Launching a rewards network for people under 30 that raised $60M in one year

As VP Design at Verve, I oversaw a lot of product elements, comms and users. Verve itself as a Saas is a complex beast - it allows you to sell event tickets to your friends to get rewards. This sounds simple enough, but with events spanning from sports to travel to festivals, with ticket integrations like Ticketmaster, See Tickets and more - and multinational e-commerce - US, UK, Germany, Belgium to name just a few, the amount of possible user interactions become huge. A user booking a sports ticket for 3 friends and wanting to choose where they sit is vastly different to someone buying a festival ticket and needing the tools to sell 30 more.

The one consistency across the Verve universe was the concept of rewards. You sell X tickets, you get Y reward - it became core to the user interactions on the site, and we saw it becoming a core driver for both acquisition and retention. Your average Verve user sells a few tickets and will probably earn themselves free access to the event - for these users this is realistically their end goal. Free is a great driver and for many it allowed them access to events they usually couldn’t otherwise afford. What we noticed though was there was a growing super user group who didn’t stop at a free ticket. They wanted more. They leveraged their social networks like an influencer, and had generated an extended reach. We wanted to build a product that could harness this super user behaviour and turn these self-made ‘influencers’ into acquisition channels. We did this through the creation of Pollen.

If Verve was a Series B, 200 headcount, in-rev company, Pollen was a pre-seed startup. We had user data, 100 users and some limited resource to hit MVP.

My role as Design Leadership in Pollen was possible due to the amazing work of Antony Shaw and Peteris Bikis. Hire those guys. They’re very very good.

The Problem

Can we take the power users we have on our existing platform and, through a combination of behavioural nudging, CX support and exclusivity, deliver a rewards network that enables higher sales per user, higher retention, better network driven promotion and cheaper acquisition? To solve this our definition of success would hinge on a few key metrics - on the quantitive side we’d look at GMV, NPS, Campaigns Joined (a user must join an event campaign to sell tickets for it) and Dwell Time (time spent inside or discovering a campaign). Qualitatively we’d look at the feedback we’d get from the CX teams, the semantics of the interactions, and we’d run face to face events to align to our testing - at the events our users were selling for. This last point was key to solving the desire lines in the product, as we saw event attendance as the last part of the user flow.

Pre-MVP

Because Verve is by nature a white label platform - it allows clients to rebrand their Verve site to match their event of campaign - we began initially by using this as the basis for what we were now calling Pollen. This allowed us to first of all test the flows we already had created in Verve, and work out which would be repurposed, but it also allowed us the ability to begin approaching the market with something to show. If Verve is a single sided marketplace through it’s white labeling, then Pollen, with it’s own branding would be a double sided marketplace - we needed users and events to transact at the same time. Without events we couldn’t get users, without users we couldn’t get events.

Reutilising existing flows and product elements was key in initial acquisition

This pre-MVP approach allowed us to begin building acquisition channels and gave us the breathing room we needed to begin brand and product scoping without blocking operations and sales. It also allowed me to quickly create pitch environments for client side partners.

Brand is everything

Building brands for the under-30s market is easier than it’s ever been in terms of broadcasting, but tougher than it’s ever been in terms of communication. Channels like Instagram, Snap, TikTok have given brand and content publishers the ability to reach people visually, way beyond the known limitations of Facebook. Paid acquisition is more impactful, and the organic story features with a focus in video and image get significant reach if done well. In researching these channels heavily with agency The Big Thinks, we defined some key focuses for communication:

1. Authenticity - Facebook has killed trust, and users are more cynical than ever before. Truth is king, and it has to look genuine - no gimmicks, nothing that can be conceived as BS, transparency is key, be bulletproof, but not to the point of being arrogant.

2. Don’t look like a pyramid scheme - Seriously, open up your Facebook and find one old school friend not trying to recruit you into a dubious scheme selling protein or moisturiser or something. Users are tired of this ‘friend drag’ and see right through it - in fact this came up again and again in our qualitative research. If it’s too good to be true it usually is, and the people your want to acquire will write you off as a scam. If your friends is offering free tickets on your feed and all you have to do is ‘click here’, people turn off.

3. Vibrancy - Seems obvious but not to be overlooked. Vibrant content really stands out and can act as a Trojan horse for complex messaging and boring terms and conditions. It’s actually hard to find good, vibrant content - people talk about their content or story being vibrant, but in truth it rarely is. Just because something is colourful doesn’t make it vibrant, and maintaining vibrancy from acquisition to process to retention is unbelievably hard to do.

4. Elevation - AKA ‘aspiration’ in some circles, but rarely is UX or CX ‘aspirational’, so ‘elevation’ it is. Elevation is usually the missing piece in product and brand flows, and it’s the oil in the engine. You can ‘nudge’ users and force desire lines, but elevation is more effective if it’s baked into the DNA. It’s the ‘too good to be true’ with real Authenticity and ‘not looking like a pyramid scheme’. It’s also incredibly difficult to define and needs to be user led.

I’m not immune to the brand think cynicism and it’s sometimes hard to type brand think into product design without feeling a little like you’re talking fluff to logic, but time after time I’ve seen products fail over and over again due to a lack of foresight and ‘feels’ when it comes to user acquisition. Acquisition is a lifeblood, yet it’s often talked about in terms of part of a paid strategy and usually nothing more. If you want to encapsulate a true sense of usability and elevation in a product, you have to have some mind in brand and marketing. This blending of the marketing and product, the ‘cross-pollination’ if you will that I was able to place into the teams at Pollen, in retrospect, was key to its organic success. Why? Pollen never lied, and its marketing and voice never wrote a cheque it’s product, CX and real-world presence couldn’t cash.

If Marketing and Product converge (which they should) on one action it should be ‘managing expectations’. Marketing should not be acquisition at all costs, nor should Product be ‘execute flows on the acquired user, everything else can be dealt with by CX’. One of the biggest issues I think we solved at Pollen was applied constraints. While it’s easy to promise the earth and tell the stories you want to seal the deal, it creates a falsehood where the brand overtakes the product and disappointment or suspicion gets seeded in the community. If there’s one thing to be learnt from Fyre Festival, it’s probably that.

The outcome of this Branding period was Product was educated enough to consult on imagery and copy for campaign, organic and paid. In fact post-MVP at Pollen saw Product Design working directly on Marketing campaigns alongside Marketing Designers with equal positioning.

MVP - There’s an app for that

You don’t need an app for that usually. In fact an app usually slows things down when trying to find the thing you need to build. In the case of Pollen however we wanted to build an app because our data told us to build one. In Verve world, where the web is accessible to everyone, an app isn’t a priority (it’s something that’s always in early stage development at Verve, but never ships). Pollen needed an exclusivity angle and using React Native allowed us to make an exclusive container for Pollen while keeping web on target. There are pros and cons to React Native, I’ve read them all. Pollen was the perfect React Native candidate.

Pollen MVP powered by React Native

If Verve was single campaign driven (one user, one festival), then Pollen was Multi-Campaign driven. This created a lot of complexity in both our user testing, but also in our personas and known quantitive data. Verve had been built and tested on a single campaign approach, focused flows, simple checkout - in fact simplicity was the 100% focus with Verve.

Pollen was very different. It was a discovery product first, with user choice and communication coming at the top of the funnel, leading to multiple checkout flows, options and end goals. If a Verve user had one definition of success to manage (sell X tickets, get Y reward), then there were potentially unlimited defined successes in Pollen, with different actions delivering different rewards at scale.

To navigate this we couldn’t equate X=Y anymore. We had to develop a different system for rewards based on a framework that could adapt to different values and different frequencies. If selling 5 tickets in one campaign as a Verve user equates to 1 free ticket as a reward, what is the reward if a Pollen user sells 1 ticket in 5 different campaigns? Do they get a free ticket? And if so, for which campaign? What if the campaigns are priced differently? Does a $400 holiday score more highly than a $10 gig ticket?

We began to build a rewards system that assigned point values to sales, weighted towards cost of sale. A user sells a holiday - 10pts, a user sells a gig ticket - 2 points. This made logical sense. Something costing more is harder to sell, therefore would get more rewards. We created Blue, Gold and Black statuses, each requiring a certain point value to access. Each ‘tier’ entitled the user to a range of rewards they could claim relating to their interests. The limitations of this system however began to show itself immediately in testing with users.

Reward behaviours

In testing rewards behaviours we began to realise that a weighted point based system wasn’t elevating our users. There were a few reasons for that. The logic of more cost equating to more reward makes sense (AMEX has built an empire from this dynamic) but for Pollen users the dynamic was too capitalistic. What we were doing was equating an experience someone was having with the cost of its purchase, which is to say that something that costs more MEANS more in our ecosystem. That just wasn’t the case. Our users saw straight through it - they saw money meaning access and immediately we began to loose authenticity. Money meaning access is something we’d originally sought to avoid with our users - those under 30 for the most part don’t see their personal wealth as an elevating element. We saw claims of unfairness, of ‘so if I’m rich I get more stuff’, of ‘this isn’t for me, because I can’t afford a holiday’. The reward behaviours we began to see positively emerge were non-hierarchical - every sale gives you a something towards a reward. And it was this non-hierarchical reward structure that unlocked the retention model we were looking for.

What we discovered was that ‘fandom’ played a large part in retention. In the world of ‘fandom’ someone who sells 5 tickets weekly to a club night has more worth than someone who purchases one large ticket item. In turn that ‘fan’ has more leverage with their network through constant broadcasting, making them effectively a micro-influencer within their domain.

‘Fairness’ in this world then becomes less about spend and more about usage. It’s less AMEX and more Foursquare. Removing money as a contributing factor began to make the reward dynamic more trustworthy, and more accessible. It removed the unnatural calculation of cash vs points.

We switched the tier access in Pollen to ‘amount of sales’ rather than ‘sales amount’ for MVP and it immediately opened up the concept of value to the users. Value that wasn’t monetary. Value that was calculated by the user, not Pollen itself.

Aiding Discovery

The quickest way to gain a user and retain them in Pollen was to give them the opportunity to discover new and exciting things, but more importantly let them find things they were looking for. With a marketplace built for scale, there’s only so many items you can place in a carousel before the intent to search is lost. We began with a powerful search tool front and centre in the UI, with predicative search display and the added options to request a search item if none was found - this allowed us to poll our users in real time.

The smart search feature in Pollen

For many products (such as Citymapper) search is bread and butter. It’s the 99% use case. In Verve world, search was non-existent, so integrating search into the product was a big deal. In fact it was such a big deal that multiple revisions of the discovery product were trialed before we tackled search being given such prominence in the UI. Search really began to take centre stage (and become more smart) when we finalised the first quarter of MVP trials in London. Pollen was expanding quickly and was showing astonishing GMV values, and the pressure was on to begin rolling Pollen into the US as soon as possible.

Adding location switches into Pollen

Much like Citymapper, we added a city switcher into the product which put more and more pressure on search, along with hyper-localised discovery. The back end required work - more tagging, more controls and the partner requirements equally grew as we added new locations. We also began reworking our initial MVP style guides to incorporate geographies. The UK is largely a single territory clubbing / festival market, with hyper-local users attending hyper-local events, save for annual festival pilgrimages. Stylistically and semantically its London - sorry to say that, but its one city that aligns the country in ‘how things look and feel’. The US is vastly different. We found our users travelling from NYC to attend a Diplo event. People from LA travelling Okerchobee, people from the east coast making the a pilgrimage to Coachella. Would someone in London use the city picker? Probably not. Would someone in NYC? Most definitely.

The more time we spent with our users on the ground in the US the more we realised that culturally the geographies were very different. Can we display an event in London the same as in LA or NYC? This opened up design framework discovery and a focus on an ever simplifying style guide.

Management

One consistency across all the locations we added was the core functionality of sales management. With users selling tickets and experiences across multiple events, locations and dates, to multiple friend groups, through WhatsApp, SMS and email - we needed a robust sales flow that would be easy to use and not require large amounts of CX spend.

Sales management in Pollen

Integrated into that sales management was SMS flows, allowing a user to send their friends checkout links and tickets, along with ‘sales links’ giving the user the tools to pitch their friends events they wanted to sell from within the app interface. These integrations were significant complexity - event discovery through to event selection through to adding friends, pitching and checkout - and getting them operational across multiple ticket providers in multiple currencies in different geographies was an impressive feat to achieve for MVP, but it was so core to the product experience that they couldn’t just be a redirect to mobile web. React let us build cross platform delivering consistency in our flows and user data.

What now?

Pollen continues to grow. In fact what was an MVP in one territory now controls the majority of collegiate travel in the US and Canada and continues to dominate in Europe. That growth was so strong that Pollen - in the space of two years - replaced Verve as both the core product and the core brand - and now leads strategy. Verve is no more - Pollen took over, and seems to be continuing to ‘pollenate’ within the experience and events market year on year.

Pollen continues to dominate

Update

In October 2019, Verve raised $60M in Series C financing with Kindred and others on the strength of the Pollen marketplace - and in turn rebranded Verve to Pollen. Pollen now leads the influencer sales market in both EMEA and the US, and through M&A owns the lion share of student travel in the US.