The Autopsy of a failed web app

It’s time to stop pretending; Cart45, the web app released in July 2008 was not the resounding success I or my employers had hoped for. This post, will undoubtedly burn some bridges, but I hope something can be learnt from it.


Whilst, Cart45 is still available for purchase, I am no longer involved in its development in any way. Thus, I cannot answer questions about it nor offer any kind of support.

Cart45 was built (in theory) to rival Actinic – as an easy to use ecommerce platform that would be as easy to install and operate for a online shopkeeper as WordPress is for a blogger.

The product was built, it had some great features but it didn’t sell like hotcakes… so why not?

1: A lack of features

If you’ve read Getting Real, you’ll know it’s good to launch a product with fewer features than your competition, but sadly our web app was lacking in some really important areas – mostly payment gateway integration. Launching with only Protx/Google Checkout was a big problem and, although I may be wrong, I think it made a lot of people not want to buy an online store that couldn’t process payments with their bank and/or Paypal.

I have to take a share of the blame here; my lack of API integration skills at the time meant that it took a long time to get those APIs working and with no budget for extra developers – that played havoc with the schedule.

2: A lack of time

I wrote Cart45 on my own, in its entirety, within 7 months, with around 2 months spent working on other projects, that is a long time but for a one man development team it wasn’t long enough. For a piece of software where the target market involves people downloading and installing the product upon countless different server setups – there simply wasn’t enough time to cover the test base and get everything right.

By the time the product was finished, we had just about run of time (I suspect, we’d run out of budget too but that was not my area so I can’t confirm that.) Consequently, there was no steam left to power on with the most important aspect of the launch – sales. There wasn’t even time to launch a proper demo version of the site and populate it with products/content.

3: A lack of time management

More important than the aforementioned lack of time was a lack of a proper time/project management. Having never written a full web app before I could only estimate the time it would take. We also lacked a definitive feature set – and as such, time overran whenever someone came up with a new feature or a new way of doing things. Failing to document, at the start, the exact requirements and define a schedule (that couldn’t be interrupted by other projects) was a major flaw in the plan.

4: A lack of budget and resources

As much as I like to work alone, getting things done efficiently without having to worry about fixing other people’s errant code – I can struggle at times. During development there were times where I had to do things I wasn’t 100% confident with: Protx’s API was one such thing but there wasn’t anyone else about I could outsource to and so it took longer and wasn’t done as well.

Now that I work for myself, I always factor into project quotations the cost and time of a subcontractor to perform specialist tasks.

5: A lack of interest from the world

My boss took on the task of promoting the app and did a fantastic job driving traffic to the sales site but after a week or so that traffic trailed off quite sharply and, sadly, never picked up again.

The people we contacted to try and get buzz weren’t overly interested and didn’t share our enthusiasm. Once again, this wasn’t my role so I can’t really state how well oiled our PR machine was but it’s clear that we didn’t gain the necessary trust and therefore exposure in blogs, trade magazines that would be needed to sell to hundreds of people which would have made the venture profitable.

The fact that there was very little marketing budget, meant that sustaining a promotional campaign without paying for advertising* wasn’t possible. I’m not saying that paying for advertising is the only way to market your product – you can do it in other ways but, whether you pay for marketing with your time and effort or with bank notes – it’s never free.

* Disclaimer: We did take out some banner space on some targeted websites which did bring in some good traffic.


I think a lot of people, myself included, think that creating a web app is an easy option. You can code it for free then post a link to it on twitter and suddenly you’ll sell thousands – but it doesn’t work like that. Some web developers with huge blog subscribers can work on an app/ebook and rely upon those followers to spread the word and keep marketing costs down but most people can’t do that.

Also, the success of a web app does not depend solely upon the greatest and/or uniqueness of the idea. It’s how well you apply that idea along with hard work and a little bit of luck that will make it successful.

I hope this has been good a good read and I hope it encourages some wannabe web app developers out there to perform a little bit more professionally through the process of conceiving and launching a new business.

14 responses to “The Autopsy of a failed web app”

  1. Very honest analysis. You’ve encountered some very common pitfalls – time management being the biggest.

    Creating a web app is most definitely not easy. I’ve worked on a number of v. large web projects (banking, National Rail etc.) and they are treated as software development projects – proj. managers, test teams, release teams, Gantt charts, documentation. It can be painfully slow and v. expensive. But I’ve worked on projects that have adopted a more XP (Extreme Programming) approach, and it’s been possible to develop some great stuff, with a small team, quickly.

    I now develop on my own, and still adopt XP methodologies. Time to market is important, so being able to deliver small releases, often is the best approach. There is less risk for everyone.

    The thing is web apps are apps, not web sites. They need to be treated as such. Web apps are becoming more and more common as they replace traditional (installed) software, and agencies need to adjust to this change. Static/brochure web sites are rare, and even a basic site might require custom CMS, which already makes it a web app.

    Anyone who seriously wants to develop a web app, should be prepared to invest in development and testing time. I occasionally get enquiries from people who have an idea and want to develop it – e.g. something to rival BetFair, or a dating website. When I quote them thousands, I never hear back.

    • Great comment Howie.

      I met with some people a few months back who nearly fell off their chair when I told them how long their web app would take to write (properly). From the looks on their faces, they clearly thought this would be a one-man, two week turnaround.

      I think we can take a good lesson from Facebook’s (admittedly, much derided) 5th birthday. Facebook is seen by the press as an overnight success but Mark Zuckerberg is happy to state that this isn’t the case.

  2. I like the candid nature of this post. It’s great that you critique your own work and admit to your weaknesses. That’s the best way to improve and I don’t think enough people are prepared to do that today. It would be interesting to see how the lessons learned here are applied in the future.


    • While I understand that there are times to keep schtum and pretend everything is good to avoid PR disasters. I think in this instance it’s safe to admit: we tried something, it failed, here’s why.

  3. Think two key things come out here, Cart to rival actinic and one man development team. I feel there is now a good knowledge out there of using web apps or building more than a simple site however there is a vast gap between this knowledge and an understanding of what this takes. The fact is a project such as Actinic or any other successful web app either grew over time or had a large development team with an even larger budget for marketing than development.

    Unless you have a very unique project that innovates and sells itself by creating little buzz through blogs and twitter etc then you are looking at competing against market players that have large marketing budgets that the releasing company needs to rival.

    I personally stay away or advise an employer to stay away from clients or projects where this expectation cannot be managed or is not realistic. A lot of people think online is easier because its a lower barrier to entry but there are far too many unrealistic people with unreasonable expectations of costs to launch something with the kind of return to rival the big market players.

    What you said about meeting a client who were suprised about the length of time etc is all too common in my experience and isn’t helped by yes men unprofessional developers who go in with silly promises that at the end of a project they never deliver on and waste a clients money.

    • Ray you’re very right. It’s easy to look at products, like Actinic, with huge sales and seemingly blatant problems (usability being one) and thinking that to usurp its marketshare – all that is needed is a better product but it’s not as simple as that.

  4. Excellent analysis and thanks for sharing it. Too often the web is inundated with “How to…” articles and advice on what to do – often it’s just as educational, if not more, to examine a failure and learn what NOT to do.

    Thanks for your honesty.

  5. I’m in total agreement with Tim. Being critical about your own work is so important. At my workplace I am always trying to critique my own work and get people to be critical of mine and each others work. As for the app, you know you gave it a shot, fair play it hasn’t worked out like you thought but you’ve learn’t a lesson and you’ll move on. Whats next in the pipeline?!?!

    • I did have plans to write and release my own web app but to be honest I haven’t had the time nor funds to be able to do it. Therefore, I’m concentrating on client work.

  6. Another very frank article there Phil!
    In hindsight, perhaps developing with a framework (such as Kohana or Cake) would have given you access to an existing base of potential users – dev time would have been reduced as well.

    Adopting a framework and participating in the forums could have helped out both parties, because it adds to the interest/hype surrounding the chosen framework as more and more people use it, and then people associate all that with your app which relies on it

  7. Good on you for being honest. Tough pil to swallow given the time you put into it.

    I agree with Andrew. Given your resources, I’ve no doubt you would have had more success had you (at the very least) built upon an existing MVC framework like Code Igniter, Cake, etc. It would have allowed you way more time to focus on the real problems associated with module development.

    Another alternative would be to take something like Magento, and churn off a few modules that you could sell on a commercial license basis (as well as positioning yourself as a sole provider of installation for a suite with your module[s] included). I’ve recently paid $500 for a small module for one of my clients to solve what should be a relatively straightforward problem, but would have cost me around £3000 in my time to develop or get developed.

    Glad you had a go though, you’ve no doubt learned a bucket load from the experience that you can either take with you into the next venture, or at least make you be more scrupulous in your preparation and architecture.

    All the best, Dave

    • Dave I think you’ve hit the nail on the head here. A far superior business plan would have been to develop on top of an existing product like magento instead of trying to completely reinvent the wheel.

  8. I knew there was something I forgot to mention…

    Code Igniter’s younger brother, Expression Engine, an open-source core and commercial licensing, is yet to develop (or have developed) a good eCommerce plugin/module, as far as I’m aware. Something like this would have been ideal for you/someone to push through, reap the free marketing and everything that goes along with it.

    cheers, Dave