Hello DOT builders, developers, independent researchers, and community!
We would like to present to you a user behavior research product that will help improve the effectiveness of application development in the ecosystem and identify possible synergistic partnerships. Our goal is to study user behavior based on the ALREADY existing onchain information on the blockchain. 

[Original discussion post >>](https://polkadot.polkassembly.io/post/1679)

### Context of the proposal
We propose web3 user-research tool for the Polkadot ecosystem that enables developers to analyze and score user behavior on-chain, providing a deeper understanding of the current state of the ecosystem at the user, dApp, and parachain level.

With real-time detailed snapshots of the existing Polkadot community, developers can improve their products, create stronger communities, and effectively incentivize newcomers using Motif's insights. 

### Reference/experience
Our team has conducted and is actively engaged in multiple research projects to battle-test our event-based extraction method. Examples of completed and ongoing studies include:
- dYdX (complete)
- Astar (complete)
- Lido (in progress)

We have made an effort to provide a comprehensive explanation in a [full proposal >>>](https://docs.google.com/document/d/1WEhn-kQQjYxpWD8hC05Nc4VK9oPYROFl/edit)

In the following sections, we will present the results of these studies and demonstrate that the event-based method is sufficient for the purposes of research within the Polkadot ecosystem. 

### Problem statement
The operation of P2P systems is based on the P - peers (or people)  themselves. Due to the nature of independent economic systems with internal accounting units, users not only generate revenue but also directly influence the economic parameters of the system. It is the behavior of users that is the main lifeblood and success factor of that kind of economic systems. To operate successfully within such ecosystems, the creators of both the ecosystems themselves and the services and applications within the ecosystem must have a good understanding of user behavior, tools to influence it, and effectively track results.
- How can we identify the users who effectively and consistently use the dApps?
- How can we identify the users who fraud actions and try to deceive the system to obtain a reward?
- What motivates users to provide liquidity and how does the ecosystem incentivize them to continue doing so?
- How can we track the last actions of users who have transitioned from one ecosystem to another?
- How can we motivate average users to engage more actively with the system?
- How can we retain existing profitable users?

Answering these simple questions or finding relevant information within Polkadot is practically impossible. The main reasons for this are:
- Blockchain user data reality itself: a huge, interconnected and decentralized mess which is difficult to explore. 
- The modular multilevel structure of Polkadot is hard to overview.
- The limited number of functioning dApps and zero-to-none smart-contracts as a source of customer behavior leads to a lack of information.

### How is this problem being solved today?

One of the current solutions is one-by-one research. However, "wallet-by-wallet" research is extremely difficult. Researchers have to go to different wallets, look at the experience of interactions, compare similar wrapped/transferred tokens/coins/portfolios, actions, etc. Comparing users, let alone profiling them is almost impossible. It leaves no chance for factually understanding users' behavior and providing targeted incentivization to wallets with relevant experience. 

Due to this complexity, analysis often boils down to two simple metrics: the number of tokens on the relay chain or the number of wallets (for example, how many came and performed some action). This leads to a distortion of reality and inefficiency in analysis and results: prioritizing “big balance” leads to the dominance of "whales" behavior, while “big numbers” lead to the domination of "bot farms." However, these cases do not lead to growing the numbers of the active users, and such metrics do not allow the application to improve its efficiency. On the contrary, they provide an opportunity for potential fraud from such participants.
Even for the whole ecosystem itself these "bot farms" and "whales" can create a centralization effect, exposing the ecosystem to potential manipulation by centralized third parties. 
Therefore, a significant portion of tools based on user behavior aimed at incentivizing them, such as airdrops, retrodrops, quests, whitelists, and others, do not produce the desired results.


### Motif’s solution

Motif provides 3 main tools: 

**1. Parachain and Relay Chain Overview User State (B2B)**
This tool provides an overview of the current holders, including their profiles, assets, and level of activity. It can help identify the most active and engaged users in the ecosystem and their holding patterns. 
![](https://gateway.pinata.cloud/ipfs/QmVyzQMfwgt2veQQuS7GmstE7XSHwyhTrBPqS6p23oGwXM?_gl=1*5q5xpv*_ga*MTc4ODEzODIxNy4xNjc4MTIzMTc1*_ga_5RMPXG14TE*MTY3ODEyMzE3NC4xLjEuMTY3ODEyMzI0MC41Ny4wLjA.)

**2. Individual Wallet Cross-Parachain Traction Analytics for Personal Use (B2C)**
Any user can input an existing native wallet address and receive comprehensive information about the user's activity, presented in an easy-to-use format. This includes details on the assets the user currently holds or has held in the past, the specific assets they have traded, and the most active protocols or parachains the user has interacted with.
![](https://gateway.pinata.cloud/ipfs/QmPNUF5NUWu3GvCuJtdRiCCJzCcYXSNt8bUjYRtxi3PPLk?_gl=1*1s9z1iz*_ga*MTc4ODEzODIxNy4xNjc4MTIzMTc1*_ga_5RMPXG14TE*MTY3ODEyMzE3NC4xLjEuMTY3ODEyMzI0MC41Ny4wLjA.)

**3. Batch Scoring of User Sets (B2B)**
Motif provides the opportunity to score a set of users in a batch, allowing for research and analysis of user sets. This helps to identify the best users for DeFi, which users will hold long-term, and which users are most active members of the Polkadot community.
![](https://gateway.pinata.cloud/ipfs/QmVMTw3JordrdZjrcjYXhpC9hxN3TzKbrkRvQmsuexU3Ps?_gl=1*1s9z1iz*_ga*MTc4ODEzODIxNy4xNjc4MTIzMTc1*_ga_5RMPXG14TE*MTY3ODEyMzE3NC4xLjEuMTY3ODEyMzI0MC41Ny4wLjA.)

The goal of Motif is to help developers improve the user experience of their dApps and incentivize the most active and engaged users. Our core concept is to research and connect the most valuable users with the most suitable dApps to accelerate the growth of the Polkadot ecosystem.


### Research method 

Our research method at Motif is event-based, focusing on events within transactions and associated data.
We generally work in two major steps: Extraction and Calculation (or orchestration). The extraction step involves collecting and organizing data from various sources, while the calculation step involves analyzing and interpreting the data to gain insights and produce results.

You can learn more about our methodology in the [full proposal >>>](https://docs.google.com/document/d/1WEhn-kQQjYxpWD8hC05Nc4VK9oPYROFl/edit#heading=h.3dy6vkm)

Here we will provide an overview of the main capabilities of our proven methodologies.

**Event-based extraction method can produce enough data for the web3 research**

Furthermore, the discovered clusters can be statistically described, which can aid in making predictions about the performance of future users and their actions 
![](https://gateway.pinata.cloud/ipfs/QmdF7CKBzpjgCbmwjih38oQvhY4tcmXCG1V4JWdjbdBUMa?_gl=1*7jddbl*_ga*MTc4ODEzODIxNy4xNjc4MTIzMTc1*_ga_5RMPXG14TE*MTY3ODEyMzE3NC4xLjEuMTY3ODEyMzI0MC41Ny4wLjA.)
![](https://gateway.pinata.cloud/ipfs/QmPNxxdmdPfPJZhUFzxE7BojRNM8xbJrqpSA9W9KdohbSr?_gl=1*twtwv8*_ga*MTc4ODEzODIxNy4xNjc4MTIzMTc1*_ga_5RMPXG14TE*MTY3ODEyMzE3NC4xLjEuMTY3ODEyMzI0MC41Ny4wLjA.)


**Users can be profiled by their activity or assets**

Users can be profiled based on their activity or assets. This is known as user profiling, which involves dividing users into groups based on their interests.
![](https://gateway.pinata.cloud/ipfs/QmVmCqXYve9QypMMPKiXw542L8jSusC4sjzPcBtvEra2ZC?_gl=1*4fhpby*_ga*MTc4ODEzODIxNy4xNjc4MTIzMTc1*_ga_5RMPXG14TE*MTY3ODEyMzE3NC4xLjEuMTY3ODEyMzI0MC41Ny4wLjA.)


**Motif can be used for dynamic research too.**

This is how the clusterization looks after event-based extraction and aggregate recombination 

![](https://gateway.pinata.cloud/ipfs/QmT1Cx4RGnfUa53a7ASc7N7EnqFkgzCbnbqMCVty1faTj1?_gl=1*dz7fmg*_ga*MTc4ODEzODIxNy4xNjc4MTIzMTc1*_ga_5RMPXG14TE*MTY3ODEyMzE3NC4xLjEuMTY3ODEyMzI0MC41Ny4wLjA.)

Additionally, we can use other hypotheses to define clusters, such as the economic value of the user being proportional to the amount of gas spent. The more gas a user spends, the more money they spend. This can be used to further define and analyze the clusters.

![](https://gateway.pinata.cloud/ipfs/QmWH7ggYDw4i6hRXW5y84hyMbeFyjo3rzfEchCZaSoLyHX?_gl=1*17w0ipi*_ga*MTc4ODEzODIxNy4xNjc4MTIzMTc1*_ga_5RMPXG14TE*MTY3ODEyMzE3NC4xLjEuMTY3ODEyMzI0MC41Ny4wLjA.)

**Event-based Method battle-test results**

- Wallets with at least 10 transactions can be studied using transaction data, which provides a variety of features (such as type, timestamp, gas spent, and method) that are suitable for statistical and machine learning analysis.
- A daily average of 80 unique active wallets (UAW) is sufficient to uncover correlations and create user profiles.
- Wallet profiles can be predicted using research data and on-chain metrics such as minimum token levels, chains, or key actions.

You can learn more about our methodology in the [full proposal >>>](https://docs.google.com/document/d/1WEhn-kQQjYxpWD8hC05Nc4VK9oPYROFl/edit#heading=h.3dy6vkm)


### Stakeholders benefits
**DApps**
Our goal is to improve the efficiency of dApp onboarding and operation by providing opportunities for researching users, rival dApps, and the parachain itself. This allows dApps to easily navigate different ecosystems and onboard the most relevant communities. Additionally, dApps can better understand user needs and experiences, and reward the best users for their on-chain activity and holding without fear of flipping. This can streamline dApp development and operations and save on marketing and liquidity costs.
For example, dApps can incentivize users with different profiles (Top degens, Strong Holders, Early Adopters, etc.) using modern mechanics such as retrodrops, airdrops, and claims.

**Ecosystems (parachains)**
Ecosystems will have the opportunity to onboard relevant dApps (if they want to work on top of parachains), which can increase the use of native coins and grow the user base organically. This will lead to rewards for user loyalty and experience. Additionally, ecosystems can incentivize users to explore new experiences without the risk of losing funds to bot farms or flippers.

**Polkadot community**
Polkadot community members, whether experienced or new, will have the ability to be recognized for their loyalty and experience, and will be able to receive benefits for their engagement with the ecosystem. This can include rewards for on-chain activity, holding native tokens, or participating in research and development of the ecosystem.

### New use cases for the Polkadot

**Updatable/dynamic Web3 id.** Motif offers an updatable and dynamic Web3 ID that can meet the demands of on-chain user recognition, scoring, and profiling systems. This ID can be integrated into smart-contracts or routing systems, providing enhanced security for users and customization options for dApp developers.
The general idea is to enable blockchain applications to provide personalized conditions based on a user's real on-chain experience, without using any information from outside the blockchain (off-chain can be easily manipulated). The existing "HYPE concept" of soulbound tokens does not seem suitable, as the verification should occur at the moment of interaction (there can be a gap between the emission of the SBT token and the interaction).This case can be solved by implementing a scoring smart contract standalone or within the routing circuit (as part of it or as part of the route when conditions occur). Such a "smart contract factory" can work based on the solution we propose.

**DID.** Motif can also provide a decentralized digital ID based on the user's fingerprint (user trace in one wallet types) and predicted level of on-chain activity.. 
Current solutions for user address unification based solely on public addresses are not viable as they can be easily falsified. In our view, a usable digital fingerprint is only possible with transactions that have internal economic value. This combination of transaction patterns with economic significance and related information increases the chances of a real human subject managing the given address. Again, thresholds and identification of such transactions are possible on the proposed platform.

**Web2 scoring** where Web3 data can be used by Web2 apps for various purposes.

**Web3 scoring interfaces** for the Polkadot. 

**Web3 scoring backend** for Polkadot, such as Nansen/debank look-alike dApps that currently do not exist in the ecosystem.

These use-cases can provide opportunities to onboard new dApps to the Polkadot ecosystem, increasing its potential for growth and usage.


You can learn more about our Motif in the [full proposal >>>](https://docs.google.com/document/d/1WEhn-kQQjYxpWD8hC05Nc4VK9oPYROFl/edit#heading=h.3dy6vkm)


We would greatly appreciate your feedback and ideas for improvement. We thank the participants of the Polkadot ecosystem and colleagues from Parity for their intensive work on this proposal.

- Motif is proposing a user behavior research tool for the Polkadot ecosystem.
- The tool will help developers analyze and score user behavior on-chain to improve application development and identify partnerships.
- The current solutions for understanding user behavior are difficult and inefficient.
- Motif provides tools for overviewing user state, analyzing individual wallet activity, and scoring user sets.
- The goal is to improve the user experience of dApps and incentivize active users in the Polkadot ecosystem.

Requested: 0.22 undefined

#271 Motif: Web3 user research platform