Signal, no signal, and anti-signal

  • Release version: Washingtondc
  • Updated February 1, 2024
  • 2 minutes to read
  • Summarize
    Summarized using AI
    This content was generated using new OpenAI-powered functionality. Results are provided on an as is basis and are not guaranteed to be accurate or complete.

    Summary of Signal, No Signal, and Anti-Signal

    The KPI Signals application helps identify variations in Key Performance Indicators (KPIs) through the generation of signals and anti-signals. Signals indicate abnormal variations, while anti-signals indicate stability within a workflow. Understanding these elements allows for timely decision-making regarding process improvements.Key FeaturesSignal Generation:Signals are triggered by specific behaviors, such as outliers, short runs, and long runs. An outlier is defined as a score outside three standard deviations, while a short run consists of four consecutive scores near the limits, and a long run consists of seven consecutive scores on the same side of the central line.Anti-Signal Generation:An anti-signal is generated when no signals are detected for a specified duration, calculated using a formula involving the baseline calculation period and an anti-signal factor. The default is set to 28 days for daily indicators.No Signal Notification:When no signal is detected for a shorter duration than that required for an anti-signal, a "No signal" message is displayed, but no notifications are sent.Key OutcomesBy utilizing the KPI Signals application, ServiceNow customers can effectively monitor their workflows. The presence of signals highlights areas requiring attention, while anti-signals confirm stability, allowing for informed decision-making to enhance process efficiency. Notifications ensure relevant stakeholders are aware of significant changes or stability in performance metrics.

    Show full answer Show less

    When KPI Signals detects abnormal variation in the scores of a KPI, it generates a signal. When KPI Signals does not detect abnormal variation for a significant amount of time, it generates an "anti-signal." The anti-signal lets you know that your workflow is under control.

    Both the presence and absence of signals conveys information:
    • The presence of signals lets you know that a workflow has changed, or at least is not statistically stable.
    • The long-term absence of signals indicates that a workflow is statistically stable. With this information you can decide to launch a process improvement initiative, for example.
    Note:
    By default, dashboards that show indicators with KPI Signals enabled have insight cards for signals that get generated. For more information, see Proactive analytics insights on dashboards.

    Signals

    The KPI Signals application considers the following behaviors to 'signal' special cause variation:
    Outlier
    Every score beyond a three standard deviation (3-sigma) upper or lower limit.
    Short run
    Four consecutive scores on the same side of the central line, with three of the scores close to the upper or lower limit. "Close to the limit" is defined as outside two standard deviations.
    Long run
    Seven consecutive scores on the same side of the central line. That is, seven scores in a row all greater than or all less than the average value.
    Note:
    The KPI Signals app counts outliers (dismissed or not) when calculating whether a short or long run has occurred.

    In the following example, you see a "Long run" signal: seven consecutive scores above or below the central line.

    KPI Signals panel open on the KPI Details tab, showing a Long Run signal of seven consecutive scores all greater than or all less than the average value.

    Note:
    The KPI Signals application can detect signals for an indicator with no more than one first-level breakdown element. The application does not support multiple breakdown elements or a second-level breakdown.

    Anti-signals

    The application issues an anti-signal when it has not detected a signal for a length of time equal to the following formula:
    Length of time to calculate the baseline for the KPI * Anti-signal factor

    The application takes 14 days by default to calculate the baseline for a daily indicator. The default anti-signal factor is two. Thus, the default length of time to detect an anti-signal on a daily indicator is 28 days. For more information about setting the time to calculate a baseline, see Configure signal detection. For more information about setting the anti-signal factor, see Configure signal notifications.

    In this example, the default configuration values apply to a daily indicator. The application has not detected a signal for 28 days, so it shows a 28-day anti-signal.

    KPI Signals showing an anti-signal, where no signals have been detected for an extended period of time

    Although responsible users are notified when an anti-signal is generated, no action is required.

    No signal

    Usually no signal is detected for a shorter amount of time than it takes to detect an anti-signal. In this case, the application shows a simple "No signal" message. No notifications are sent and no action is necessary.

    In the following example, no signal is detected for five days after the baseline was most recently reset. The baseline has been reset three times in the period being shown.

    KPI Signals showing no signal for a KPI: neither a signal nor an anti-signal