Inspiration

What it does# About QuestLoop

Inspiration

Most loyalty and referral programs treat every customer the same.

Businesses usually offer a generic incentive such as:

Refer three friends and earn $10.

But customers influence people in very different ways. One customer may enjoy posting publicly, another may bring coworkers, while someone else may be great at organizing family or community events.

At the same time, businesses have highly specific growth problems:

  • A restaurant has empty tables on Tuesday afternoons.
  • A fitness studio has unfilled classes.
  • A salon needs to recover cancelled appointments.
  • A local event still has unsold seats.

We wanted to connect these two sides.

Instead of giving every customer the same referral link, we asked:

What if every customer received a personalized mission based on how they naturally influence others and what the business needs right now?

That became QuestLoop.


What It Does

QuestLoop is a personalized, gamified loyalty and referral platform.

The business starts with one measurable objective:

Fill quiet Tuesday afternoons with new customers.

QuestLoop then generates a different quest for every customer.

For example:

  • A social creator may be asked to run a poll and bring one friend who voted.
  • An office worker may be asked to invite two coworkers for lunch.
  • A community organizer may be asked to host a four-person tasting event.

Each quest includes:

  • A personalized action
  • An XP reward
  • A business reward
  • A completion condition
  • A playable digital experience

Depending on the customer, the quest may require:

  • A poll
  • A landing page
  • An RSVP form
  • A QR code
  • A private invitation
  • A social graphic
  • A referral link

Customers earn XP when they complete quests and progress through loyalty tiers:

Explorer → Connector → Organizer → Ambassador

Higher tiers unlock more valuable challenges, rewards, and experiences.

The core difference is:

Traditional loyalty programs personalize the reward. QuestLoop personalizes and builds the action required to earn it.


How We Built It

We designed the project around one simple flow:

Business goal
→ customer context
→ personalized quest
→ playable quest experience
→ verified completion
→ XP and tier progression

Nexla

Nexla provides the normalized customer and business context used by the quest-generation agent.

That context can include:

  • Purchase history
  • Visit patterns
  • Referral behavior
  • Preferred sharing style
  • Current XP and loyalty tier
  • Previous quest performance
  • The business’s current growth objective

Nexla helps turn fragmented customer information into a consistent profile the agent can reason over.

Its role is:

Understand who the customer is and what the business needs.

Zero.xyz

Zero.xyz is the core execution layer.

The system does not know ahead of time what every personalized quest will require.

A social quest might need a poll and a graphic.

A workplace quest might need a private invitation and RSVP page.

A community quest might need an event page, registration form, and QR code.

After the agent creates the quest, it determines the required capabilities and uses Zero.xyz to discover and execute the necessary services on demand.

This changes QuestLoop from a fixed library of referral templates into an open-ended quest engine.

Its role is:

Turn the personalized quest from an idea into a working experience.

Pomerium

Pomerium protects the identity and reward boundary.

It ensures:

  • Customers access only their own private quests.
  • Business controls remain restricted.
  • Quest-completion endpoints are protected.
  • XP and rewards are awarded to the correct user.
  • Unauthorized users cannot claim another customer’s quest.

Its role is:

Make personalized quests, completion, and rewards trustworthy.


The Demo

Our MVP focuses on one business objective:

Fill quiet Tuesday afternoons with new customers.

We show three customers:

Maya — Social Creator

Maya receives:

Let your friends choose your Tuesday order and bring one person who voted.

Her quest may include:

  • A voting page
  • A social graphic
  • A tracked referral link

Omar — Office Connector

Omar receives:

Bring two coworkers for Tuesday lunch.

His quest may include:

  • A private invitation
  • An RSVP page
  • A QR code

Lena — Community Organizer

Lena receives:

Host a four-person Tuesday tasting.

Her quest may include:

  • An event page
  • A registration form
  • An invitation card

The important demonstration is that the business goal remains the same, but the referral mechanism changes completely for each person.

When a quest is completed, the customer earns XP and can move to a higher tier.


What Was Newly Possible

Before Zero.xyz, we would have needed to prebuild every possible quest format.

Developers would have had to:

  • Select providers in advance
  • Create accounts
  • Configure subscriptions
  • Manage API keys
  • Build every integration
  • Maintain every campaign type

The AI could personalize the wording, but the underlying experience would still come from a fixed set of templates.

With Zero.xyz, the process becomes:

Understand the customer
→ invent the best quest
→ identify the required capabilities
→ discover services at runtime
→ build the playable experience
→ launch it

The quest does not have to exist before the customer receives it.

That is the central innovation behind QuestLoop.


Challenges We Faced

Keeping the Idea Focused

Our first versions became too broad.

We explored:

  • Full loyalty dashboards
  • Multiple business objectives
  • Large quest libraries
  • Reward marketplaces
  • Leaderboards
  • Complex progression systems
  • Many customer-data integrations

We realized the strongest demo was not a complete loyalty platform.

It was one clear moment:

The same business goal creates completely different playable quests for different customers.

We reduced the MVP to three customers, one objective, one quest-generation flow, and one visible Zero-created experience.

Making Zero Essential

A major challenge was avoiding a shallow sponsor integration.

It would have been easy to use Zero.xyz only to generate an image or publish a fixed webpage.

Instead, we designed the system so the agent first invents the quest and then uses Zero to obtain the capabilities required to execute it.

This made Zero part of the product architecture rather than an added API call.

Balancing Personalization and Privacy

Quest generation becomes more useful when the system understands customer behavior, but that also creates privacy and authorization concerns.

We limited the customer context to information directly relevant to quest creation and used Pomerium to protect customer-specific experiences and reward actions.

Reliability

Runtime service discovery introduces uncertainty.

A service may fail, respond slowly, or return an unusable result.

To keep the demonstration reliable, we designed fallback behavior:

Primary service fails
→ record the failure
→ try an alternative capability
→ use a local fallback if required
→ continue the customer experience

The agent should not become stuck simply because one provider fails.


What We Learned

We learned that personalization is more valuable when it changes the action, not only the message.

Changing:

“Invite your friends”

to:

“Invite your coworkers”

is not enough.

Real personalization means changing the complete mechanism:

  • Who the customer invites
  • How they invite them
  • What digital experience supports the action
  • What reward is appropriate
  • What happens after completion

We also learned that the most powerful use of Zero.xyz is not access to thousands of services by itself.

The real power is that an agent can encounter a capability gap, acquire what it needs, and continue working toward the user’s outcome without forcing the user to configure another tool.

Finally, we learned to keep the product focused on business value.

QuestLoop is not primarily a points game.

It is a customer-acquisition engine designed to help businesses activate existing loyal customers and fill unused capacity.


Business Value

QuestLoop can help businesses:

  • Reduce customer-acquisition costs
  • Increase referral participation
  • Fill unused or perishable capacity
  • Generate more group visits
  • Increase customer lifetime value
  • Build stronger customer communities
  • Measure which customer behaviors create revenue

The business can track:

  • Quests launched
  • Quests completed
  • New customers generated
  • Revenue attributed to quests
  • Cost per successful referral
  • Capacity filled
  • Reward cost
  • Tier progression

The strongest initial markets are businesses with repeat customers and unused capacity, including:

  • Boutique fitness studios
  • Restaurants and cafés
  • Salons and wellness businesses
  • Climbing gyms
  • Local classes and workshops
  • Community-based membership businesses

What’s Next

The next version of QuestLoop would add:

  • More dynamic business objectives
  • Real transaction and booking verification
  • Adaptive quest recommendations
  • Customer choice between multiple quests
  • Better reward optimization
  • Quest-performance analytics
  • Multi-location business support
  • Continuous learning from completed and abandoned quests

Over time, the system could learn:

Which customer should receive which growth action, at which moment, and for which reward?


Final Summary

QuestLoop turns customer loyalty into a personalized game.

A business defines one growth objective. Nexla provides the customer context. The agent creates a unique quest. Zero.xyz builds the required experience on demand. Pomerium protects identity and rewards.

Customers complete missions, earn XP, unlock new tiers, and help the business grow in the way that feels most natural to them.

One business goal goes in. A different playable growth quest comes out for every customer.

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for Quest Loop

Built With

  • and
  • built-with:-zero.xyz
  • codex
  • framer
  • gpt-5.6
  • nexla-adk
  • next.js
  • pomerium
  • react
  • tailwind-css
  • typescript
Share this project:

Updates