How I Built and Launched Find a Delivery

A couple of months ago, I launched my side project - findadelivery.com.

This was special for me because I have many ideas, but I often don’t follow them through to completion. In fact, most ideas don’t make it past my notes app.

This time I followed through and managed to amuse even myself.

“So Karan, how did you pull this off?!” I hear you not asking.

Story time.

Conception

Not too long ago, supermarkets looked like this in the UK.

Should you turn to online delivery - things weren’t looking too good either. Slots were booked weeks in advance.

So, selfishly I decided to write some scrappy code that would monitor Amazon Prime Now (my online supermarket of choice) for delivery slots. It automatically checked their page every 2 minutes in the background until it found an opening, at which point it would log a message in my terminal like this and send me an email.

It’s not the most elegant thing in the world, but it does its job.

Pleased with myself, I decided to tweet it out to my fictional Twitter audience who find me funny and smart.

Since Amazon Prime Now release their grocery delivery slots randomly throughout the day, I wrote a script that scrapes the checkout page every 5 mins and sends me an email whenever a slot opens up. 👨‍💻 pic.twitter.com/DTcugQxcOF

— Karan Navani (@TrappedInKaran) April 5, 2020

I got some validation in the form of likes, but also I got a few responses of people wanting this for themselves. So, with my sample size of 3 people, I decided that it was worthwhile to pursue this idea further and build something out of it.

I decided to create a service that would let people enter their postcode, select the supermarket they wanted to be notified about, and then receive an email when slots became available.

Building

Alright, I’ll have a go at it this weekend to put it online. Will let you know :)

— Karan Navani (@TrappedInKaran) April 9, 2020

It took me 14 weekends.

Before this point, I had only hacked together an inelegant solution for myself that worked with some fiddling around. Now, I had to make this easy to use for anybody and cover the scenarios that come with building something that can be used by >1 person, on the internet.

As I later found out, I had completely underestimated how challenging this would prove to be and how much I would have to learn in the process.

I had to figure out everything from maintaining a queue of searches from different users, to being able to feed back to the users that their search had failed because the supermarket does not deliver to their address. All this, while maintaining a legible user experience, was an adventure.

Moving fast

I planned to build an MVP (Minimum Viable Product), which basically meant - what’s the most basic thing I can create that can get the job done? But the to-do list of things that needed to get done to even be in this basic stage was so long.

Eventually, I created a basic v1 of the app and got my friends to try out the underlying tech.

I did this so that I could keep myself motivated by sharing something concrete with others.

The next step was to polish it up and make it launch-ready. I kept thinking “I’m almost there”, and would keep finding there was much more to do.

---

Side note: In retrospect, I think I should have time-boxed each feature that I was trying to build because I ended up spending weeks working towards dead-ends. With that being said, I also spent time working on things that turned out to be non-essential, as I learned from speaking to users. I coldly axed those pieces from the product. It is critical that you do not dwell over sunk cost if it hampers the usability of your product.

---

At this point, a couple of friends (Kane and Curtis) decided to jump on board to help me out and built some critical pieces that I needed to get the app launch-ready. This provided me with much-needed impetus to push forward.

The only thing missing was a deadline.

"Work expands so as to fill the time available for its completion"

– Parkinson's Law

The best antidote to this is to have a deadline that you cannot miss. I created this deadline by scheduling a slot to demo what I had built to my colleagues at work. The thought of failing in front of 40 people was enough to have something ready and working.

This call got postponed, unfortunately, but at this point, I was itching to get this in people’s hands and I decided to launch on that weekend.

Launch

Finally, it was launch day, I had knots in my stomach, but the website looked like it could hold up, and I was ready to push this in the world. It was time for the world to see my baby. In my mind, this app would revolutionise online grocery shopping in the UK (blame The Social Network).

3…2…1…publish to all platforms!

Comments and likes started to flow in and the support was incredible! However, the metric that mattered - active users, wasn’t where I expected it to be. I feel like I had mismanaged my expectations about how many people I could get to sign up to the app and actually use it. In the first couple of days, I had ~35 users, which was fine but it was nowhere close to the lofty numbers I had in mind.

I realised that I needed to better target people who needed a service like this because most of my network was comprised of people my age who probably were ok with queuing and going to supermarkets in person.

Finding the right channels

The most effective channel I found was Reddit.

I targeted threads of people sharing similar services and tried to convince them that I had a superior (aka free) offering.

Posting in relevant subreddits was also a successful strategy.

This got me lots of great feedback and added ~30 users overnight.

A happy plateau

After sharing online and some other word of mouth promotion, I managed to get ~65 users on the service. I could keep going and target more specific groups and potentially reach out to the press but I realised that I was no longer having fun doing this. It made me realise that I needed a partner in the future who would be able to help me out with this aspect of launching a product online.

At the same time, lockdown measures were starting to ease and supermarkets were managing their stock better. I was driven by the immediacy of the service at the time I was building it, but that ship was pulling out of the harbour by the time I was done building it. That’s not to say that I failed, because I had no idea how to build a service that others could use when I started this project, and a few months later I had put something into the world that existed because I typed some gibberish on a computer. That means a lot to me.

Hey Karan,

I was on the all hands meeting we had earlier today and absolutely loved your work on your website.

Just dropping you an email to say my Mum lives in Cornwall and has struggled the whole way through Covid-19 to get a delivery slot.

I sent her the website and your fantastic idea has sorted her out a slot for Sunday which she is delighted with!

Thank you very much for this, it’s a great idea.

– A colleague from work

What’s next?

For now, the service is live and completely free to use. I don’t plan on promoting or growing it all that much unless I see a pressing need for it again, like...a second lockdown. I do, however, intend to start working on another side project after a brief break. If you’re working on something interesting, reach out to me!

Karan