This is a tale of my time working in management for Australia's biggest crowdfunding platform Pozible.com.
My weekly meetings with the CEO of Pozible were leading up to the same topic a few weeks in a row. Pozible was ticking away, but growth had slowed down and the disruption period for reward based crowdfunding was well and truly over. Equity crowdfunding was showing potential, but we had a team already working on that (see birchal.com). We both realised together that it was time to experiment. we needed something new. It was time for some intrapreneurship, time to get out of the comfort zone of our successful brand.
Instagram stories was just becoming popular around this time. They were bringing out new features every month. You could tell what they were going to build next, just open Snapchat and see what they were doing.... Whether the idea was stolen or not, one feature in particular sparked my interest. I noticed the “swipe up” feature, which allows verified Instagram business pages to link to external websites. This was something very different for Instagram. Before this the only way to link to a webpage was from a profile. The power of posting with links is huge for business. So I started to investigate this feature to see how people were using it.
What I found was brands, mostly smaller craft goods pages, were linking to their shopify pages. Then it hit me.
A lightweight flash sales tool. Something that was crazy simple and aimed at minimising the clicks to make a purchase. After all the user never actually leaves the Instagram app. They use the in-app browser with seemingly the safari engine. The next consideration was the lifespan of Instagram stories. At only 24 hours this would change the tool completely. So I got the team together and we started brainstorming.
Assemble the Troops
It’s important to get the team involved early. Nobody wants to work on something that they have no ownership over. Bringing all the brains together and deciding together is how you make people feel they’re a part of something. They have some say in the direction of the product, and with this they feel more comfortable making decisions on their own. Application are essentially a compilation of all the decisions your team has made, no matter how small. Getting staff to care is incredibly important, because you’re not there for 95% of the small decisions, when people are in a positive mindstate they perform much better. If they’re just coming to work robotically everyday then these ill thought small decisions will stack up and have an insidious effect.
The meeting with the team paid off. We realised the 24 hour window of Instagram stories was an advantage, not a limitation. It made the product centre around urgency. In crowdfunding this is huge. Did you know the start and end of a crowdfunding campaign are the busiest times for pledges? The middle is always slow, hardly anyone donates, the reason is they can do it tomorrow, or next week, or when they finish eating their burger, or when they get home from work. There needs to be strong incentive for people to act NOW.
What we left the meeting with was a flash sales tool with a 24 hour countdown timer. In as few clicks as possible to checkout. Sellers would exhibit their products with a set of Instagram Stories, then link to the check out form. What we were building was not a tool to exhibit products, it was designed for the final phase of the buying cycle. The user simply wanted to pay when they swiped up on an Instagram story. It was going to enable sellers to sell a limited run of a one size fits all item. This eliminated the need for extra forms and menus involving sizing etc.
We decided we’d need to see really strong growth early to follow through with the project. After all it was hugely dependent on Instagram, if they changes their rules or decide to make their own checkout tool we’d be over. So we set a goal. To get 3 launch partners who were excited to use the tool again after their initial sale.
So we put together a specification and some wireframes then sent them to an agency we were working with in need of UX/UI designs.
A couple of days later we had the layout with a few different colour options and we were ready to start building.
We got together again, made the final product design decisions, then started creating and sizing user stories. One big question we had though was what options we had for collecting payments, we assigned this a maximum story point value and treated it like a spike. After all we were on 1 week long sprints so 5 working days was a good amount of time to find answers to questions we had. The other thing we agreed on was that as it was an MVP, that hadn’t proven itself yet, we weren’t going to spend any time building an administration area. Sellers would need to fill in a form and we would paste the details into the database directly. This was fine for our goal of having just 3 launch partners.
As the marketing team worked on a pitch deck and researched the Pozible database for potential launch partners. The engineerings researched the best payment providers that worked in the Instagram in-app browser.
Unfortunately, we learned that apple and google pay were out the question. At the time, they weren’t supported by instagram’s in-app browser.. so we opted for Stripe. They have a lightweight and flexible payment solution called Stripe React Elements. Which offers a React component that’s super easy to style. Stripe also has an in-built email receipt function, saving us precious time for this MVP experiment. Along with this you need a backend to handle the request, which we ended up hosting for free on google app engine.
As the week came to an end we almost had a marketable product, but we needed to do some solid testing and finalise things with the brands. Plus we hadn’t chosen a brand name yet. We wanted pozible FOMO initially but found it hard to get a suitable domain name. Plus we wanted to avoid confusion between our products. This is when our CEO overheard us talking. He had a project from years ago that he’s spent way to much money buying a domain name for. He also had a logo ready to go. Why not right? It was an MVP after all. So we jumped on it and kept moving forward. From that day it was called Billycart.com.
This is the beauty of open plan offices, there’s a lot of communication that is collected ‘osmotically’. Unintended snippets of information can make all the difference sometimes. We would have never had this breakthrough if the CEO wasn’t in earshot.
By the weeks end we had a product that from the sellers perspective, looked like a well finished marketable product. We also had some launching partners ready to go! With their product details already added to the database and unique URLs ready to go.
We coordinated with the launch partners, scheduled the sales and watched with a careful eye.
The Second Iteration
The marketing team kept researching and onboarding new clients. They could setup a dummy page before even speaking to the brands and we could Instagram message pages with the link. This saved us a lot of explaining.. Meanwhile the developers worked on a landing page for sellers. But really what this iteration was all about data collection, user feedback and analysis.
We set up a slack notification for every sale made through the app, as the slack notifications came through we’d all make eye contact gleefully, it was great for team morale!
We learned pretty quickly that the tool worked best for items with a lower price point. It was essentially an impulse buying tool and people will spend under $10 impulsively. We also learned that because it was a sales tool, we needed to to the UI that the item was discounted for a limited time. We went ahead and mapped out an “original price” display while also making the remaining stock more prominent on the sales page also adding the original price.
It was making sales, nothing huge, but sales none-the-less. Usually under 20 units per post. We continued to experiment and tweak things but we reached the point were we had to make a tough choice.
The idea had a few issues. The first was that people wanted custom fields… We battled with it but people were asking daily, the problem is that our only point of difference was a super low clicks to purchase.
The items the tool did work for was impulse buying, but stores with for impulse products is very niche. Most seller have an online shop, where users head to with the intentions to purchase. Sellers didn’t quite get their head around exhibiting the item first, then asking users to swipe up. Most sellers would stuff a product in front of their fans faces without inspiring them to buy. Instagram is about giving value through content, not ads. So users shouldn’t post product all that often.
The tool may have worked well for fidget spinners (as an example), but it’s a weird corner of the internet and very short term. Essentially people also just couldn’t be bothered typing their credit card details in. The price became irrelevant, the effort was too high.. It may have been different with Apple pay who knows.
Either way, as we guessed, Instagram brought out an inbuilt feature to sell products from stories just after us. Which was the final nail in the coffin.
We had a small funeral for Billycart, and by that I mean we played ping pong a drank beer like we did every Friday. We spoke about the problems and lessons and moved back to working on our core product.
In the world of startups there is no failures, only lessons. It’s important to keep a growth mindset and continue to try new things. It’s about focusing on the process, not the outcome...
The biggest benefit of all was learning together as a team. The basis of a friendship is shared experiences and shared values. Really makeing a not of every stage formalizes this for a team and improves the way people work together moving forward.