Without tracking how work is actually performed, you are flying blind. You might think your IT team is doing a great job, but you would have no evidence to prove it. This is where Performance Analytics is introduced as the solution: it is the system for defining, measuring, and improving Key Performance Indicators (KPIs) so you can make data-driven decisions instead of guessing.
Jump to a section
A simple way to picture Performance Analytics and Metrics
The Head Chef at a busy restaurant kitchen is the person responsible for making sure every dish that leaves the kitchen meets the standard. To do this, they don't just taste every single plate – that would be impossible. Instead, they rely on a system of key measurements. The Head Chef watches the 'temperature of the grill' to ensure steaks are cooked properly. They check the 'time it takes to plate a dessert' to see if the pastry section is keeping up. They monitor the 'number of orders returned by customers' to spot a problem with a particular recipe. These specific, repeatable measurements – grill temperature, plating time, return rate – are exactly like performance analytics metrics. The Head Chef doesn't care about how many peas are in the walk-in fridge that morning; they care about metrics that directly tell them how well the kitchen is running right now. If the return rate for the special sauce jumps up, they investigate. If plating time for starters slows down, they shift a prep cook to help. This system of picking the right things to measure, watching them over time, and acting on what the numbers show is precisely what Performance Analytics in ServiceNow helps an IT department do. The Head Chef uses these measurements to keep the kitchen performing at its best, just like an IT manager uses Performance Analytics to keep services running smoothly.
The Head Chef also uses a 'recipe book' as a baseline for what should happen. A metric becomes useful only when compared to a target or a previous period. The Head Chef knows that a grill temperature of 200°C is the standard. A deviation from that is a signal to act. In the same way, in ServiceNow, you don't just collect data; you set targets. You compare this month's incident resolution time to last month's. The Head Chef's focus on specific, actionable metrics, rather than just 'everything in the kitchen', is the core lesson for understanding Performance Analytics.
Performance Analytics (PA) in ServiceNow is a tool that helps you measure how well things are working. It is not just about collecting random data; it is about choosing the right data to measure, tracking it over time, and using those measurements to improve. Think of it as the dashboard of your car: without it, you might know you are moving, but you wouldn't know your speed, fuel level, or engine temperature. PA gives IT professionals and managers that same kind of dashboard for their IT services and processes.
The first important concept is the Key Performance Indicator, or KPI. A KPI is a specific, measurable value that shows how effectively a company is achieving a key business objective. For example, a KPI might be 'Average Time to Resolve Incidents'. This is a number that tells you how long, on average, it takes the support team to fix a problem. Another KPI could be 'Number of Critical Incidents per Week'. KPIs are not vague; they are precise numbers that you can track over time. The goal of Performance Analytics is to measure these KPIs and identify trends.
How does it work in ServiceNow? The system relies on data from the existing records in the platform – like Incidents, Changes, Problems, or Service Requests. You create a 'Metric' that defines exactly what you want to count. A Metric is the definition of the calculation. For the KPI 'Average Time to Resolve Incidents', the Metric would be the formula that calculates that average from all closed Incident records. When you set up Performance Analytics, you are essentially telling ServiceNow: 'Look at all my Incident records, find the ones closed this week, calculate the average time from creation to closure, and store that number as a data point.'
But a single data point is not very useful. You need to see it in context. That is where 'time series' data comes in. Performance Analytics takes that calculation and runs it automatically on a schedule – typically daily or weekly. It stores each calculation as a separate data point, creating a history. You can then view these data points on a chart or a dashboard. A line chart might show the 'Average Time to Resolve' over the last 30 days. If the line is going up, it means resolution is taking longer, which is a bad sign. If the line is going down, it means the team is getting faster.
To make this even more useful, you can set a 'Target' for each KPI. A Target is the goal you want to achieve. For example, you might set a target that all high-priority incidents must be resolved within 4 hours. Performance Analytics can then show you not just how you are doing, but whether you are hitting your target. The chart might colour the area below the target line green and the area above it red. This makes it instantly clear if performance is acceptable.
Another key component is the 'Widget'. A widget is a small, reusable piece of content that displays a single chart or score on a dashboard. ServiceNow has pre-built widgets for things like a line chart, a bar chart, a gauge (like a speedometer showing good/bad), and a 'Score' that shows the current value relative to the target. You can add these widgets to a dashboard to create a single view of all your most important KPIs.
Why does this exist? Before PA, managers had to manually run reports, export data to spreadsheets, or rely on gut feelings. This was slow, error-prone, and didn't show trends easily. PA automates this entire process. It collects data continuously, stores it forever, and presents it in an easy-to-understand visual format. It replaces the need for someone to constantly 'run a report' to check on things. It replaces guesswork with facts.
Performance Analytics also includes a feature called 'Break Down'. This allows you to look at your KPI from different angles. For instance, instead of just looking at the average resolution time for all incidents, you can 'break down' that metric by 'Assignment Group'. This shows you which team is the fastest and which team is the slowest. You could also break it down by 'Category' (e.g., Hardware vs Software) or by 'Priority' (e.g., Critical vs Low). This granular view helps you pinpoint specific problems.
The final important concept is the 'Indicator'. In ServiceNow, an Indicator is the engine that collects the data. It is the automated job that runs the metric calculation on a schedule and stores the result. KPIs are the high-level business questions you want to answer (e.g., 'Are we getting faster?'). Indicators are the technical definitions that produce the numbers. You typically design an Indicator to collect data for a specific KPI.
Define the Business Goal (KPI)
Start by deciding what you want to measure. This is your Key Performance Indicator (KPI). For example, 'We want to reduce the average time to resolve high-priority incidents'. This step is crucial because it determines what data you will collect and how you will measure success. Without a clear KPI, your Indicator will have no purpose.
Create the Metric Definition
In ServiceNow, you create a 'Metric Definition' which is the specific calculation formula. You tell the system what to count or average. For the KPI above, you would create a Metric that calculates the average 'Calendar Duration' for Incident records where 'Priority' equals 'High' and 'State' equals 'Resolved'. This Metric is the recipe.
Configure the Indicator
Now you create an 'Indicator'. This is the automated job that runs the Metric on a schedule (e.g., every day at midnight). You specify which table the data comes from (e.g., Incident) and what condition filters the records (e.g., State = Resolved). You also link the Metric you created in Step 2 to this Indicator. This step makes the calculation happen automatically.
Set a Target for the Indicator
A target defines what 'good' performance looks like. For your KPI, you might set a target that the average resolution time should be under 4 hours. You configure this target on the Indicator. The target will be used later to colour the Score green (if met) or red (if not met). Without a target, you have a number but no way to determine if it is good or bad.
Build the Dashboard with Widgets
Go to the Dashboard module and create a new dashboard. Add a 'Widget' (like a line chart or a Score gauge) and connect it to the Indicator you created. Choose the type of widget that best displays the data. For average time, a line chart showing the trend over the last 30 days is useful. For real-time status, a coloured Score widget is best. This is where you visualise the data.
Monitor and Analyse with Breakdown
Once the dashboard is live, monitor it regularly. If you see a problem (e.g., the average time is increasing), use the 'Break Down' feature on the Indicator. Break it down by 'Assignment Group' to see which team is causing the delay. This step is key to turning raw data into actionable insight.
An IT professional, let's call her Priya, is the IT Service Desk Manager for a company with 5,000 employees. Her team handles about 200 support requests per day. She needs to know if her team is providing good service, but she can't watch every single call. She decides to use Performance Analytics.
First, Priya opens ServiceNow and goes to 'Performance Analytics' in the navigation menu. She creates her first KPI: 'Average Time to Resolve Incidents'. She defines the Metric by telling ServiceNow to look at all Incident records with a 'State' of 'Resolved' from the last 7 days and calculate the average of the 'Calendar Duration' field. Calendar Duration is the time between when an incident was opened and when it was resolved, excluding business hours. She schedules the Indicator to run every day at midnight. This is her baseline metric.
Next, Priya wants to know if her team is actually keeping end-users happy. She creates a second KPI: 'User Satisfaction Score'. In ServiceNow, every time an incident is resolved, a survey is sent to the user. The survey has a score of 1 to 5. Priya sets up a Metric that averages all survey scores from closed incidents in the last week. She also breaks this down by the individual technician. Now she can see which technicians get consistently high satisfaction scores and which ones might need training.
Priya then builds a dashboard in ServiceNow. She adds a widget for the 'Average Time to Resolve' line chart. She adds a gauge widget for 'User Satisfaction Score', setting a target of 4.5 out of 5. She adds a third widget called a 'Score' widget for the number of open incidents, with a target of under 50 at any given time. She makes this dashboard her homepage.
Every morning, Priya looks at her dashboard. One morning, she sees the 'Average Time to Resolve' line has spiked up. It used to be 4 hours, but now it is 8 hours. She uses the 'Break Down' feature to see which assignment group is causing the delay. She finds that the 'Network Team' has an average resolution time of 18 hours. That tells her where to focus. She talks to the Network Team manager and learns they are short-staffed due to a holiday. She can now make a data-backed decision to temporarily shift some incidents to another team or to escalate the issue to management to get temporary help.
Without PA, Priya would have only known something was wrong when angry users started calling. With PA, she caught the problem early and had specific data to act on. She also uses PA to report to her own boss. Once a week, she exports a PDF of the dashboard. The chart shows that for the month, the team resolved 95% of critical incidents within the 4-hour target. This kind of concrete data proves the team's effectiveness and helps justify future budget requests.
In a more advanced scenario, Priya might set up 'Predictive Analytics'. This is an add-on that uses historical data to predict future performance. For example, it might predict that if current trends continue, the average resolution time will exceed the target in two weeks. This gives her time to prevent the problem before it happens.
The IT professional's daily work with PA involves:
Defining new KPIs as business needs change.
Monitoring dashboards for unexpected changes.
Using 'Break Down' to investigate performance issues.
Setting and updating targets for KPIs.
Sharing dashboards and reports with stakeholders.
Troubleshooting why an Indicator might not be collecting data (e.g., incorrect filter conditions).
The SNOW-CSA exam tests your understanding of the core concepts of Performance Analytics, not how to configure it in a complex way. You do not need to know how to create an advanced script for an Indicator. You need to know what the parts are, what they do, and how they relate to each other. The exam is very specific about terminology.
Here are the exact concepts the exam loves to test:
The distinction between a KPI, a Metric, an Indicator, and a Widget. These terms are not interchangeable. You must know that a KPI is the business goal (e.g., 'Improve user satisfaction'), the Metric is the calculation definition (e.g., 'Average of satisfaction surveys'), the Indicator is the automated job that runs the metric and stores the data, and the Widget is the visual element on a dashboard. A common trap is to call an Indicator a 'Metric' or vice versa. The exam expects the correct ServiceNow term.
The definition of a 'Time Series' and 'Score'. Know that a Score shows the current value relative to a target, often using a colour (green/red). Time series data is a record of values over time, displayed on line or bar charts.
The relationship between Indicators and Data Sources. An Indicator pulls data from a Data Source (e.g., a table like Incident or Task). You must know that an Indicator has a 'Condition' (which records to include) and a 'Formula' (what to calculate, like count or average).
The concept of 'Break Down'. The exam might ask: 'If you want to see the average resolution time per region, which feature do you use?' The answer is 'Break Down'. They might phrase it as a scenario: 'A manager wants to see satisfaction scores per department. What should they configure?' Answer: 'Break Down on the Indicator'.
Typical exam questions include:
Multiple choice asking which component is used to display data on a dashboard. Answer: Widget.
Scenario-based: 'A KPI is showing a red score. What does this indicate?' Answer: 'The current value is below the target threshold.'
True/False: 'A Metric is the same as an Indicator.' Answer: False. A Metric defines the calculation; an Indicator is the system that runs it.
Drag-and-drop: You might be asked to match terms (KPI, Metric, Indicator, Widget) with their definitions.
Fill-in-the-blank: 'The ______ shows the current performance relative to a target.' Answer: Score.
Traps the exam sets:
They will give you two similar-looking definitions (e.g., 'A calculation to be performed' and 'A job that runs automatically') and ask which one describes an Indicator. The trap is that both sound technical, but only one is the correct, standard definition from ServiceNow documentation. Memorise: Indicator = the running engine.
They might ask about the 'Performance Analytics' module being available. It is a premium plugin. It is not included in every base system. The exam will ask if it requires a separate license or is it included. The correct answer is that it is a plugin that must be activated. They might ask about 'Performance Analytics' vs 'Reporting'. PA is for proactive, time-series monitoring. Reporting is for one-off, historical data. Knowing when to use which is tested.
They might test the order of operations: You first define a KPI, then create a Metric, then configure an Indicator, then add a Widget to a Dashboard. The steps are sequential, and the exam may ask 'What comes first?'
Key definitions to memorise for the exam:
Indicator: An automated data collection job that runs on a schedule.
Metric: A definition of a calculation (e.g., count, average, sum).
Score: A visual representation of performance against a target (often a colour-coded number).
Widget: A reusable visual component (chart, gauge, score) placed on a dashboard.
Dashboard: A collection of widgets showing multiple KPIs.
Breakdown: The ability to view Indicator data segmented by a field (e.g., by category, by group).
Time Series: A sequence of data points measured at successive times.
Target: The defined goal for a KPI, used to determine if performance is good or bad.
A KPI (Key Performance Indicator) is a high-level business goal, while an Indicator is the automated ServiceNow job that collects the data to measure that goal.
Performance Analytics data is time-series data, meaning it is collected on a schedule (e.g., daily) and shows trends over time, not real-time snapshots.
A Score in Performance Analytics is a visual representation showing how current performance compares to a defined Target, often using a green (good) or red (bad) colour.
The feature 'Break Down' allows you to slice Indicator data by a specific field, such as Assignment Group or Category, to identify root causes of performance issues.
A Widget is a reusable visual component (like a line chart or gauge) that you place on a Dashboard to display your KPI data.
Indicators rely on a Condition filter to define exactly which records should be included in the calculation; without a condition, the result will include all records, which is often incorrect.
You must first define a Metric (the formula) before you can configure an Indicator to run it and produce data.
These come up on the exam all the time. Here's how to tell them apart.
Indicator
An automated job that runs on a schedule and collects data.
Stores the results as time-series data points.
Has a status (running, completed, failed) and a last runtime.
Does not define the calculation formula; it uses a Metric Definition for that.
Metric Definition
A definition of a calculation, like average, count, or sum.
Does not run on a schedule; it is just a formula.
Does not store data; it is only the rules for the calculation.
Can be reused by many Indicators.
KPI
A high-level business goal or objective (e.g., 'Improve user satisfaction').
Is not a technical component in ServiceNow; it is a concept.
Does not have a technical configuration of filters or schedules.
The 'what' you want to achieve.
Indicator
The technical tool used to measure the KPI.
A ServiceNow record with configuration settings (table, condition, schedule).
Collects data automatically and stores it.
The 'how' you measure the KPI.
Score Widget
Displays a single current value (e.g., 82%) compared to a target.
Colour-coded (green = meeting target, red = not).
Best for showing real-time status of a KPI relative to a goal.
Does not show trends over time.
Line Chart Widget
Displays a line of data points over a period of time (e.g., last 30 days).
Not inherently colour-coded based on target; shows the actual trend.
Best for showing change and identifying upward/downward patterns.
Shows historical data, not just the current value.
Mistake
Performance Analytics is the same as standard Reporting in ServiceNow.
Correct
Performance Analytics is specifically designed for tracking KPIs over time (time series) with automated data collection, scores, and targets. Reporting is for creating one-off tables and lists from existing data and is not inherently time-series focused.
Both involve displaying data and charts, so beginners naturally assume they are the same thing. The trap is reinforced when users see similar-looking chart options in both areas.
Mistake
An Indicator and a Metric are the same thing.
Correct
A Metric is the definition of what to calculate (e.g., 'average of closed incidents'). An Indicator is the automated job that runs that metric on a schedule and stores the results. You create a Metric, then you configure an Indicator to use it.
The terms are used interchangeably in casual conversation. The ServiceNow documentation uses them very precisely, but beginners hear 'metric' and 'indicator' used loosely and incorrectly in practice.
Mistake
I can create a KPI directly without defining a Metric.
Correct
A KPI is a high-level business goal. To measure it, you must first define a Metric (the formula) and then an Indicator (the process that collects the data). The KPI is the 'what', and the Metric/Indicator are the 'how'. You cannot display a KPI without underlying data collection.
Many beginners think KPIs are just labels they can type into a field. They do not understand the technical pipeline required to produce the numbers.
Mistake
Performance Analytics will show me live, real-time data.
Correct
Performance Analytics data is collected on a schedule (e.g., daily, hourly). It is not real-time. The data you see on a dashboard is from the last time the Indicator ran (the 'last runtime'). It can be a few hours old.
The word 'Analytics' suggests instantaneous analysis to beginners. They expect it to update the second a new record is created. The scheduled nature is a surprise.
Mistake
If I create an Indicator, it will automatically collect data from all records in the table.
Correct
An Indicator has a 'Condition' filter that specifies which records to include (e.g., 'State is Resolved'). If you do not set a condition, it will include all records, which often produces misleading results (like including unassigned or open records in a 'resolution time' calculation). You must define the condition carefully.
Everyday software like Google Analytics tracks 'everything'. Beginners assume ServiceNow works the same way and do not realise they must define the data scope.
Reveal each answer, then mark whether you got it right. Score 60%+ to unlock the next chapter.
Yes, Performance Analytics is a separate, purchased plugin. It is not included in the base ServiceNow platform. You need to activate the 'Performance Analytics' plugin (com.snc.pa.*) and may need additional entitlements depending on your subscription.
A 'Score' widget displays a single number (e.g., 85%) with a colour (green/red) based on the target. A 'Score Bar' widget shows a horizontal bar representing the value relative to the target, often used in lists or for multiple KPIs side-by-side.
Yes, you can create Indicators on any table in ServiceNow. You simply select the table as the Data Source when configuring the Indicator. Common examples include tracking 'Number of Change Requests by State' or 'Average time to close a Problem record'.
Indicators run on a schedule you define. By default, they might run once daily. You can configure them to run every hour if needed, but real-time (second-by-second) updates are not possible with standard Performance Analytics.
If you delete an Indicator, all the historical data collected by that Indicator will also be deleted. The Widget that was using that Indicator will then show an error or no data. It is generally better to 'deactivate' an Indicator (stop it from running) rather than delete it, to preserve history.
Yes, you can export dashboards and individual widgets as PDF files. The data itself (the time-series points) can be exported as a CSV from the Indicator record, allowing you to analyse it in Excel or other tools.
You've finished Performance Analytics and Metrics. Continue through the SNOW-CSA study guide to build a complete picture of the exam.
Done with this chapter?