Rumor vs LunarCrush
Rumor and LunarCrush both measure what social media says about markets. They often give very different numbers, for two reasons.
First, engagement: how many people respond to a post. Rumor measures it the industry-standard way. LunarCrush also counts every time a post is shown on someone’s screen, so its engagement figure mostly measures what a social network chose to show.
Second, what a post is about. Rumor has a model read each post and check that it is about the company or coin. LunarCrush counts any post that uses the name, so posts about the football player Tom Brady are counted for Brady Corp, an unrelated maker of safety signs and labels.
Rumor is built for investors, with 17 years of stock, ETF and crypto conversation. LunarCrush covers more ground: more coins, TikTok and Instagram, and sports teams, brands and people.
Below is where each one is stronger, feature by feature.
Last reviewed
The Data
| Rumor | LunarCrush | |
|---|---|---|
| History | ||
| History | 17 years Since September 2009 | 6 years1 Since January 2020 |
| Social networks | ||
| Social networks | X, Reddit, YouTube and Stocktwits | X, Reddit, YouTube, TikTok and Instagram1,2 |
| Stocks | ||
| Stocks | ~7,500 Delisted ones included | ~4,2304 |
| ETFs | ||
| ETFs | ~6,700 | ~100, listed inside Stocks4 |
| Crypto | ||
| Crypto | ~880 | 5,5525 |
| Sports, brands and people | ||
| Sports, brands and people | — | 3,9305 In 34 categories |
| News | ||
| News | 60 publishers About 90,000 articles a month | Publishers' posts on X3 |
Rumor covers more stocks and ETFs, and goes back to 2009. That depth is the point: it lets you see how attention and sentiment move through whole market cycles.
LunarCrush covers more ground: more coins than Rumor, plus sports teams, brands and people. Its history is shorter, starting in 2020, but it counts two networks Rumor does not, TikTok and Instagram.
Measuring Conversations
| Rumor | LunarCrush | |
|---|---|---|
| Engagement | ||
| Engagement | Interactions Likes, replies and shares | Reach10,3 Mostly views, labelled engagement |
| Share of the conversation | ||
| Share of the conversation | Attention Share of engagement in the stock, ETF or crypto conversation | Social dominance6 Share of posts within its category |
| Sentiment | ||
| Sentiment | Bullish and bearish Counted separately | One score7 Percent of engagement that is positive |
| Topic models | ||
| Topic models | Multiple
| None3 A few themes for the current day only, written fresh each day |
| Summaries | ||
| Summaries | Daily Any day or month back to 2009 | Current day8 |
| User metrics | ||
| User metrics | Newcomers and regulars Engagement and summaries for each | Creator counts and top creators10 Creator pages, time series and rankings |
| Composite scores | ||
| Composite scores | — | Galaxy Score and AltRank7 |
Rumor and LunarCrush disagree strongly about how to measure engagement. In social media analytics, engagement is the number of times people act on a post: liking it, replying to it or sharing it.
Rumor follows the industry standard. Engagement counts how many people responded with an action. Each like, share or reply is a person who stopped, read the post and chose to act on it. An interaction records that the post held someone’s attention. That makes it the best measure we know of what investors are actually paying attention to.
LunarCrush mixes two different terms. They measure engagement by also adding impressions and views, which is a different measurement: reach.
Why is this bad? X counts a view whenever a post appears on someone’s screen, including when they scroll past it without reading a word. The result is misleading: its “engagement” numbers mostly track how widely X distributed a post and how large the account behind it is.
Cleaning the Data
| Rumor | LunarCrush | |
|---|---|---|
| Spam filtering | ||
| Spam filtering | Yes | Yes11 |
| Matches posts to the right asset | ||
| Matches posts to the right asset | Yes A model checks whether “Brady” means the company or the quarterback | No Counts every post that uses the name or the ticker |
Labeling conversations on social media with the correct topic is hard.
Rumor uses dozens of proprietary models to ensure we are representing each asset accurately.
LunarCrush matches on the name or ticker alone, so a common word pulls in conversations that have nothing to do with the company.
Features
| Rumor | LunarCrush | |
|---|---|---|
| Screener | ||
| Screener | Filter on hundreds of columns | Sort by one column, filter by one sector12 |
| Heatmaps | ||
| Heatmaps | Yes | Yes |
| Interactive charts | ||
| Interactive charts | Pan and zoom through the full history Drag, scroll or pinch to move through time | — |
| Post viewer | ||
| Post viewer | — | The top posts behind each topic9 |
| Alerts | ||
| Alerts | — | On sentiment flips and volume spikes13 |
Rumor is built for research: searching a market for the assets that fit what you are looking for, and following their conversation back through the years. LunarCrush has better tools for alerting, and lets you read the posts behind a topic.
Access
| Rumor | LunarCrush | |
|---|---|---|
| REST API | ||
| REST API | Yes | Yes14 |
| MCP server for AI tools | ||
| MCP server for AI tools | Yes | Yes15 |
| Command line | ||
| Command line | — | Yes10 |
Rumor and LunarCrush both connect to AI assistants such as Claude through an MCP server, so you can ask questions of the data in plain English: which stocks are drawing the most bullish attention this week, or how the conversation about a company changed after its last earnings report.
For programs that read the data directly, both sell a REST API. LunarCrush also has a command-line tool.
Plans and Prices
| Rumor | LunarCrush | |
|---|---|---|
| 100 API calls a day | ||
| 100 API calls a day | Free, $0/mo | Free, $0/mo14 |
| 2,000 API calls a day | ||
| 2,000 API calls a day | Personal, $49/mo $39/mo billed annually | Light (Individual), $90/mo1 $72/mo billed annually |
| 20,000 API calls a day | ||
| 20,000 API calls a day | Pro, $199/mo $159/mo billed annually | Medium (Builder), $300/mo1 $240/mo billed annually |
| 100,000 API calls a day | ||
| 100,000 API calls a day | — | Heavy (Scale), $900/mo1 $720/mo billed annually |
| Free trial of a paid plan | ||
| Free trial of a paid plan | 24 hours | None16 |
Both Rumor and LunarCrush charge by the number of API/MCP calls per day, and both offer 100 free calls per day. Rumor’s paid plans cost less, and you can try one free for 24 hours before paying. LunarCrush has a larger plan for heavy use.
Which to Choose?
Choose Rumor If
- You want a tool built for stock, ETF and crypto investors
- You want 17+ years of history, back to 2009
- You want industry-standard engagement metrics
- You want to ensure data is labeled correctly
- You want to try the paid plan first (24 hours free)
Choose LunarCrush If
- You trade small coins Rumor does not track
- You follow brands, sports or people
- You need TikTok and Instagram counted
- You want to read the posts and know who is writing them
- You need alerts