7 ways to start small

If you’ve been thinking about testing an idea you’ve had, pick a path and take a few steps. You won’t regret starting something.

November 30, 2020

I’ve always been a big fan of iteration. It’s a valuable exercise to try to solve a problem for someone with the minimum level of initial effort. The role of failure has been argued to death, but the merits of having an idea and testing it quickly are solid. You usually don’t know where you’re going to end up, even if you think you do. So it’s worthwhile to pick a direction, take one or two steps, and then evaluate your situation to decide what’s next.

Image for post

With TeamKit we’ve been working on a more ambitious project — a new way for team managers to build and run remote teams. We’re sprinting forward with that mission, but we knew we also wanted to try out some other quick experiments in parallel. One of these is #WorkingRemoteToday, a web app we built in a day of work to test a theory we had about team members sharing their remote working habits. #WorkingRemoteToday is an ultra-simple app that lets you build and share a tweet about your day out of the office.

Building something at all may sound like too much of a challenge for you. Or maybe you’re a habitual over-builder (guilty as charged). Read on for ways to start small in your own experiments.

1. Minimize scope

Stick to one screen or page, maybe two. Try to provide value with one web form or activity and one result or shareable report for the first iteration. In the case of #WorkingRemoteToday, the result of our single form is a tweet that can be shared. There’s no user account system, no preferences, no history, nothing fancy. All that can come later if it’s needed.

2. Don’t start from scratch

Consider using an existing tool as a starting point. Can you solve a problem for someone by creating and publishing a Google Form? Try it. If coding is not your forte, don’t worry. The #NoCode and low-code trends have gained momentum recently, and there are lots of useful tools to get started. If you want some help on a quick experiment, try Fiverr or Upwork.

3. Minimize server needs

Consider going serverless. Sidestepping the misnomer debate, services like AWS Lambda, Google Cloud Functions, or Azure Functions mean you can deploy tiny bits of code to handle your server needs without spinning up a full-blown container instance. We used AWS Lambda recently for a video processing task on TeamKit, without requiring us to setup and manage another instance.

4. Stateless

Even better, go stateless. Or have someone else store state for you. Like Twitter. Or LinkedIn. With #WorkingRemoteToday, our only state is managed Twitter in the form of tweets. That made for a quick development process. We can always add a database for increasing the state capabilities later.

Image for post

5. Debt is a tool

Overbuilders, this one’s for you. Engineers are conditioned to design things in the most proper way practical, and to avoid technical debt. But debt is a powerful tool that has made many new innovations possible. Taking on some technical debt for your small project is often worth it for the increase in speed. We spent a day working on #WorkingRemoteToday, in the midst of other work. Avoiding our natural tendencies to strive for perfection allowed us to ship something quickly. When starting small, it’s OK to embrace technical debt.

6. Integrate

Build an integration with a new feature on your favorite platform. This approach can give you useful constraints that limit the scope, a quicker alternative to starting from scratch, and an existing audience of users who can easily try what you’ve built. The awesome folks at OneBar inspired this. After Slack released the new App Home Tab, OneBar started thinking about how to make use of it. The result was ToDoBot. Give it a try if you’re in Slack with your team. You can also read more about their experience.

7. Observe & react

Watch the results of your experiment and react accordingly. This goes along with the first tip. How are people using your simple creation? Be prepared to be surprised. Then decide: do you need a new feature, a new integration, a bug fix? That decision process is a huge topic, but there are simple approaches that can help.

Summary

If you’ve been thinking about testing an idea you’ve had, pick a path and take a few steps. You won’t regret starting something. Are you doing it from outside the office? Use #WorkingRemoteToday to share your experience.

Share this post

Enjoyed it? Found it useful? Spread the word...

Take the Remote Hiring Survey

A quick survey for team managers who recently hired remote team members (or plan to do so). Your answers help us shape and refine TeamKit.

Recent posts

Sharing a sense of place

Sharing a sense of place

When your team can't be together, try sharing a virtual tour to get a common sense of place

Meeting agenda templates: how to run your team meetings

Meeting agenda templates: how to run your team meetings

December 18, 2020

Looking for a starting point for how to run team meetings of various types? This is your list.

Holiday icebreaker questions

Holiday icebreaker questions

December 15, 2020

Holiday ice breakers for your team

Christmas icebreaker questions

Christmas icebreaker questions

December 15, 2020

Festive Christmas themed ice breakers

Icebreaker questions

Icebreaker questions

December 15, 2020

Questions to ask in your next team meeting, or use with a team building activity