Website not giving you the results you expected? How about some non-judgemental feedback from a fresh set of eyes? Apply for an online marketing tune-up with Paul Montreal

What If You’re Selling To More Than One Customer? (CodeCombat) A Y Combinator Website Marketing Tune-Up.

by Paul Montreal. Average Reading Time: almost 14 minutes.


Today I’m looking at Nick’s website He’s got a Y Combinator backed start-up that’s helping kids learn programming through games. Let’s dive in and see what we can learn…

Codecombat Y Combinator Website Marketing Makeover.

Click for full size homepage image.

1. Enough clarity to continue, logically and emotionally.

I love what I see right away. Whilst the opening promise, or headline, doesn’t quite roll of the tongue, it does give me a very clear idea of what this place is all about. “The most engaging game for learning programming.”

I see a lot of random adjectives thrown around on landing pages, “The fastest, the cheapest, the bestest on the interwebz”. But when you’re a teacher trying to get kids to focus, “engagement” is probably your biggest problem. And honing in on the biggest problem is exactly what that opening promise is all about.

But what I love even more is that there’s a big human image accompanying the promise and the image communicates the exact same message as the headline. You can see, and your brain can “feel” the happy, engaged kids. Happy kids = happy teachers.

Y Combinator marketing teardown.

2. Make the aesthetics congruent with the key idea.

The only improvement I’d make is to work on your color scheme. If you applied the word “engagement” to the overall aesthetic, you’d end up with a slightly different color palette. That great image has a dark filter over it. We’ve essentially sucked the color and some of the life out of it.

Our brains encode images in all sorts of interesting ways. Those “pictures in our mind” determine how we remember things. And how we remember things is how we think about them.

We’ve seen over several of the recent tune-ups, that people who are skilled at thinking in a technical manner, will often be very engaged with the technical aspects of a product, but less engaged with the human aspects. So, we see big screenshots of computer screens but tiny images of human faces.

When your job is staring at a computer screen all day to create amazing new products, that’s exactly how you want things. But when the task is connecting with the humans on the other end of a sales conversation, we have to change mode.

As well as the size of an image, and the shape of the image, we use color to store images in our memory. Think about this for a second – how do you know that a memory is a memory and not a dream about a future event that has yet to happen?

One of the common ways we know a memory is a memory is because we color it differently. Memories tend to be dull, or black and white, at least initially. We don’t think about this consciously, we just “know” it’s a memory and not something else.

So, when it comes to setting the “tone” for our aesthetics, if “engagement” is the vibe we’re trying to communicate, then I’d recommend turning up the vibrancy. Lose the dull color palette, change the image filters. Turn up the color. Bring people out of the past and into the present. In the present, the time when we’re really engaged, everything is about as bright and shiny as it gets.

Of course there’s a real skill to aesthetics, a professional skill. So, once you really understand what you want to achieve, it can be worth the time and cost of having a designer with an exceptional eye spend the many, many hours it takes, to find and manipulate those half a dozen images that will perfectly communicate the human feelings behind your pitch.

Because when you get it right, your idea and its value will just make more sense, unconsciously, to your customer. They won’t be able to tell you why, they’ll just feel like it all makes sense.

Codecombat website marketing tune-up.

3. Who are the individual players, what can they do here?

It’s great to have the most important call to action above the fold. But the 3 buttons and their descriptions aren’t particularly clear here.

Is the play button really a “Demo” button?

The “Join class” button leads to another login page which itself is just as confused and seems to be in the middle of some transition.

This is where the customer starts having to suffer the realities of developing. And we don’t want that. We want clarity, with very little thinking. Decisions that people can’t really screw up.

The third button says:

Teachers & Educators
Learn how our classroom-in-a-box platform fits into your curriculum.

But the button doesn’t seem to be taking them to a “learning” part of the site. The “learning” part of the site seems to me to be the marketing copy directly below. And that link is for people who already understand the pitch and want to get actual lessons set-up?

Define the path by the person taking it.

If you’ve got different types of people using your website, the most useful way to organize that website is around the PEOPLE, not around your functions and features.

In this case, we’ve got at least 2 distinct groups. Kids/Students and Teachers.

But in this market, I’m sure there’s likely to be a third group, I’ll call them “Administrators” for now.

When we’re doing our job right, the Kids are tugging at the trousers of the Teachers, saying “please Sir, buy CodeCombat” and the teachers are tugging on the trousers of the Administrators saying “please Sir, can we buy CodeCombat”.

Beyond those 3 groups, you may even have an important 4th party, Parents. Parents can be supportive, or parents can be disruptive, so you better have a plan for persuading and convincing them that their kids “playing more computer games” is a good thing, and not the beginning of the end.

So, how do you eat the 4 elephants in the room? One at a time.

I’d keep the big happy kid picture and the opening promise / headline, then I’d identify and silo those different parties off onto individual pages that apply and speak only to their unique interests.


Below each header (and you can maybe include a small image representing each different party), reinforce the unique value that you represent to each group with a single sentence…

Learn programming by playing games.

The language of science and success, now available to every family.

All your computer science lessons covered (even if you’ve never taught tech before).

An affordable way to raise computer science grades in your region.

(I’m making these up off the top of my head. Yours should reflect whatever is the most important point you want to communicate after having thoroughly worked out the long-form sales pitch for each group).

Then, give them one or two very clear ‘call to action’ buttons each. They either want to start using the product if they are already familiar, or they want to “learn” what it’s all about if they are new.

Think about it, your marketing site is a sales pitch. A sales pitch is just a persuasive conversation. If you had a kid, a parent, a teacher and an administrator in the same room, they’d all have totally different perspectives and concerns. You’d have the best chance of convincing them all, if you spoke directly to each of them individually about their problems from their perspective.

You’ve already attempted to do this, we just have to go all the way.

Codecombat website teardown.

4. Don’t undersell.

One of the advantages of creating different pages for different parties, is that you now have the space to go deeper into really selling each important point you want to make.

I think the copy on the homepage is better than average and hits a lot of the right notes. There’s clearly a connection being made between the product and the human emotions that people actually care about. Let’s look at a few…

Our courses have been specifically playtested to excel in the classroom, even by teachers with little to no prior programming experience.

From the Teachers perspective this is really important. There are lots of teachers who end up running a class that is not their area of expertise. They’ll be feeling totally out of their depths. So by making it clear that this is for people with no prior programming experience, we’re solving a major problem.

Remember how we determine what’s important to people? It’s not our features, it’s “what keeps them awake at night?” Can you imagine a teacher being told “you have to take this computer class next year” with no prior experience? Can you imagine that teacher having some sleepless nights over it? I can. And I bet it happens all the time.

So give it the space it deserves. Go deep into showing how you remove that pain for inexperienced teachers in a pitch devoted just to them.

Democratizing the process of learning coding is at the core of our philosophy. Everyone should be able to learn to code.

These days, coding is being talked about in the same way that “learning English” used to be talked about, in countries around the world, by parents who wanted their kids to have more opportunities in the future. There’s a genuine, and very positive movement to level the playing field, and see those who are willing to study and act, succeed. Above and beyond how lucky they were to be born in one area, or another.

This kind of thing really can be part of a wider movement that makes the world a better place for us all to live in. So, sell the shit out of it! Tap into those positive emotions. Martin Luther King didn’t make change with a 3 minute speech full of one line bullets. He spoke for 17 minutes.

Studies suggest gaming is good for children’s brains. (it’s true!)
When game-based learning systems are compared against conventional assessment methods, the difference is clear: games are better at helping students retain knowledge, concentrate and perform at a higher level of achievement

I’d recommend never qualifying a statement with “It’s true!”. It makes your argument sound weak. In fact, when people aren’t telling the truth, they often start their sentences with “To tell you the truth…”

That isn’t to say that you shouldn’t pay attention to the concerns and cynicism of your audience. I’m sure that some parents (especially as you expand beyond the tech world bubble) will question the validity of using games to teach “serious” topics. And it’s OK to acknowledge that in your copy and reply to it directly. You don’t need to pretend that no one ever says anything negative. Just state your persuasive argument without over qualifying it.

A classroom in-a-box for teaching computer science.

What is a classroom in a box? I think I know what you’re trying to get at here. You’re appealing to the same teacher who isn’t necessarily a computer scientist themselves, and they just want a “one stop” solution. Buy it and it takes care of everything.

But I’d prefer to see you lay that out, in full, so that those teachers really understand it. Everything does not have to be tweetable. When we condense our selling point too much, they become cliches or worse, incomprehensible. Clever, or cute, is somewhere in-between. Where we kind of know what it means, but don’t really feel it deep down.

You’d have to have more conversations with your teachers to find the right combination of words. To hear the problem from their angle, before you could come up with a better headline. I don’t know if a teacher thinks: “How am I going to fill X hours per week with computer science this year?” or whether they think “How am I going to get through teaching two years of computer science?”. But there will be some metric, some hole they need to fill. Once you better understand the nature of that hole, you can fashion a better peg to fill it.

So, rather than saying “A classroom in-a-box for teaching computer science” you might say, “Absolutely everything you need to teach computer science twice a week for the next 3 years. (Even if you haven’t taught it before).”

The main point is, try to sound human and specific, rather than clever or cute.

Secret Squirrel Stuff
to view this part of the content. (It is the best bit! Muhahaha).


This is definitely a site that’s doing more things right than they are wrong. I’d split the pitches up, so you can talk specifically to the different needs of the different parties involved. That will allow you to do deeper into solving their individual problems and providing lots of social proof in the form of testimonials and (over time) case studies. The existing copy is good, but it’s just the tip of the persuasive iceberg. I’d also make sure to include pitches for Parents and Administrators.

Just as the games themselves are colorful, don’t make your marketing aesthetic dull and distant by comparison. Dial up that saturation, that intensity of color, especially in the human images. As you create the individual pitches, reduce the number of generic icons and replace them with more relevant human images.

Marketing, just like coding is a complex process. But it does follow rules. Rules governed by the human personality. Piece by piece, person by person, we can work out how to communicate in the language that makes our customers move through that maze. Collecting Scooby snacks as we go. That’s magic as well!

I’d like to thank Nick and the CodeCombat team for sharing this with the community, so we can all get a little smarter. Until next week, stay the course, see it through, make your mark!

Vote on Hacker News

Website not giving you the results you expected? How about some non-judgemental feedback from a fresh set of eyes? Apply for an online marketing tune-up with Paul Montreal

What do you think?...

This is a safe place for learning, so don't be shy. But do be cool.

  1. nick08 says:

    Love this! One question: with regard to splitting up the initial call to action based on role (student, parent, teacher, administrator) instead of action (play now, join class, get started)–do you have any examples of sites that do this well?

    One thing that is always a bit tricky is that there’s crossover between the categories a bit, where “students” may want to join a class or play on their own, which is two different parts of the site. Maybe an interstitial question once they hit some sort of “Students” button to then route them to the right place? What do you think?

    • Paul Montreal says:

      I didn’t have any particular sites in mind when made that suggestion. Don’t think of it from a design perspective, just think about it from a conversation perspective. Try to get to the point where you’re having a one-on-one conversation that’s specific and tailored to each individual group as quickly as possible. Think of it as if you have 4 homepages, rather than one. And each one of those homepages speaks only to the needs of one particular group.

      Interstitial questions, or choices sound fine. It matters less than you might think how many clicks or pages a user goes through on a journey to learn / achieve something, as long as each choice is easy to understand and they feel like they are getting closer to a goal at each step.

      I’d rather send a user through 3 pages, where they only have to make one decision, A or B on each page. Than have them make 3 decisions all on one page. But I’m not sure your process is that complex. After you’ve identified who you’re talking to it’s just a process of deepening the pitch and the education. And as you go deeper into the pitch you just regularly ask “want to learn more” or are you ready to “sign-up” where sign-up is whatever the next logical step in the commitment process is. A kid might want to just dive in. A teacher might want to sign-up for more training / education. An administrator may want to talk to someone about a deal that covers a region etc. This software means something different to each of these parties, it solves different problems for different people, that’s why we need to talk to them on their own.

      Does that make sense? It’s hard to communicate these slightly abstract ideas without doing a wireframe or something and I’m afraid my time is limited, I was 2 days behind schedule getting this out this week. Tune-up Tuesday ended up being Tune-up Thursday :)

Leave a reply