It’s been a long time in the making and so we’re excited to be ushering in the next phase of the people.io app marketplace. If you’re curious about the how, what and why of how the new marketplace came about — read on.
As many of you will know from our weekly updates, this has been a tricky and lengthy process, hindered by many of the typical challenges any startup faces including time, money and resources i.e. the people who can make it happen.
As you’d hopefully expect though — everyone in our team has been working hard to make this happen and continues to put their energy into the growth and evolution of our marketplace.
As a personal data startup, our focus is to help people realise the value of their data. Importantly though, we don’t believe value is purely monetary and that value can be achieved in any number of different forms, where often those forms and their perceived value will differ from person to person depending on factors such as life stage.
Version 1 of our marketplace was designed to provide a simple way of demonstrating ‘quick win’ value to users, to prove the point that taking control of their data can be rewarding. Our original plan was to then quickly introduce new product features and services that would then also enable our users to realise the value in other ways — creating a balance of value types.
We were incredibly excited by how well received our app was (thanks to you!) and our user base grew incredibly fast — too fast in-fact. That unfortunately meant we were so focused on scaling our platform and technology that we simply didn’t have the time or resources to put into the other demonstrations of value. Fast forward a few months and we were simply caught in a catch 22 where we couldn’t sustain the growth with the same model and tech.
Now, after months of internal preparation (thinking, hiring, building) we’ve been able to launch the next version of our marketplace. While on the surface it might seem similar or even the same, behind the scenes we’ve built a number of different advancements that will enable us to grow and evolve the proposition in the ways we had originally intended to.
We’ve already got lots of ideas in the pipeline focused on how we can demonstrate value to our users. They’re on our product roadmap and some are even in development already.
This marketplace relaunch is the start, not the end, and we hope you’ll continue this journey with us. In the meantime, here’s a few specific updates around the most frequently asked questions:
Why is the marketplace more expensive?
When our app first launched we kept it simple and only had gift cards with a fixed credit to £ ratio. That meant some of our early users were able to redeem 1500 credits for £10.
With that first version of the marketplace we’ve already paid out over £180,000 of items to users (which is more than we’ve actually earned in revenue). As a business, that’s just not sustainable, and so to be in with any chance of rewarding our users far into the future we had to rethink how we manage our marketplace, its stock and the pricing models surrounding it.
We’re doing our best to work with brands directly so that we can introduce gift cards at the best possible credit price, but our ability to do that is largely dependent on those partnerships as we’d quickly hit a wall if we keep funding that all ourselves.
What’s next for the marketplace?
The relaunch of the new marketplace is a start, but it’s definitely not the end. As with any feature, it’s a work-in-progress, and we’ll be working hard to give you a better, fairer marketplace experience in the weeks and months to come.
Why did things keep changing over the last few months?
When we first took the marketplace down, we didn’t know what would happen to it. We wanted to believe that we could give people the same value as before, and said on record that a £10 gift card would still be worth 1500 credits.
As we started to build out the new functionality and look at our relaunch plans, it became clear that we simply couldn’t do that without the right brand partnerships. We should have clearly stated that we would have to raise prices for the majority of users until we had these partnerships, and for that we are sorry.
What’s behind this new account health scoring?
You might recall us talking about some of the interesting challenges we’ve had over the last 12 months regarding fake users or spam accounts, with a number of registered users attempting to game the app for their own personal gain, with little regard for the impact this might have on our ability to reward other users.
Our approach to resolving this was the introduction of a user account health score that aims to identify users whose profiles have given off signals that, based on other known spam users, suggest there may be reason for concern with the account. This process isn’t infallible and given that some of the largest tech companies in the world struggle with this very same problem, we know we have a long way to go!
So, how does this mythical account health score work?
Once a day we pull together the various information you’ve provided, either through responding to questions or integrating various data sources into the app. We then run a multi-stage process over this collection of information in order to generate your account’s daily health score. This health score is the result of three categories of metrics, each of which is weighted according to its type, form and importance. These sections can generally be described as reliability, engagement and integration metrics, outlined below.
1. Reliability
We realise that under ordinary circumstances, it is extremely difficult, even impossible, to figure out whether users’ answers are reliable, due to the nature of online questionnaires. To help with this problem, we use techniques often employed in surveying and other anomaly detection problems, such as asking the same question differently several times within a very short amount of time, using machine learning methods to spot repeated patterns in your answers that are highly unlikely to occur by chance, and similar metrics to try and spot unusual activity. This category is currently highly weighted as it represents an important measure of not only capturing potential misuse of the app, but also helping you to build the most accurate, and therefore, valuable, digital representation of ‘you’.
2. Engagement
We want to ensure we are rewarding the users who have made it this far with us and engage with us frequently in various ways, such as visiting the app regularly or checking out their matches. We completely understand that during the time our marketplace was down, you might not have kept up on our update videos or answering questions. We also wouldn’t expect you to visit us every day, especially if no new content is available, and so frequent visits and engagements will be all it takes. This currently has low weighting, but will increase over time as more possibilities for engagement are added.
3. Account integrations
In order to get the most of the value of your data and complement your journey with people.io, you can enrich your profile by connecting one or more of the current possible integrations: location, Gmail or Spotify. We realise you may not be using all of those or simply not feel comfortable sharing them with us right away, but this won’t prevent you from reaching our top tier as it will be easily balanced by other metrics. For instance, as most of us are using smartphones to access our app, which have GPS, location will be given more weight than the rest.
We’re always working to improve the product. With that said, our account health scoring model is currently in its early days and we will be iterating, reviewing and improving it over time. As always, all feedback is welcome on every aspect of what we are developing.
