You take out a post-it and write a note, but you don’t have a tool to write with.
So you think can build your own pencil?
Because, you will have to mine the graphite from Ceylon, mix it with clay, cut it into strings and dry it in ovens. Then you will to increase the strength on the lead by treating it with candelilla wax from Mexico, paraffin wax, and hydrogenated natural fats.
After that you will need to saw a California Cedar tree, cut it into small pencil-length slats, wax the slats and kiln-dry it and then, glue the graphite. After that, you would need to shape the pencils into perfect hexagons so they don’t keep rolling over and then and lacquer the pencil to smooth the surface.
You would then need to apply heat to carbon black mixed with resins to make the labelling. And then… Well there are a dozen other tasks that we ran out of space. 🙂
So, do you still think you can build a pencil?
Or, would you rather go to the neighboring stationery shop and buy the pencil, so you can write on your post-it as soon as you can.
Now hold that thought.
You have an idea. You want to get started as fast as you can. You throw this idea around and it sticks. Now you have to build on this idea. You don’t have to build your own servers. You don’t have to develop your own email system. You don’t have the build your own payment system. You just have to focus on your product and leave the rest to existing solutions.
That is what successful businesses did using services like Sendgrid, Braintree etc. Startups like Pinterest, AirBnB, and OpenTable, didn’t have to build these systems from scratch.
Now hold that thought too.
An eCommerce company built its own hosting service, purely for the lack of an existing solution.
Another company built on top of the previous company, scaled rapidly, and now built its own hosting service.
A third company, on a hyper growth stage, shut down its own servers and moved into the first company to continue to scale in business.
Build, buy, or wait, what would you do?
That which is seen, and that which is not seen
In an essay, published in 1850, Notable French Economist, Frédéric Bastiat, explored the consequences of this idea in economics. The underlying lesson is a siren-call for most decisions that you’d be making as a startup.
He illustrates this with an example. When a shopkeeper’s son broke a square of glass from the window, the all too common consolation from spectators was, “it’s a good thing to break the glass, for, if not, how would the glaziers make money”. The theory did not hold good, and is confined only to what is seen.
Alternatively, if the accident would have been prevented, the amount that was to be spent on fixing the glass, would have been spent on the adding another book to his library or replacing old shoes. And the money would have still been spent on which is not seen.
As a business that is building the next best thing, you spend so much time, energy, and capital in hiring developers (who don’t come easy today) and they join your team with dreams of making a mark with their new job.
It only makes sense to employ their skills towards everything that they can build not just your product. After all, it would just be a weekend, or a few weeks. At most. They could hack something, you might think. And they wouldn’t fret. That’s what is seen.
What’s unseen is everything that befalls the task.
Building a peripheral platform will, in fact, make it a second product within your core product – a product that demands resources and thus adds to the overhead. Developing the solution, testing and releasing the application, fixing the bugs, ensuring security and compliance, providing ongoing support and maintenance requires dedicated resources.
“Moreover, when you begin a project, the software that you are “going to build” always looks better than the software someone else already has because you haven’t yet run into the limitations that inevitably show up in software engineering.” – Timothy Campos
In the landscape of businesses evolving and scaling up, business needs will change (and increase) from time to time. And such a growth is complemented by changes in pricing, or by offering more options and features to customers.
Most SaaS businesses opt to build their own internal solutions when they find the available solutions lacking a feature that is important for their business. It could be an integration with a certain application that is crucial to the working of the SaaS company, something that evaluates tax in different geographies, and so on.
“The danger in building internal tools is not that it saves an insignificant amount of cash, but that it stifles future cash”, Josh Pigford, Founder of Baremetrics, has said in his blog. The idea is simple – spend on tools that will generate money in return, than spending it on engineering which may not guarantee a long term impact. Baremetrics has a great ‘Build Vs Buy Calculator’ to evaluate if you want to build your peripheral solution, or buy one.
And if you still decide to build, ask yourself – ‘Is the time right? Is there no existing solution, that does the job better?’
So you have been holding on to a couple of thoughts for some time now.
Amazon’s explosive and dominant growth hosting services with Amazon Web Services (AWS) from being one of the biggest eCommerce giants, is a great story of how build worked over buy.
Today, AWS has become one of the most dominant businesses, supporting powerhouses such as The New York Times, FourSquare, NASA & Netflix. When Amazon started building AWS, it was because of the lack of an available solution.
A fast-growing cloud storage startup, Dropbox relied on AWS for its infrastructure for several, ensuring that its engineers could focus on building the actual product than building an expensive infrastructure.
About two and a half years ago, this changed, with Dropbox moving out of AWS with close to 500 petabytes of data, as they had built their own infrastructure stack, as it would have been much more expensive to host on AWS.
This came on the heels of Netflix doing the opposite, moving about 3 petabytes of its data into AWS, after shutting down its own servers.
Basically, businesses have to ask themselves – at what stage are you willing to build vs buy? Do you have the resources and time to build solutions that are not core to your product, like AWS? Is this the right time to build? Is there no existing solution that will fulfills your requirements?
Ultimately, it boils down to just one thing – and that is, choice. If you are starting out, it clearly makes sense to buy what works and then you can change based on your needs.
by Divya Ganesan
Divya is Product Marketer at Chargebee Inc, an online subscription and recurring billing solutions provider for SaaS and Subscription-based companies.