<aside> 💡 Alert is an incredibly useful tool for businesses that want to track their key performance indicators (KPIs) and quickly take action based on the data they see.

</aside>

Setting alerts:

Setting alerts is done through the KPI view. Choose your KPI and Edit your Alert

Choose:

1- Condition: Above/below to track the increase or decrease of the KPI

The below condition can be used during an ongoing crisis to track whether there has been a decrease in trends on a specific KPI.

2- Check [the frequency of alert]

Hourly: The KPI is checked every 2 hours. If the alert is triggered, users will be notified via email and notification.

Daily: The KPI is reviewed daily at 22:00 (UTC+3). If the alert is triggered, users will be notified by email and notification.

Weekly: The KPI is checked weekly, specifically on Sundays at 22:00 with a UTC+3 timezone. Users will receive a notification and an email if the alert is triggered.

3- Specify the email recipient [up to 10 email recipients]

Untitled

Key Information provided in alerts:

I1 / Percentage of change

Indicates the percentage of the difference between the old values and new values

Hourly: compare with last 24H hourly average

Daily: compare with last week's average daily

Weekly: compare last Month's weekly average

Formula: [100 * (recent_data - past_data) / past_data ]

How to interpret this indicator?

In case of an increase, two scenarios can be interpreted:

I2 /Recent Trends:

Fig1: Trends chart

Fig1: Trends chart

Indicates the volumetric change of data in the comparison window and the threshold set by the user.

How to interpret it?

This chart showcases 2 key insights:

2- Amplitude: How far the new value is from the threshold set by the user?

3- Speed of change: How fast is the increase or the decrease? [In Fig 1 the chart value is above the average threshold but it tends to decrease over time]

Edge Case study:

Untitled

In this case, the KPI had an hourly increase over the threshold compared to the last hour but checking the graph this trend is not an anomaly or a crisis but it is more of a repetitive pattern in that hour.

I3/ Hot topics: Root cause understanding

Hot Topics provides information on the top topics associated with new data. It can be used to understand the primary reasons for changes in the data.

Edge case: If the data is not related or contains insufficient data, this part will indicate an error that no top topics were detected.

â›” Indications

Alert cannot be triggered in these cases or can contain some empty indicators, cases are summarized in this table:

Case ID Case
W1 if new data < 10 reviews
W2 if there is no old data to compare with.
W3 TOP topics can be empty in case new data is not enough to generate significant topics
W4 Trends charts can have no continuous line in case of fewer data points.