As a Co Founder and Product Lead at a Start Up like Fantastec, one can appreciate that producing more while costing less was a primary business goal. While there were many tools and software plugins available for a tech company to subscribe to, they weren’t always a perfect fit for our new and unique NFT business. And they certainly weren’t cheap for the volume and scale we were audaciously expanding towards. So like all innovative problem solving personalities, I sought to challenge the developer teams through mini sprints and hackathons to build our own software systems to support business operations. They were not as pretty as a commercial SaaS product might be, but they got the job done and saved us hundreds of thousands of pounds. Here are four of them.

NFT Content Preference Testing

Situation

As a football NFT (collectibles) product offering, every month and every season we had to test our content ideas and design ideas for the next ‘collection’ we were thinking of making. That meant we had to learn from our users what they liked and didn’t like before committing time and resources to producing final designs to mint.

Problem

To use software survey providers like Usability Hub cost thousands because of the the amount of content data we needed to have hosted and the amount of user testers we needed to get results that were statistically significant.

Solution

Using a cheaper hosting site called Elastic Beanstalk, we were able to build internally our own ‘preference testing’ system. Innovatively it cycled through all our content options (rather like Mark Zukerbergs’s infamous ‘Face Smash’ from his pre The Facebook Harvard days) so that our testers kept choosing between option A and option B, to eventually be comparing between all first choices cycling down towards a final top 10 content winners from a previous spread of hundreds.

Benefit

Using this system we built internally, using the usual product development MVP approach, we could be 99.9% confident that the content styles and creatives we were going to mint into NFTs were going to be popular and present the in-app engagement stats and sales. The software also saved us £30K a year in third party subscriptions for software that didn’t provide the same granular analysis.

UX/UI Prototype Testing - Video capture technology

Situation

Fundamentally the role of the ‘Product Team’ is to de-risk any features making it onto the developer backlog that could fail to deliver against product KPI’s and business objectives. So we tested everything in terms of problem solving, value and feature usability. One essential test was high fidelity prototypes UX/UI tests - i.e Could people use the app and new feature intuitively. We had an 80% minimum pass rate requirement.

Problem

To host tests on specialist software platforms at the frequency that we needed to test our features was hugely expensive. We estimated that it would cost us between £55k to £75K each year and that might only increase. So we needed another solution.

Solution

In Summer 2020 we decided to build a simple platform we called ‘capture tech’ in Python that could display either images or import interactive high fidelity prototypes from Figma. It also allowed us to onboard testers, who could be paid, and guide them sequentially through a series of tests. The platform was able to record their interactions with the prototypes so we could see what they were doing, and also record audio so testers could articulate or express their views and describe problems they are having. All recordings were stored to a database for the product team to review and score in order to conclude the feature reached the pass score of 80% or refinements were needed.

Benefit

Our ‘Capture Tech’ product was hugely successful. After a few teething problems it became a regular part of our product development process for the mobile and web app, and also for more simple preference testing. Its flexibility and video capture capability saved the company £65K every year and was a home grown tool our department was extremely proud of.

Rewards Ladder Gamification - Dopamine Parameter Testing

Situation

In 2020 we wanted to introduce a Rewards feature to the Fantastec SWAP app which could earn users an array of benefits like free or special edition NFTs, merchandise or entry to competitions for VIP football related experiences.

Problem

The key to a Rewards feature is finding the sweet spot for a user to be tempted to ‘buy’ more NFTs because there was a chance he’d get enough points from their rarity to move them up the ladder, earn another free pack which creates a flywheel effect and get’s them even closer to the prizes and experiences they really want. Building the feature in the app and missing the sweet spot could lead to lots of developer tweaks which would take time, cost resource and could play negatively to our users. We needed a way to find the right points threshold for the rungs of the ladder, and the right potential points to be earned in free packs that gave a dopamine hit, encouraging users to keep buying, collecting and climbing, especially if they were new to the SWAP app.

Solution

We built a test model with python that was linked to our NFT database. It allowed us to adjust the parameters of ‘points’ assigned to an NFTs rarity, the average number of points that could be received for NFTs in a free pack, the amount of coins needed to buy a pack, and the points threshold needed to be reached to climb the Rewards & Prizes ladder. Once tested internally we invited 100 SWAP users to test the model, giving them 2500 coins. Coins could also be earned by selling NFTs. Over a period of 4 weeks we held four sessions with the aim of getting as many to climb the ladder as possible.

Benefits

Each week we were able to analyse the progress of users and get feedback. We then adjusted points per NFT, average NFT points per pack, points ladder thresholds to find a sweet spot where 80% of user testers reached Achievement Level 7 out of 20, or higher with Level 1 being the highest. With this knowledge we could set the parameters for the developer team to build into the SWAP app, and within 4 weeks of launching the rewards feature we had a 17% uplift in pack sales and a 34% increase in user engagement on the app through the Rewards feature.

Fantastec SWAP CMS

Fantastec SWAP Content management System

Fantastec SWAP was a pioneering NFT platform that had a complex integration of tradition web2 components like graphic images and content that lived in Amazon cloud servers, web3 Blockchain structures where NFT metadata was stored and assigned to users, and payment systems that handled fiat and crypto currencies.

This meant we needed a user friendly CMS (Content Management System) to allow different stakeholder teams such as production or marketing to be able to set up clubs, players and teams, and to create collections, mint collections, investigate users, user purchases, set up Rewards features and so on.

The CMS was a product epic that required a disciplined production management approach in order to provide the features and functions necessary to efficiently operate the business and provide fast accurate data for customer support.

Unfortunately it was rare that I took screen grabs so the full extent and grandeur of our CMS cannot be fully appreciated. None the less it was a hugely unique system which, through upgrades, in 2023 operated significantly faster in the way we could add and create collections, manage the online store, and seamlessly assymilate the hundreds of athletes and athlete details from the US colleges and college teams we were now incorporating into the platform. As head of product working with the developer team I learnt through experience and implementation a huge amount about how important these C.M. systems are and what it takes to make them one of the most valuable and cherished assets of a company.