Alternate States
in Qlik Sense

What is an Alternate State?

By default, a Qlik app has one selection state: the user clicks "Southeast" and every object reflects that selection. Alternate States break this constraint, enabling different objects to operate in independent, simultaneous selection contexts.

The Problem It Solves

Consider this scenario: an analyst needs to compare the performance of two products with different filters applied to each, without creating a second app, without opening two tabs, without hardcoded Set Analysis values. Alternate States is Qlik's native answer to this problem.

Both states coexist in the same app at the same time. A single chart can display both values side by side. Users make selections in each state completely independently.

Set Analysis Syntax with States

// Default state - same as {$}
Sum({$} Revenue)

// Alternate state by name
Sum({Scenario_B} Revenue)

// Alternate state + additional modifier
Sum({Scenario_B<Status={'Active'}>} Revenue)

// Difference between the two states
Sum({$} Revenue) - Sum({Scenario_B} Revenue)

// YoY % comparison between states
Num(
    (Sum({$} Revenue) / Sum({Scenario_B} Revenue)) - 1,
    '+#,##0.0%;-#,##0.0%;0.0%'
)

State vs. Hardcoded Set Analysis

// Always ignores ALL selections
// Not interactive - never changes
Sum({1<Region={'South'}>} Revenue)

// Value is fixed regardless of
// what the user selects

// Uses selections of state Scenario_B
// The user can change those selections!
Sum({Scenario_B} Revenue)

// Value updates as user interacts
// with objects assigned to that state

Why Use Alternate States?

Alternate States are not a feature for simple scenarios. They exist to solve comparative analysis problems that would otherwise require complex, fragile workarounds.

When NOT to Use Alternate States

• When the comparison data is fixed and comes from the model (e.g., a separate "Budget" column) - use simple Set Analysis with a field modifier.
• When the comparison is a fixed period (YTD vs LY) - use the dynamic period expressions from Part 2.
• When you need more than 3-4 states - maintenance complexity outweighs the benefit.
• When end users do not understand the concept of "two independent filters" - the UX becomes confusing.

Creating and Configuring States

States are created in the Qlik Sense interface and then referenced in any Set Analysis expression by the exact name you defined.

Step 1 - Create the State in the Interface

Step 2 - Create Filter Panes for Each State

Step 3 - Reference in Set Analysis

// The comparison chart stays in the default state
// It simply has two measures:

// Measure 1 - uses default state selections
Sum({$} Revenue)
// Label: "Revenue - Selection A"

// Measure 2 - uses Scenario_B state selections
Sum({Scenario_B} Revenue)
// Label: "Revenue - Selection B"

// Measure 3 - absolute difference between both states
Sum({$} Revenue) - Sum({Scenario_B} Revenue)
// Label: "Variance A vs B"

// Measure 4 - percentage variation
Num(
    (Sum({$} Revenue) / Sum({Scenario_B} Revenue)) - 1,
    '+#,##0.0%;-#,##0.0%;0.0%'
)

Production Patterns with Alternate States

These battle-tested patterns combine Alternate States with the techniques from the previous articles: modifiers, P(), dollar-sign expansion, and dynamic periods.

Pattern 1 - Dynamic Regional Benchmark

// Setup: Two Region filter panes (State $ and State Region_B)
// User chooses any region in each filter independently

// Dynamic label showing which region is active per state
'Selected: ' & Concat({$} DISTINCT Region, ', ')
'Benchmark: ' & Concat({Region_B} DISTINCT Region, ', ')

// Revenue comparison
Sum({$} Revenue)           // Region A
Sum({Region_B} Revenue)  // Region B

// Variance
Num(
    Sum({$} Revenue) / Sum({Region_B} Revenue) - 1,
    '+#,##0.0%;-#,##0.0%;0.0%'
)

Pattern 2 - Cohort Analysis with State + P()

// Default state: customers who bought Product A
// State Cohort_B: customers who bought Product B
// Question: what is the cross-revenue between both groups?

// Customers exclusive to State A (bought A but not B)
Count(
    {$<CustomerID =
        P({$} CustomerID)
        -
        P({Cohort_B} CustomerID)
    >}
    DISTINCT CustomerID
)

// Customers in BOTH states (bought A and B)
Count(
    {$<CustomerID =
        P({$} CustomerID)
        *
        P({Cohort_B} CustomerID)
    >}
    DISTINCT CustomerID
)

// Revenue from overlapping customers
Sum(
    {$<CustomerID =
        P({$} CustomerID) * P({Cohort_B} CustomerID)
    >}
    Revenue
)

Pattern 3 - State + Dynamic Period

// State A: South Region, user's default period
// State Period_2: North Region, different period

// YTD for default selection
Sum({$<Year={$(=Max({1}Year))},
     Month={"<=$(=Month(Today()))"}>} Revenue)

// YTD for alternate state - same period logic, different selections
Sum({Period_2<Year={$(=Max({1}Year))},
              Month={"<=$(=Month(Today()))"}>} Revenue)

Pattern 4 - What-If Scenario

// Default state: actual performance
// State WhatIf: simulated scenario with different filters

// Actual
Sum({$} Revenue)

// Simulated: WhatIf state + Premium segment only
Sum({WhatIf<Segment={'Premium'}>} Revenue)

// Impact (absolute)
Sum({$} Revenue) - Sum({WhatIf} Revenue)

// Impact (percentage, signed)
Num(
    (Sum({$} Revenue) - Sum({WhatIf} Revenue)) /
    Abs(Sum({WhatIf} Revenue)),
    '+#,##0.0%;-#,##0.0%;0.0%'
)

Quick Reference and Common Mistakes

Full Syntax Reference

Common Mistakes and Fixes

Production Checklist

• Clear naming: Comparison_Period is better than State2
• Document which objects are in which state - use different background colors on filter panes to visually signal the state
• Limit to 2-3 states per app. More than that usually means the solution architecture needs rethinking
• Always add a "Clear State" button for each alternate state linked to ClearAll() for that state
• Use Concat({StateName} DISTINCT Field, ', ') in text objects to show users what is selected in each state
• Test with empty alternate state: what happens when the user has not selected anything in State B? Does the expression return total or zero?
• In QlikView, Alternate States exist but the configuration path differs: Tools > Document Properties > Alternate States