If you think that scaling is one of the main issues you need to focus on while building your new disruptive web application…well think twice. It is quite sure you won’t have anything near to a mass adoption anytime soon. Instead you will have a quite slow user-curve acceptance and maybe some peak. But focusing on the architecture and scaling possibilities of your app for millions of users is just plain dumb. Millions is quite a lot. By now you probably will struggle to arrive to one thousand users. Probably less if they are (paying) customers.
Concentrate on marketing. Concentrate on user experience. Concentrate on marketing (did I say that only twice?).
Thinking about scaling is an easy task for technical ninjas like you are. But that’s not the big part of starting up something. The hard and tough part is:
- creating something people would like to use
- because you are solving a problem (that they know or don’t know having)
- and spreading the word, marketing your product in-and-out the web, so that people know you exist
- and they can buy your stuff (that’s what advertising is made for since the beginning of time)
So, talking about scaling is just fear and resistance to change your tech cap with a marketing cap, because you have no idea how to market your product. Really.
That’s the reason why you have to love what you are building. Love and believe, so you can start talking to people telling them not what your wonderful web app can do, but what they can gain using it or what the app can do for them and how wonderful their life will be after.
Because lots of startup die for not having customers enough (or at all!), but I haven’t heard of any died because at first was not able to scale millions of users.
UPDATE
I posted a long comment on Highscalability blog, in response to a post discussing about how and when and why to think about scaling. There I also talk about the difference between users and customers and why it is important to you.
I think startups that think about how to scale prematurely are simply lingering in their comfort zone, defiantly unwilling to tackle these bothersome things called “customer development” and “marketing.” Good post, I think it hits the nail right on the head for the startups that have priority issues.
P.S. JuicyCampus was one startup that failed for unmanaged growth.
Pingback: Tweets that mention Startups die for not having customers, so STOP thinking about how to scale | Wild Bunch -- Topsy.com
Hi Alvin, thanks. I think you are right when you say that the first priorities are “customer development” and “marketing”. Probably there should be a startup movement named “Go out of your comfort zone!” like the 37signals “Getting Real“.
I looked at JuicyCampus and in their faq they state:
“JuicyCampus’ growth outpaced our ability to muster the resources needed to survive the economic downturn and the current level of revenue generated is simply not sufficient to keep the site alive.”
and
Online ad spend dropped across the board as the economy took a turn for the worse. Our advertisers have spent less, but have remained loyal[...]
I’m not sure was all about scaling. In the sense that: yes if their infrastructure could sustain the weight of the new users without increase in spending…well…yeah probably they were still there, but it seems to me that their business model based on adv revenue was probably not able to generate a consistent amount of money (as of FB or Twitter, but JuicyCampus was probably not VC funded), because at a certain point they could have stopped for a while adding new Campus to the site.
I’m just guessing, but probably, in the middle of the crisis, the revenue, only ads based, just went down.
By the way, that’s the problem having users and not having customers.
Like all things in life it is about balance. Focusing 100% on a single aspect of a startup is a great recipe for failure. You have to keep scale in mind when building an app but it shouldn’t be all consuming. There are many challenges along the way and being nimble enough to address the issue at hand is key. I think the reason we read about scale so often is that it is easy to talk about. It is a concrete…tangible part of business that can be easily communicated. Marketing and driving traffic is somewhat abstract.
Pingback: STOP thinking about how to scale, Startups die for not having customers. | Apple Ipad Secrets
Friendster! They had plenty of users. They very much died because they didn’t scale.
+1
No, they died because their site sucked.
Friendster died because of spam. People who could tolerate the spam were increasingly influenced by their peers who have migrated to Facebook, and all was left to do was for Metcalfe’s Law to slowly decimate Friendster’s traffic.
Friendster died because it became irrelevant. I don’t recall scale/capacity ever being the issue, rather than it simply losing relevance.
Friendster hit major scale issues. Here’s a mention from CNET ca 2006:
“It’s possible that there’s some larger reason for the outage, like that they reached the limits of scale,” he observes, citing the example of Friendster, which previously was the dominant social-networking site and had occasionally crashed due to excessive traffic.
I see this a lot in the blogosphere and it does make sense. But if you’re designing an application and not at least pausing for a moment from time to time and asking yourself, “If I had to scale this aspect out, what would I do?” and letting your answer inform your design, aren’t you shooting yourself in the foot? Do you really want to build a successful business by concentrating on marketing and UX and then find that you have to re-write it because you can’t handle your traffic? I agree you may not want to build the ability to scale in front the start but shouldn’t you at least be thinking about it when you’re designing the application?
We struggled with scaling as we brought on customers. There is a fine balance especially with enterprise applications. If you do not tackle scaling early enough you make mistakes in application architecture which will really bite you in the butt and be very costly. If you did not design your app carefully and have to scale quickly due to new customers, then your business can go up in smoke just as easily as not having enough customers.
Plan for the worst and hope for the best but only scale as much as you have to.
Tackling the right thing at the right time – and nobody knows either for sure – is necessary to succeed.
Over-developing on any front – scale, reach, user-experience, customer-adoption – is an expense and effort which cannot be justified at that point in time.
If you believe in the product that you have developed – go out and get more potential users and try to sell to more potential customers as much as possible.
How the product and business will evolve at a later stage is something you can only take a guess at – but spending more on any one aspect without properly seeing/sensing users/customers need for it is a very big gamble which more often than not, doesn’t pay off…
Pingback: Two questions « Rubber Tyres –> Smooth Rides
Good page load times require good site architecture, that may include scalability. Also simple things tend to scale better – so scalability is sometimes a good protection against unnecessary features. Scalability isn’t hard – it’s all about making right choices in the beginning.
Title of article could be also: “Startups die for not having customers, so STOP thinking about how to do trick Google” or any other internal or short-term aspect.
Pingback: Twitted by alexatkeplar
Pingback: One Million users is a lot of people, more than you think | Wild Bunch
Pingback: Twitted by startupangel
These are all very good points and I think you are spot on, developers and marketers just need to have the realization that there is a hidden danger to “reacting” to scale.
It’s awesome if you get a lot of big clients, but if you end up under delivering or god forbid not delivering then the bad publicity or reputation will kill you fast.
The marketing section needs to be aware of the scaling limitations of the application, otherwise they will sell something you can’t deliver and all your engineers will work round the clock trying to make something happen.
This is a wonderful post and may be one that you should followed up to see how things go
A pal e mailed this link the other day and I’m desperately looking your next post. Carry on on the impressive work.
Discovered your site via google the other day and absolutely like it. Carry on the truly great work.
I dont think abandoning scalability is a good idea – but there should be more emphasis on marketing as you mentioned.
It really depends on the type of application being developed though. Refactoring things like ‘social networking’ apps for better efficiency can be a huge project, so its better to start with some scaling built in.
Its good for you to presently now find a location where the blogger knows certainly well about his subject.