In 2013 YCombinator co-founder Paul Graham famously wrote that one of the most common types of advice he gives to start-ups is to "do things that don’t scale".

At the core of Graham's principle is the belief that when you start out, adopting a handmade approach to all parts of the business gives you an opportunity to learn and develop an understanding about your customers and products that are difficult to achieve later on. Indeed the learnings you make in the early stages of your business compound as you begin to properly scale and start to automate, hire or outsource them. This ultimately drives greater growth later and prevents costly mistakes.

There's also the reality that as a founder you're going to have to do a lot of things that you may feel is below your pay-grade to get some early momentum. Any founder who thinks they are above taking on repetitive tasks is going to miss some significant product and customer insights.

The much-quoted story of Airbnb's Brian Chesky and Joe Gebbia going door to door taking photographs to improve the quality of their listings on the website, or of Pinterest's Ben Silbermann recruiting his first users at design conferences, are just two legendary examples of founders that weren't afraid to get their hands dirty.

Speaking to people in the street

When it comes to acquiring your first customers, adopting the "do the things that don’t scale" mindset often means speaking to your customers directly.

At my last start-up, Hoop (a marketplace for family activities), we acquired our first customers by standing on our local street in London, handing out branded balloons to families to canvas their thoughts on the app we'd just built. A few weeks later we had 3,000 customers and a marketing pitch refined to perfection. Nothing will improve your understanding of what marketing messages work better than trying to convince an exhausted parent to download an app while dealing with a grumpy toddler!

The insights from speaking to thousands of families set the foundation for our comms and product marketing strategy for years to come, utilizing parent feedback to refine our PR and paid acquisition campaigns. We could never have scaled our business speaking to parents on a one-to-one basis to secure a new customer, but these early conversations gave us the insights that allowed us to optimist our acquisition strategy at scale. Four years later we had 1.5M customers and had twice been picked as one of Apple's 10 Best apps of the year.

In my experience it is best to approach the philosophy of "doing things that don't scale" as a mindset to engage the early problems of building a business. Starting out, you're likely to have more product ideas than you can ship. Adopting the "do things that don’t scale" mindset forces you to manually hack near-term solutions and learn the right path to build a winning product before you commit significant engineering resources.

Computers are good at repetitive tasks

As a developer-first company, we believe technology can solve most problems, not least automating repetitive tasks. We've already began to build a number of internal tools to streamline the day to day operations of the newsletter.

The Google Apps Scripts we've written makes the editorial process more efficient, but there remain a range of tasks we continue to do manually to track the breadth of new tools and beta programs available for developers. We have a lot of Google Sheets where we sort and filter a range of data sources to find all the companies operating in a particular sector, making announcements, raising money, releasing software, being acquired, running events...

Undoubtedly, we'll be able to automate some of these processes but for now there is value in being close to sources we monitor to understand which are most useful.

But, how does it scale?

At Console, we're making an effort to speak to our subscribers directly. Subscribers have been replying to the newsletter directly, which goes right into our inbox. "No reply" addresses are an unnecessary barrier to getting feedback, and several of the tools we've featured have come directly from inbound recommendations. Sometimes these are from the vendor, but often they are from the end-user who just likes the tool they've picked.

In the coming weeks will be personally reaching out to our top subscribers to say thank you. We know that to do this forever will be unsustainable, but the direct feedback about what our subscribers like and don’t like will help shape the product going forward.

There will be a stage later this year where it becomes no longer practical to adopt such a hands-on approach to both the day-to-day running of Console and how we invest our time as a founders. Our editorial process currently involves just us, but it is essentially practice for when we have more people contributing and pitching ideas.

We're not a venture backed company so having invested our personal funds to bootstrap gives us some natural constraints on making sure we focus on the right things. This means we're never going to have a huge team, but we expect to hire more team members to pick up the slack. This will including building more tools to streamline the processes we see being core to our business. We currently hand-write the Markdown for all the website content, but at some point we'll need a simple CMS!

Indeed, there’s undoubtedly a contradiction in doing things you know you cannot do forever as a strategy to grow, but in my experience these early learnings set a foundation for future growth that is difficult to replicate when you're moving at pace and start to reach meaningful scale. For now, I value the time cutting pasting data into spreadsheets staying close to the research, and enjoy the direct conversations we're having with our subscribers, and thanking them for joining the community.