The Code is your Enemy


You’re a builder, a creator — whether a back-end programmer, a Linux hacker, a Javascript ninja, a UX magician, a designer. You make stuff.

That’s great of course, because in a new startup everyone needs to be either making stuff or selling stuff — there’s no room for managers and executives and strategists. But this also produces a natural weakness, and when I look at what made me a successful entrepreneur — not just a great coder — it’s that I acknowledged and overcame that weakness.

The weakness is your love of creation. You love to write clean, tested, scalable, extensible, beautiful code. You love converting marketing requirements into 960-wide artwork, replete with whitespace, custom fonts. You love developing an entire app in the browser against a scalable back-end.

And because you love it, it’s what you do. You wake up in the morning thinking about what you can make. You open Photoshop before you consult your to-do list because there’s something you just need to tweak. You launch xterm before Highrise because the server was running just a tad more slowly than you’d expect and you want to paw through log files.

The trouble is, this is almost certainly not the activity that would most benefit your startup.

As much as you’re a minor deity when it comes to TextMate or Balsamiq, so are almost all the technical founders of all high-tech startups. We can all write code — at least well enough to get a product launched and through a few iterations. We can all make a functional website — at least well enough to produce orders.

Most startups fail, despite excellent coding and/or design skills. Yet, because that’s your love, your passion, that’s the stuff you dwell on.

Most startups fail because not enough people show up on the home page, or people show up but they don’t try the product, or they don’t pay for the product, or it’s too expensive to get them to show up, or they don’t tell their friends to come along, or because it’s not solving a pain that people have, or it’s not solving a pain that people know they have, or it’s too hard to explain the pain, or a bunch of other things that are not whether the code works or whether it looks good.

Customers don’t open their wallets based on your unit test coverage or whether you used Bodoni instead of Times New Roman on the home page. In fact I’ve made millions of dollars on companies with hideously ugly websites and buggy code. Those things are actually not the most important things.

So if these things — the raw materials and skills used in web-based startups — are necessary but insufficient, what are those things outside your comfort zone which nevertheless are the things that are actually valuable to your company?

Here are two:

1. Have you talked to 50 potential customers? By that I mean fifty, not a dozen. I once vetted an idea and after the first 10 interviews I thought I was really on to something. Suddenly things changed and future interviews weren’t so clear. Turns out there was accidental bias in the people I selected, obvious only in hindsight. Another time, everyone said it was a great idea, but it wasn’t, which was only clear after dozens of interviews. Do you find it hard to locate that many people? Well it will still be hard to locate them after you’ve built a product, and then it’s unlikely the product matches what they want! So solve that hard problem now. Don’t forget to vet the price at the same time, otherwise it doesn’t count.

2. Are people coming to your website every day? If not, solving that is much harder and much more outside your control than building software. Consider: Would you rather get hired as the CTO of a company with 1,000 daily new, unique, qualified visitors with no product, or the CTO of a company with a stable product and 10 uniques visits to the home page? You know you can solve for the first case; who knows about the second? But if you stay nose-down in the code instead of working on getting attention, that’s exactly the company you’re building.

Here’s a coder-centric way of thinking about all this more generically: When you tackle a large development project, do you tackle the high-risk, inadequately-understood modules first, or leave those to the end? First, of course, because you understand the rest and you need to solve the unknown problems while you still have lots of time to pivot and re-plan.

This is the same thing, it’s just that in a business it’s the attention, marketing, positioning, selling, defining part that’s high-risk and inadequately-understood, and all of the coding and design is low-risk and well-understood in comparison.

So force yourself out of your comfort zone. You’ll also do coding and design and that’s fine of course. But force yourself to mostly do those other things that create a valuable business.

29 responses to “The Code is your Enemy”

  1. This is a really great post! I definitely struggle with this, and I see a lot of people who wont acknowledge it.

    People think the biggest problem with starting a software business is developing the software, but actually its whether anybody will buy it!

  2. This is the number one mistake I made with all my previous “startups”. They were really all just an excuse to code something I thought was cool. They weren’t businesses at all.

    If I was to try to found another startup today, one of the first things I’d do would be to just set up a blog about the target market.

    – Want to build software to help organise around the world trips? Start writing a travel tips blog.
    – Want to build software for printing companies? Start writing about designing leaflets and running a print company efficiently.

    This will do a number of important things for you

    1. Test your knowledge. If you don’t know enough travel tips to fil a blog then perhaps you should pick another domain to build a startup in.
    2. Test your interest. Don’t want to write about travel tips all day? Why are you building a business in that area?
    3. Build an audience. If you do this successfully, in a few months you can be in the exact position Jason is describing – you’ll have 1,000 daily uniques to launch a product onto.

    • interesting. another issue to solve for is once you do start blogging, how do you get awareness for your blog? this is a real need for a startup business in the real world, so again if you can’t figure out how to get exposure and build a following, don’t bother to start writing code yet. this is basic marketing, find a group of people whom might be leads, target them with messages they want to get, and keep them engaged with value. i agree, start there first. if your blog allows comments, and you get comments, you’ve just started your customer discovery and development exercises.

  3. Smart down to earth post Jason. Your content is great to read because you can tell you actually lived it.

    PS – David Tuite, I think you’re on to something…

  4. Nice and wise post (as usual!). This also means that once a startup becomes sufficiently successful, it has to invest into fixing the debt representing by the poor code base written quickly, else at a later stage this debt could be well impair drastically the business.

    • … drastically affect the scalability to grow, ie twitter’s early days. but still be focused on delivering value to the customer, not simply re-writing code to be the most excellent code you’ve ever written, “crushing it” :)

      • Sure the challenge is to add value and reimburse the early days debt at the same time. Both are not incompatible but still it is a challenge, this is exactly what I am facing with my own software business!

  5. Dear Jason, your comments are very valid and its obvious you build on lots of valuable experiences. Running a SaaS startup for years and coming from a sales background (with quite some past success) I very much do focus from das one on customers. Still its not easy and we also had to learn that good sw – if solving a complex topic – might not be selling easily. We have extended our SaaS product range with an very simple and easy online offering which serves as a go to market vehicle. Getting visibility with that helps cross selling to the others.
    Another point I personally believe in that US based customers are by far more willing to ake a risk and test a new offering from a startup. This is not the case in Germany, leading to further challenges for German based startups in b2b.

  6. It’s hard!

    The only fun is coding!

    Tech assistance is ok!

    Selling is not fun, SEO can be ok.

    But managing is really dreadful and abominable!!!

  7. You are right, you touched my weak spot. The problems is that we (computer scientists, engineers) are wired to create products (and by create, I mean the build part), and our training just reinforces that behavior. It takes real discipline to get out of that vicious circle (think, code, admire creation, repeat “ad infinitum”). Learn how to write good copy, send a bunch of emails to prospects, cold calls just makes us nervous… Only thinking of it makes me want to open my IDE.

    And the ultimate problem is that writing the code you feel you are moving forward; marketing, you feel like you are not moving at all (at best). I think that to be able to follow you advice, the first thing to be convinced of is that marketing is a VERY productive activity. Then you step out of your confort zone and at least feel good about having done it.

    Thanks for the post.

    • what do you get rewarded for? if you are a founder, do you get rewarded for writing great code? or do you get rewarded for solving a customer’s problem? is payment your reward, or is praise? do customers praise you for your excellent implementation of the singleton pattern, or do they pay you to make it easier for them to gain more customers?

      your code can form the foundation of these rewards, but the code itself will not get those rewards, you have to get people aware of your creation. without that awareness you’ll never get your rewards. it’s almost like writing code but never shipping it. change your reward to be customer recognition and you will improve your outlook on the required activities of marketing / selling.

      • well spank me with a horizontal rule! i just rewrote a sql query to “improve” the dataset that would be presented to the customer WITHOUT the customer telling me they needed that! Sometimes you gotta guess, other times you need to test first. I couldn’t STOP MYSELF this time.

  8. No matter what we create, code we have to know that nothing is precise, or without errors. We need to understand the variables that are all around a platform. So i guess that when you code something you need to be ready to the unknown.

  9. Great tips. I couldn’t agree more. Coding always feels like the “real work” because you are creating something. But before you do, you need to figure out what is it you are creating and who will use it.

    A friend of mine used to say “coding is like a drug, avoid it when you can and tackle the real problems”.

  10. During a typical trial their will be opening statements by both the prosecution and the defence. The next step will be where the prosecutor presents their case. This will be done by calling witnesses and presenting evidence to the court.It is what you keep that matters. Good Revenue per sale.This is often missed by many.The Ayurvedic view of eye colour is that your eyes are the colour that you have in “excess” (namely, in the astral or spiritual body). That view supports the idea that changes in eye colour may be Nike Air Max For Sale possible due to the deep cleansing provided by a raw food diet. More interestingly, there are cases of Multiple Personality Syndrome in which a person’s eye colour changes according to which personality is occupant at the time!

    Rolling up to the park you can probably catch me blasting some “These are my people” by Rodney Atkins. Nothing gets me more pumped then a good ole’ country boy playing the fiddle. What gets you pumped?The first impulse may become to look the net for internet vendors, second hand stores as well as market websites. This is without a doubt a smart thought since right now there are so many online stores offering fabulous discount rates on cheap Jordans. You may also obtain a cheaper deal on market internet sites because customers presently there won”t be prepared to spend top dollar for secondhand products, and retailers realize that too.So you love making your own college basketball picks and can’t help but bet a few college games every week. You know you can only follow so many teams so you limit college basketball betting mostly to your own conference. You only bet teams you have actually watched play.

    You will discover price reduction UGG boots are offered on the web now.Thereby, obtaining the boots because the current to suit your needs as well as your family or associates is incontrovertibly a pleasant alternative. This is a advantageous alternative, is not it?Styles who definitely have developed from the item of clothing add some cover sleeve, polo nck, tank, camisole, long sleeve, and additionally frilled and skirted leotard. The most common leotards may be the particular camisole. The most important element is its formfitting superior and suppleness.Low top Trainers. These are light shoes that offer greater maneuverability where speed and dexterity improve on the court. The ankles are exposed; therefore, movements are not restricted.

    This is certainly not favorite, “vulgar,and happy” them, advocate one of the most favorite brand name identify, within this age, but in addition the happiest. I checked out the shop, when yet again lamented his individual spirit and tips of aging: The footwear? A number of people get them?Thanks to MJ’s timely return which made this new release become a hit at cheap nike air max 2012 its first release. Also due to MJ’s return, Nike successfully reversed its declining sales. With AJ 11 Jordan Melo M7 on his feet, MJ led Bulls to another Final Champions.They can be classy, sporty, casual, cute or fashionable Order Retro High SVSM 1S Online. Ankle high more locks feet with the tradition of the Air Jordan 13 Retro Black / Red 309259-061 Retro High SVSM 1S For Sale. all of medicines buying in our store are what we after special skill refineses, so the price is more a little bit more expensive, wear a medicine 200 dollars, if you decide to want, I sell to you, if you dislike words of expensive price, I can open other medicines for you.

    He fished by obstinate isles. The artist private break-down and disintegration was very aptly shown by T. Eliot in The Waste Land: On Margate Sands, / I can connect / Nothing with nothing.) Be sure to check with local organizations and schools to see if they offer financial help, too. As always, beware of scams; you should never have to pay money to win money. The free grants are probably your best bet.”Un tuareg golpeo a un songhai. El songhai fue a buscar a sus amigos. Asi comenzo la pelea”, segun Ansari.Pink poodle had been to introduce camouflage teddy bears and wing2.0 also no suspense in which, Meanwhile, we also see a lot of the event first appearance of the qualified goods. The design is bold, exaggerated character is still maintained consistent style of the adidas originals x jeremy scott, really,js instinct hi the people once again admire the talent of a devil’s designer jeremy scott.

    If you have a lot of items and a great theme, break them up into several smaller baskets. Ideal basket prices are in the $50-$75 range.Package it right – Keep the theme tight.Pleasing nike blazers women New kind of development Nike air jordan shoes or boots Next has now get the actual physical corner of countless jordan trial boot footwear your own web shop: Webroot Spam Shredder Whitelist and the Webroot Spam Shredder Blacklist features let you identify e-mails that should be immediately delivered or instantly deleted. If you use Outlook, Outlook Express or Eudora for your e-mail, you can effortlessly import your contact list into your Whitelist. When friends, family, colleagues etcetera send you e-mail those letters while be instantly delivered to you inbox and you won have to make any decisions regarding them.

Sign up to receive 1-2 articles per month: