cook anything before the Valentinos got back.

Younger Cop: Actually every meal can be prepared in 40 minutes or less with step by step recipe cards. They’ll never know we were there!

Older Cop: Ugh fine, don’t make me regret this. Wow, there’s exactly enough here for two people. I’m always buying too much and have to throw things out! So which recipe should we make?

Younger Cop: The meatball with tomato sauce with asparagus and creamy rice sounds amazing.

Older Cop: Whoa this would pair great with wine! Go see if there’s any in the cellar will ya?

Younger Cop: Hey remember that rumor about the decapitated heads? I found them … and they’re still breathing!

Narrator: Check out this week’s menu and get your first 3 meals free by going to Blue Apron, a better way to cook.

This is a fantastic example of an ad that blends into the show’s style (the show is a horror fiction show). Here’s another example:

Blue Apron’s Ad on “You Are Not So Smart”:

Blue Apron’s podcast ads are almost all read by the host of the podcast. They’re running ads across many different categories – from health podcasts to true crime and everything in between.

Here are a few tips for podcast advertisers:

  • Typical rates are $8 to $40 per thousand downloads (CPM). The rates vary based on the popularity of the podcast, as well as the category the podcast is in. The True Crime category might average around $10 CPM, while the tech or business category might average around $30 CPM. Then you have the really popular shows, which can command CPMs of up to $50.
  • Always go for host read ads. Don’t submit a sound clip to have them insert into the podcast. One of the reasons podcast ads work so well is because you get to leverage the connection the host has to the audience. Don’t make it sound like a traditional ad – personalize it.
  • Programmatically inserted ads usually don’t perform well. Programmatically inserted ads mean you’re buying ads in a podcast for a certain amount of time. For example, for the month of June, when a listener downloads a podcast episode, your ad will be “inserted” into the ad slot as the customer downloads it. In July, your ads are removed and someone else’s ads will be inserted. These types of ads generally perform far worse than “hard coded” ads, where the ad is actually recorded into the episode rather than inserted via software.

Sponsoring YouTube Channels:

Brands often pay YouTube stars to create videos about their brand. You get to leverage the connection YouTubers have with their audience, and get a unique piece of content created.

YouTubers have a well honed sense of showmanship, and can  create videos that really “wow” audiences. Depending on the agreement you have with the YouTuber, you might be able to use those content pieces in other places – your website, your Facebook Ads, etc.

That said, this channel is harder to manage than many other channels. Instead of buying impressions through an ad platform, you’re making deals directly with various YouTube personalities. And, like doing business with the artistic, it can be a quirky experience.

There are, of course, agencies who which specializes in managing that process for you. Typically they charge a 25% premium on top of what the YouTuber charges, and come with a a $25,000 minimum investment.

Example: Blue Apron’s Sponsored YouTube Video,
“Cooking the Perfect Meal”

This skit is short, funny, and really gets the Blue Apron value prop across. I’m a big fan – it’s worth a watch:

Blue Apron sponsoring YouTube videos

If you want to experiment with sponsoring YouTube videos, Famebit is the primary avenue for doing so. Famebit is a platform that connects YouTube influencers with advertisers who want to sponsor them. They were acquired by Google in 2016. For a more “done for you” approach, Grapevine is an agency that specializes in this.

This channel only works for certain kinds of businesses. For businesses in fashion, food, or fitness YouTube sponsorships can work great. For businesses selling intangible products – software, consulting, etc – YouTube is not a very good channel. Basically, if there are a lot of people creating YouTube channels about your topic, then it could be a good channel.

When paying for YouTube sponsorships, here are a few things to look for:

  • Typical rates are between half a cent to 1 cent per subscriber. In other words, a channel with 100,000 subscribers should cost $500 to $1,000 for one sponsored video.
  • A good view through rate is about 15%. In other words, a channel with 100,000 subscribers should get around 15,000 views per video they release.
  • The size of channels you advertise in should be between 100k to 500k subscribers. Channels with less than 100k subscribers are usually hobbyists, who might not take the video seriously. Over 500k channels often have agents and demand higher CPMs. Between 100k and 500k is the sweet spot.

Display Advertising

Blue Apron’s display ads are gorgeous. Their copy is witty. Overall, they do a great job.

That said, according to SimilarWeb, display is a very small percentage of Blue Apron’s traffic:

Blue Apron's traffic sources

Of all their marketing channels, display is likely the least effective. I’ve also usually found display to perform worse than most other channels.

It works for retargeting. It can work great for retargeting. But for bringing in new customers, I’ve generally found display to be much worse than most other channels.

All that said, I do love the photography and copy that Blue Apron’s got going. They know how to make some pretty great ads.

Blue Apron Display Ad 1 Blue Apron Display Ad 2
Blue Apron Display Ad 3 Blue Apron Display Ad 4

Sponsored Reviews via Native Ads

Blue Apron is a big advertiser on native ads. Native ads are those annoying ad units at the bottom of most news sites. They tend to blend in with the news style of the site, using attention catching headlines and photos to try and entice a click.

The three most common native advertising networks are Outbrain, Taboola, and Yahoo Gemini. Typically advertisers send readers to a “pre-sale page”, rather than a product page, which is designed to introduce people to their product.

In this case, Blue Apron introduces people to the idea via a “We Tried Blue Apron: Here’s What Happened” ad. This directs visitors to a page where the writer shares their experience with Blue Apron.

Blue Apron's native ads on outbrain

This ad then takes you to a page like this:

Blue Apron Native Ads - landing page

Native Ads work best when combined with retargeting. Since most customers aren’t necessarily in the mood to buy right now, retargeting becomes even more important with Native Ads.

Native Ads generally convert worse than Facebook, Podcasts, or paid search, but offers more scale. It’s best to get a couple other channels to work first, before scaling up by adding native ads.

Offline Marketing Channels

TV Advertising

TV is an interesting choice for Blue Apron. Traditionally, TV ads are bought by advertisers looking to buy mind share. They’re looking to increase awareness of their brand, and associate more positive emotions to their brand, so that when someone’s considering a purchase, that brand comes to mind. Colgate buys TV ads so that when you’re in front of a the tooth paste section at the grocery store, you’ll buy Colgate instead of Crest.

Blue Apron is different. They’re trying to generate sales directly. There is never a moment where you’re standing in a grocery store, choosing between Blue Apron, Hello Fresh, and Plated. Instead, they have to get people to think “oh, that’s a product I want to try – let me go online and buy that.”

Blue Apron is a bit of a pioneer in this aspect. Although “infomercial style” commercials have worked in the past, those usually come at the cost of bad branding. Blue Apron’s TV ads are beautifully branded, and aim at driving sales.

It’s a kind of hybrid between a direct response ad and a branding ad.

I don’t have any data about how these ads performed for Blue Apron, but if I had to guess, I’d bet that they performed worse (from an ROI perspective) than many of their other channels. TV is a blunt instrument – your targeting capabilities are very limited. For example, you can target left-leaning women from age 25 to 45. But you can’t target married women with 2 kids, who eat Paleo, and also shop at Whole Foods.

Compared to channels like Facebook Ads, podcast ads, and YouTube, TV’s targeting capabilities are clearly inferior. However, TV has massive scale. It’s impossible to spend $10 million on podcast ads or YouTube influencers in an effective way. It’s very easy to spend $30 million on TV.

At a certain point, scaling up becomes more important than getting the best ROI out of your marketing dollars – especially if you have venture capitalists or Wall Street to please. But then again, spending money on marketing that doesn’t pay off is a big part of why Blue Apron’s struggling to get profitable.

Radio Advertising

Blue Apron spent $293,000 in 2015 and $2.41 million in 2016.

Radio is much more scalable than podcast ads. Since radio has been around for so long, many agencies already have the knowledge and connections to buy ads at scale. Podcasts are still a bit of a Wild Wild West compared to radio. And podcasts don’t have the scale. Geico alone is estimated to spend more on radio ads than all of podcast ads combined.

For brands that need scale, podcast is not replacing radio anytime soon. Yes, podcasts are growing very quickly, but radio is still where the eyeballs (ear drums?) are.

Typical CPMs are $10 to $20, or about half of podcast rates.

Here’s an example of a Blue Apron’s radio ad:

Just like their podcast ads, this is a host-read ad, integrated into the style of the show itself. They’ve basically taken podcast-style ads and turned them into radio ads.

Direct Mail

Blue Apron is a big advertiser in direct mail. Direct mail’s one of the oldest and most proven channels for direct response marketing.

Blue Apron often participates in what’s called co-op marketing. Essentially the way it works is this:

  • Companies enter into the co-op by submitting their customers’ names and addresses.
  • The names and addresses, along with purchase data, create a database that’s shared among everyone else in the co-op.
  • Once you upload your customers, you’ll be able to access everyone else’s customers as well.
  • Using your customers as a baseline, an algorithm can find other people in the database who are similar to your customers. You can market to those people to try and get them to buy.

This is very similar to Facebook’s lookalike audiences, except applied to direct mail. It generally yields higher returns than any other direct mail strategy. Other strategies include zip code mailing, mailing based on magazine subscriptions, and mailing based on people who’ve purchased certain products.

The trade-off for co-ops is around privacy. More details on how co-ops work can be found here.

Example Blue Apron Direct Mail Piece:

Blue Apron's direct mail piece

Although Blue Apron often uses co-ops, they probably test different kinds of mailing lists as well.

In addition to the different groups of people they mail to, they also test the format and messaging of their mail pieces. Above is a three-section dropdown format. They also send postcard format, in an envelope format, and several more elaborate formats as well.

Understanding Blue Apron’s Marketing Metrics

Blue Apron’s marketing team did a lot of things right. They took a new concept and introduced it to millions of people. In a way, they are a smashing success.

Unfortunately, there’s a lot of problems with their core metrics. Let’s take a look.

Customer Acquisition Cost

Blue Apron’s customer acquisition cost is a mix of their marketing spend, as well as the discount they give to initial customers. They typically give away some number of free meals on a customer’s first order, which is included in their acquisition cost.

Blue Apron tells us their acquisition cost is $94 between 2014-2016. That’s an average, which most likely hides a rising acquisition cost. Fortunately, using data from Blue Apron’s public filings, several analysts have calculated their current acquisition cost to be between $150 and $400 per customer.


Blue Apron also doesn’t provide any data on cancellation rates.

Fortunately, a third party company called Second Measure does. Second Measure buys data from credit card and loyalty card companies. By aggregating data from thousands of credit card statements, they can figure out how often people stay subscribed to various companies.

Blue Apron customer retention

By the 3 month mark, Blue Apron loses more than half of their customers. Within 1 year, Blue Apron loses about 80% of their customers.

In other words, Blue Apron is great at getting new customers. Keeping customers? Not so much.

Lifetime Value

Lifetime value is how much a customer is worth, over the lifetime of their relationship with the business. Blue Apron’s LTV is not that hard to figure out.

Their gross profit margin is about 22%. Their average order value is $58. That means they’re making around $12.50 per order.

Blue Apron's average order value

In other words, if they’re spending $150 to acquire a customer, they’d need customers to stick around for 1 year to break even. Unfortunately, by 1 year they’ve already lost 80% of their customers.

The Big Picture

Blue Apron – while brilliant in many of their acquisition strategies – is losing money. When they first went public, they were worth $1.9 billion. Today, they’re worth $500 million. That’s about 25% of their IPO value. Most of this is because their marketing metrics don’t work out.

This doesn’t mean Blue Apron can’t be a viable business. It just might be a smaller business than investors expected. It’s entirely possible that they can scale down their marketing spend to just the most efficient channels, and have a business that grows slowly and runs profitably.

Of course, a slowly growing but profitable company is not what Wall Street wants.

Whether Blue Apron will make it is still to be seen. But, one thing’s for certain – they’ve pioneered a lot of marketing strategies we can all learn from. From their massive buys in podcast ads, to their $35 million referral program, to their hilarious YouTube videos, there’s a lot in their marketing mix we can all learn from.

What are your thoughts on Blue Apron’s marketing? Share in the comments below!

Derek Pankaew