Predictive Wallet Behaviors
Mnemonic's Audience Insights API provides predictive behaviors of wallets. Via a single endpoint you can obtain actionable insights into your audience, and build new or enhance existing applications with delightful personalized experiences.
This is a powerful tool that can be used to:
- Predict who is more likely to engage with your collection.
- Identify and engage with your influencers, or find who to partner with based on other collections.
- Better understand the sociographic profile of your community and/or your audience.
- Build delightful personalized experiences that are tailored to your community, fans and/or audience.
Note: In order for a wallet to have a proper label it requires for a wallet to have engaged with an NFT at any point in time, however, it does not require for a wallet to own NFTs at the time of the analysis.
Wallet behaviors are updated daily to enable the most accurate and personalized experiences within wallets, marketplaces and other applications.
A quick walkthrough
Our wallet behavioral categories can be used in a variety of ways and use cases, either directly or as inputs into another model or proprietary formula as a highly predictive signal of a very likely action expected from that wallet, for optimizing marketing campaigns, loyalty programs, community management, growth initiatives, pricing strategies, recommenders and more.
Behaviors categorization
The following table provides definitions of supported wallet behavior categories.
| Category | Description |
|---|---|
| Patron | A wallet that shows continued support of the NFT space through downturns and bear markets, has a consistency in collecting and supports a diversity of projects and is likely to continue collecting consistently. |
| Whale | A wallet with a history of moving large amounts of ETH and has high purchase capacity due to owning large amounts of ETH. |
| Influencer | A wallet that has access to VIP communities made out of influential figures, with access to capital and information, that also has a pattern of influencing a number of wallets, has been early on in the inflection points of very successful collections and likely to bring credibility and increased interest into new collections. |
| Tastemaker | A wallet that is frequently early in a number of the most successful collections, has a history of good taste, a large portfolio of highly valuable NFTs and likely to be ahead of everyone else in collections that have been very successful in terms of volume of sales, likely to hold and not flip, highly experienced |
| Collector | A wallet that is more likely to collect and hold, has a high purchase intent in the next 1-3 months, and is unlikely to quickly flip an NFT for short term profit. |
| Flipper | A wallet that has a high purchase intent in the next 1-3 months, and is likely to quickly sell an NFT for short term profit. |
| Programmed Collector | Similar to Collector, but has a high likelihood of using automation methods or being a wallet that is programmatically controlled, highly experienced. |
| Programmed Flipper | Similar to Flipper, but has a high likelihood of using automation methods or being a wallet that is programmatically controlled, highly experienced. |
| Advanced | A wallet that has a history of engaging with the blockchain through various complex transactions, intermediaries or has done more involved actions such as staking among other things. |
| Bad actor | A wallet that has a history of minting the smart contracts of spammy collections. |
| Creator | A wallet that has minted the smart contract of a collection which is not considered a spam, the audience size in this case pertains to all the wallets that currently own any of their collections. |
| Explorer | A wallet that has not shown a particularly strong predictive behavior of any type. A wallet can be reactivated and further subdivided based on their experience, holding or taste scores to discover the dormant collectors and advanced traders or flippers. |
| Newbie | Wallet that has a very low experience, taste and automation scores and seems to be fairly new to the space, low likelihood to purchase or sell. |
The above categories are defined within the API as follows:
enum Behavior {
BEHAVIOR_UNSPECIFIED = 0;
BEHAVIOR_WHALE = 2;
BEHAVIOR_NEWBIE = 3;
BEHAVIOR_CREATOR = 4;
BEHAVIOR_PATRON = 5;
BEHAVIOR_FLIPPER = 6;
BEHAVIOR_PROGRAMMED_FLIPPER = 7;
BEHAVIOR_ADVANCED = 8;
BEHAVIOR_COLLECTOR = 9;
BEHAVIOR_PROGRAMMED_COLLECTOR = 10;
BEHAVIOR_EXPLORER = 11;
BEHAVIOR_TASTEMAKER = 12;
BEHAVIOR_BADACTOR = 13;
BEHAVIOR_INFLUENCER = 14;
BEHAVIOR_INACTIVE = 15;
BEHAVIOR_EXPLORER = 16;
}
{
"$schema": "http://json-schema.org/draft-07/schema#",
"title": "Behavior",
"type": "string",
"enum": [
"BEHAVIOR_UNSPECIFIED",
"BEHAVIOR_WHALE",
"BEHAVIOR_NEWBIE",
"BEHAVIOR_CREATOR",
"BEHAVIOR_PATRON",
"BEHAVIOR_FLIPPER",
"BEHAVIOR_PROGRAMMED_FLIPPER",
"BEHAVIOR_ADVANCED",
"BEHAVIOR_COLLECTOR",
"BEHAVIOR_PROGRAMMED_COLLECTOR",
"BEHAVIOR_EXPLORER",
"BEHAVIOR_TASTEMAKER",
"BEHAVIOR_BADACTOR",
"BEHAVIOR_INFLUENCER",
"BEHAVIOR_INACTIVE"
]
}
Categories dimensions
Each wallet is further evaluated on various dimensions which capture their experience, taste, automation, hold, sell, spend intent and influence levels in scores which are normalized and categorized in sub categories which capture skew, ranging from very low, low, medium, high, very high.
| Score | Description |
|---|---|
| Experience score | a score that captures their level of experience with blockchain transactions and NFTs in general |
| Automation score | a score that captures the amount of automation or application of programmed methods, scripts and similar for transacting on the blockchain |
| Taste score | a score that captures their ability to identify early on the collections that will attract interest from high spenders or a demand from a larger group of people which is reflected by the quality of their portfolio as well |
| Influence score | a score that captures their ability to influence other wallets to follow them in the collections they participate in, as well as their access to information and capital through the VIP communities they belong to |
| Holding score | a score that captures their propensity to purchase and hold over the next 1-3 months |
| Selling score | a score that captures their propensity to purchase and sell over the next 1-3 months |
| Purchase intent score | a score that captures the quantity and level of intent to purchase an NFT in the next 1-3 months |
Each dimension is further provided with a score defined as follows:
enum Score {
SCORE_UNSPECIFIED = 0;
SCORE_VERY_LOW = 1;
SCORE_LOW = 2;
SCORE_MED = 3;
SCORE_HIGH = 4;
SCORE_VERY_HIGH = 5;
}
{
"$schema": "http://json-schema.org/draft-07/schema#",
"title": "Score",
"type": "string",
"enum": [
"SCORE_UNSPECIFIED",
"SCORE_VERY_LOW",
"SCORE_LOW",
"SCORE_MED",
"SCORE_HIGH",
"SCORE_VERY_HIGH"
]
}
The final wallet behavior is then defined as follows:
message WalletBehavior {
// Address of the wallet.
string address = 1;
// Behavior category.
Behavior behavior = 2;
// Wallet experience score.
Score experience = 3;
// Wallet holding score.
Score holding = 4;
// Wallet selling score.
Score selling = 5;
// Wallet taste score.
Score taste = 6;
// Wallet automation score.
Score automation = 7;
// Wallet spending intent score.
Score spend_intent = 8;
// Wallet influence score.
Score influence = 9;
}
{
"$schema": "http://json-schema.org/draft-07/schema#",
"title": "WalletBehavior",
"type": "object",
"properties": {
"address": {
"type": "string"
},
"behavior": {
"type": "string",
"enum": [
"BEHAVIOR_UNSPECIFIED",
"BEHAVIOR_WHALE",
"BEHAVIOR_NEWBIE",
"BEHAVIOR_CREATOR",
"BEHAVIOR_PATRON",
"BEHAVIOR_FLIPPER",
"BEHAVIOR_PROGRAMMED_FLIPPER",
"BEHAVIOR_ADVANCED",
"BEHAVIOR_COLLECTOR",
"BEHAVIOR_PROGRAMMED_COLLECTOR",
"BEHAVIOR_EXPLORER",
"BEHAVIOR_TASTEMAKER",
"BEHAVIOR_BADACTOR",
"BEHAVIOR_INFLUENCER",
"BEHAVIOR_INACTIVE"
]
},
"experience": {
"type": "string",
"enum": [
"SCORE_UNSPECIFIED",
"SCORE_VERY_LOW",
"SCORE_LOW",
"SCORE_MED",
"SCORE_HIGH",
"SCORE_VERY_HIGH"
]
},
"holding": {
"type": "string",
"enum": [
"SCORE_UNSPECIFIED",
"SCORE_VERY_LOW",
"SCORE_LOW",
"SCORE_MED",
"SCORE_HIGH",
"SCORE_VERY_HIGH"
]
},
"selling": {
"type": "string",
"enum": [
"SCORE_UNSPECIFIED",
"SCORE_VERY_LOW",
"SCORE_LOW",
"SCORE_MED",
"SCORE_HIGH",
"SCORE_VERY_HIGH"
]
},
"taste": {
"type": "string",
"enum": [
"SCORE_UNSPECIFIED",
"SCORE_VERY_LOW",
"SCORE_LOW",
"SCORE_MED",
"SCORE_HIGH",
"SCORE_VERY_HIGH"
]
},
"automation": {
"type": "string",
"enum": [
"SCORE_UNSPECIFIED",
"SCORE_VERY_LOW",
"SCORE_LOW",
"SCORE_MED",
"SCORE_HIGH",
"SCORE_VERY_HIGH"
]
},
"spend_intent": {
"type": "string",
"enum": [
"SCORE_UNSPECIFIED",
"SCORE_VERY_LOW",
"SCORE_LOW",
"SCORE_MED",
"SCORE_HIGH",
"SCORE_VERY_HIGH"
]
},
"influence": {
"type": "string",
"enum": [
"SCORE_UNSPECIFIED",
"SCORE_VERY_LOW",
"SCORE_LOW",
"SCORE_MED",
"SCORE_HIGH",
"SCORE_VERY_HIGH"
]
}
}
}
Example:
{
"walletBehaviors": [
{
"address": "0x3eb4b12127edc81a4d2fd49658db07005bcad065",
"behavior": "BEHAVIOR_COLLECTOR",
"experience": "SCORE_VERY_HIGH",
"holding": "SCORE_HIGH",
"selling": "SCORE_LOW",
"taste": "SCORE_LOW",
"automation": "SCORE_LOW",
"spendIntent": "SCORE_LOW",
"influence": "SCORE_VERY_LOW"
}
]
}
Each of these behavioral categories can be further subdivided into cohorts that have a combination of taste, experience, influence, automation, holding, selling and purchase scores.
Wallet behaviors can be used as one of the building blocks to build powerful recommenders within a wallet or a marketplace.
Relevant APIs:
Updated 11 months ago