Explore: How to Use Spike Detection & Analysis

Separating the signal from the noise to help you focus on what’s truly important.

Whitney W. avatar
Written by Whitney W.
Updated over a week ago

Learn more about using Spike Detection & Analysis to help you identify statistically relevant events across news and social.


What is it?

A new AI-powered insights feature on the Mentions Trend widget in Explore: Spike Detection and Analysis.

Identifying statistically relevant shifts in news and social media conversation, with summary explanations, and insight drill-ins to view mentions and additional spike analytics.

We’ve also updated our Spike Alerts with the new contextual analytics across email, in-app, and Slack! Spike Alerts include AI-powered summary explanation identifies statistically relevant shifts in news and social media conversation, with added context via plain language summary explanations that include both the cause and reason for a spike.


What’s the value?

The size of data and its pace of change overwhelms our human abilities to accurately and quickly analyze it. We struggle to separate the signal from the noise. What should we pay attention to? What should we dismiss? What should we ignore?

Powered by AI, we are now able to identify statistically relevant shifts in news and social media conversations, helping you focus on what’s truly important.

Summaries will explain why volume spikes have occurred and their relative significance.

Taking that analysis a step further, we provide the full list of mentions and relevant analytics for a specific spike, for a faster deep dive.

By automating some of these processes, we’re attempting to alleviate your lack of time and resources, as well as pointing you in the direction of the most crucial information you should be focusing on, to help with becoming data-driven.


How does it work?

In order to perform spike detection and analysis, users will find lightbulb icons over the spikes in the Mentions Trend widget.

Spike detection will then run and highlight any statistically significant spikes in the trend graph. This analysis is powered by the same technology that drives the Spike Detection Smart Alerts.

Spikes will be indicated by a lightbulb over the top of the spike. Click a bulb to launch the first analysis window.

  1. Light bulb - indicates a statistically significant spike. Click to gain insights.

  2. Spike Analysis - offering key insights related to a particular spike, including

    1. Total mentions and date

    2. Multiple more above average

    3. The following all feature mentions and comparisons to average

      1. Source driving spike

      2. Mentions spike total and average

      3. Top sources

      4. Top terms

      5. Sentiment

      6. Top Location

      7. Content with the highest reach for spike

  3. View more insights - click to open a slide-out featuring Insights, Mentions, and deeper Analytics for the spike

When you click View more insights, a slide-out content stream will appear, filtered to the day of the spike. This slide-out features three tabs, Insights, Mentions, and Analytics, plus a Search tip option at the top.

The Insights tab features the same information featured in the spike pop-up from clicking on the light bulb, however the content with the highest reach features the top three pieces of content.

The Mentions tab features a content stream that offers all the same functionality of the main stream on the page, including AI tags, search, and download options.

The Analytics tab features four widgets filtered to the day of the spike, including mentions trend by hour, top locations, top keywords, and sentiment.

The analytics widgets behave the same as they they do on the Overview tab in Explore.

We’ve also updated our Spike Alerts with the new contextual analytics across email, in-app, and Slack.

Email

In-app

Slack


💡 Tip

Need more help? Feel free to reach out to us via Live Chat or check out our Customer Community.

Find answers and get help from Meltwater Support and Community Experts.


Did this answer your question?