How the DexScreener Trending Algorithm Works
Reverse-engineering DexScreener's trending algorithm. What factors matter and how to optimize for them.
DexScreener's trending page is the single most important discovery mechanism for new tokens in decentralized finance. Millions of traders check trending daily, and a spot on the first page can drive thousands of new visitors to your token's chart. But how does DexScreener decide which tokens trend? Understanding the algorithm is the first step to optimizing for it.
This analysis is based on observed patterns, community research, and systematic testing across thousands of tokens in 2025-2026. DexScreener has not publicly documented their algorithm, so everything here comes from empirical observation rather than official sources.
The Core Ranking Factors
DexScreener's trending algorithm appears to weigh five primary factors, roughly in this order of importance:
- Trading volume (24h and 6h) — The total dollar value of trades. This is the heaviest-weighted factor.
- Volume acceleration — How fast volume is increasing compared to the token's recent baseline. A token going from $1,000 to $50,000 in 6 hours ranks higher than a token steady at $100,000.
- Unique wallet count — The number of distinct wallets making trades. More wallets signals broader interest.
- Transaction frequency — How many individual trades are occurring. 100 trades of $100 ranks higher than 1 trade of $10,000.
- Price action — Tokens with positive price movement appear to receive a mild boost, though this factor is weaker than the others.
Volume Acceleration: The Key Differentiator
Absolute volume matters, but volume acceleration is what gets new tokens onto the trending page. DexScreener's algorithm appears to measure the rate of change in volume over rolling 1-hour, 6-hour, and 24-hour windows. A token that goes from zero to $30,000 in volume over 6 hours will often outrank a token that has been steadily at $50,000 for weeks.
This is why token launches have a natural advantage for trending: the jump from zero to any meaningful volume represents infinite acceleration. However, you still need to reach minimum absolute volume thresholds before acceleration matters.
The Threshold Effect
There appears to be a minimum volume threshold below which tokens are not considered for trending at all, regardless of acceleration. Based on testing, these thresholds are approximately:
- Solana: $10,000 in 6-hour volume to enter the candidate pool
- Ethereum: $25,000 in 6-hour volume
- Base: $8,000 in 6-hour volume
- BNB Chain: $15,000 in 6-hour volume
Below these levels, no amount of wallet diversity or transaction frequency will get you trending. The thresholds shift with overall market activity — they rise during bull market peaks and drop during quieter periods.
The Wallet Diversity Weight
DexScreener significantly increased the weight of unique wallet counts sometime in late 2025. Previously, a token could trend with volume generated from just a handful of wallets. Now, wallet diversity is a meaningful ranking factor.
Testing suggests that the algorithm uses a ratio: volume per unique wallet. A lower ratio (more wallets for the same volume) produces better trending results. Optimal targets appear to be $100-500 in average volume per unique wallet for most chains.
Chain-Specific Trending Dynamics
Each chain on DexScreener has its own trending list, and the competition dynamics differ significantly:
Solana is the most competitive trending environment. With 200+ new tokens launching daily, the first page of trending turns over rapidly. Staying on the first page requires sustained volume and fresh acceleration.
Base is the second most competitive but has fewer total tokens, making it easier to stand out. A $30,000 daily volume can often secure a first-page trending spot on Base.
Ethereum has higher absolute volume thresholds but less competition from new launches. Tokens that reach Ethereum trending tend to stay there longer due to lower turnover.
What Does NOT Seem to Matter
Based on extensive testing, several factors do not appear to meaningfully influence trending position:
- Liquidity pool size: A token with $10,000 in liquidity can trend just as easily as one with $1,000,000, assuming equal volume.
- Token age: New tokens and old tokens compete on the same playing field. There is no boost or penalty based on deployment date.
- DexScreener profile completion: Having a verified logo and description does not affect algorithmic ranking, though it does affect click-through rate once trending.
- Social media links: The algorithm does not appear to factor in whether a token has Twitter or Telegram links on its DexScreener page.
Gaming the Algorithm: What Works and What Backfires
Volume bots are the most common tool for optimizing DexScreener trending performance. When configured correctly — with wallet diversity, randomized trade sizes, and sustained session duration — they can reliably push tokens into trending positions.
What backfires is using wash trading patterns that the algorithm can detect: identical trade sizes, ping-pong trading between two wallets, or extreme volume spikes followed by complete silence. DexScreener has improved its detection of artificial patterns, and tokens that trigger these filters may be deprioritized or flagged.
Optimizing for the Algorithm
Based on everything observed, the optimal approach for DexScreener trending involves:
- Generate volume above the chain-specific threshold within 6 hours
- Spread trades across 50+ unique wallets
- Randomize trade sizes between $20 and $500
- Maintain consistent transaction frequency (at least 1 trade per minute)
- Sustain activity for 12+ hours rather than concentrating in a burst
- Stack volume generation with organic community activity to amplify wallet diversity
The key insight is that DexScreener's algorithm rewards patterns that look like genuine market interest. Designing your volume strategy to mimic organic trading behavior — diverse wallets, varied trade sizes, sustained activity — aligns with both the algorithm's preferences and good practice for token visibility.