All sources
Twitter / X
Social Intel

Twitter / X

Tweets, replies, profiles

What Twitter / X is for

Twitter is where people complain in public and in real time. For anyone selling a solution to a named frustration, that immediacy is the value — a tweet about a tool someone is fed up with is a buying signal with a timestamp attached.

At one credit per result it is the cheapest way to monitor a conversation continuously rather than in occasional bursts. That matters, because the window between someone expressing a problem and choosing a solution is often days.

It suits competitive monitoring, launch tracking, and finding the specific language your market uses — which is frequently different from the language your marketing uses.

How it works

The search accepts the full Twitter query syntax, so anything that works in advanced search works here. Keywords, phrases, hashtags, from and to a specific account, language, and date ranges all compose into one query.

Filters narrow by content type and engagement: verified accounts only, exclude retweets, replies only, has media, has links, minimum engagement. Filtering to posts with engagement is the single most effective noise reduction, since it drops the automated and ignored.

You choose whether to sort by latest or by top. Latest is right for monitoring — you want the newest mentions. Top is right for research, surfacing the posts that actually shaped the conversation.

Results include the full tweet text plus the author's profile, so a monitoring run doubles as prospect discovery: the person complaining is identified, not just the complaint.

What each result contains

FieldTypeDescription
textstringFull tweet text — the field to run intent detection against.
author.userNamestringHandle, for follow-up or profile lookup.
author.namestringDisplay name.
author.descriptionstringProfile bio, often stating role and company — useful for qualification.
author.followersnumberFollower count, a rough measure of reach.
createdAtdatetimePost timestamp — recency is the whole point for intent work.
likeCount / retweetCount / replyCountnumberEngagement metrics for ranking what resonated.
urlurlDirect link to the tweet, so you can reply in context.

What you get

  • Pull tweets, replies, and profiles matching any keyword or hashtag.
  • Track competitor mentions or monitor buying-intent conversations in real time.
  • Use with AI Intent Detection to identify engaged, in-market buyers.

Example use case

A sales team tracks mentions of a competitor combined with words like 'switching' or 'alternative', filtered to posts with engagement. Intent Detection scores the results, and anything above a seven gets a same-day reply referencing the specific complaint.
See the full workflow: Competitor Intelligence

When to choose this source

Use this for fast-moving public conversation and named-competitor monitoring. For deeper, more considered discussion of a problem, Reddit Pro returns better material. For professional context and firmographics around the person posting, LinkedIn Posts is the better fit.

Limitations worth knowing

Volume is high and quality is uneven. Without engagement or verification filters, a broad keyword returns a great deal of automated and low-value content.

Short posts carry limited context. A tweet tells you someone is frustrated; it rarely tells you their company size or budget, so pair it with profile enrichment before qualifying seriously.

Coverage is limited to public posts. Protected accounts are not included and never will be.

Sarcasm and joke posts routinely defeat automated sentiment reading. Skim intent-scored results before acting rather than trusting the score blindly.

Credit cost

1

credits per result

Only charged for results returned. Failed runs are fully refunded.

Plan availability

All plans

Including Scout ($10/mo)

Start using Twitter / XSee pricing

Frequently asked questions

Can I use advanced search operators?
Yes. The full query syntax is supported, so a query you have refined in Twitter's advanced search can be pasted in and will behave the same way.
How do I cut the noise?
Filter to posts with engagement, and consider restricting to verified accounts. Between them they remove most automated content. Adding a language filter helps again on international keywords.
Can I monitor a specific account's mentions?
Yes — use the 'to' filter for replies directed at an account, or search the handle as a keyword to catch mentions that are not direct replies.
How current are results?
Sorting by latest returns posts from minutes ago. This is the most real-time source available, which is why it suits monitoring rather than one-off research.
Why is it only one credit per result?
Each result is a single short post with its author profile — a small, cheap unit of data. The low price is what makes continuous monitoring practical rather than a luxury.