ASK IAN - User Guide for working with Trackers
This guide explains how to use ASK IAN for different types of wave-over-wave comparisons in tracker surveys.
Quick Reference
| What you want | Example queries |
| Latest vs previous wave | "What changed since last wave?" |
| Same month, year apart | "Year over year comparison" |
| Full year vs full year | "Compare 2022 to 2023" |
| Year to date | "YTD comparison" |
| Last 3 months vs prior 3 | "Quarterly comparison" |
| Specific waves | "Compare January 2023 to June 2023" |
Comparison Types Explained
1. Last Wave Comparison
Compares the most recent wave to the immediately preceding wave.
Example queries:
- "What changed since last wave?"
- "What's new this month?"
- "Show me the latest changes"
- "Any significant movements recently?"
- "What moved since last time?"
What it compares: `July 2023` vs `June 2023`
Best for: Monthly reporting, catching immediate shifts
2. Year-over-Year (YoY)
Compares the latest wave to the same month/period one year ago.
Example queries:
- "Year over year comparison"
- "How does this July compare to last July?"
- "YoY changes in awareness"
- "Same period last year comparison"
- "What's different versus a year ago?"
What it compares: `July 2023` vs `July 2022`
Best for: Seasonal businesses, annual trend analysis, removing seasonality effects
3. Full Year Comparison
Aggregates ALL waves from one calendar year and compares to ALL waves from another year.
Example queries:
- "Compare 2022 to 2023"
- "How did this year compare to last year?"
- "Full year comparison"
- "What changed between 2021 and 2022?"
- "Annual comparison"
- "Show me the year-on-year summary"
What it compares: `\[Jan-Dec 2023]` vs `\[Jan-Dec 2022]`
Best for: Annual reviews, board presentations, strategic planning
4. Year-to-Date (YTD)
Compares all waves so far this year to the same months last year.
Example queries:
- "Year to date comparison"
- "YTD performance"
- "How are we doing compared to this point last year?"
- "Same period comparison"
- "First half this year vs first half last year"
What it compares: `\[Jan-Jul 2023]` vs `\[Jan-Jul 2022]`
Best for: Mid-year reviews, tracking annual targets, fair comparison when year isn't complete
5. Quarterly Comparison
Compares the last 3 months to the prior 3 months.
Example queries:
- "Quarterly comparison"
- "Last quarter vs previous quarter"
- "Q2 vs Q1 changes"
- "What changed in the last 3 months?"
- "Rolling quarter comparison"
What it compares: `\[May-Jul 2023]` vs `\[Feb-Apr 2023]`
Best for: Quarterly business reviews, smoothing out monthly fluctuations
6. Custom Comparison
Specify exact waves or periods to compare.
Example queries:
- "Compare January 2023 to January 2022"
- "What changed between March and July 2023?"
- "Compare Q4 2022 to Q1 2023"
- "Show changes from August 2020 to July 2023"
- "Compare the first wave to the latest wave"
Best for: Ad-hoc analysis, specific business questions, pre/post campaign analysis
Combining with Metrics
You can combine time comparisons with specific metrics:
Examples:
- "What happened to brand awareness in the last year?"
- "YoY comparison for consideration"
- "How has familiarity changed since 2022?"
- "Quarterly changes in purchase intent"
- "Compare satisfaction between Q1 and Q2"
Filtering by Brand
For surveys with brand loops, you can ask about specific brands:
Examples:
- "What changed for Brand X since last wave?"
- "Year over year comparison for Brand Y"
- "How has Brand Z's awareness changed this year?"
- "Compare all brands between 2022 and 2023"
Understanding Results
Significance Levels
IAN highlights statistically significant changes:
| Indicator | Meaning |
| 99% | Very high confidence (p ≤ 0.01) - Almost certainly a real change |
| 95% | High confidence (p ≤ 0.05) - Industry standard threshold |
| 90% | Moderate confidence (p ≤ 0.10) - Worth noting but not conclusive |
| 85% | Low confidence (p ≤ 0.15) - Directional only |
Reading Changes
↑ 3.2pp = Increased by 3.2 percentage points
↓ 1.5pp = Decreased by 1.5 percentage points
Green highlighting = Significant increase
Red highlighting = Significant decrease
Tips for Better Results
Be specific about metrics - Instead of "what changed?", try "what changed in awareness?"
Specify the timeframe - "Last wave", "this year", "since 2022" helps IAN choose the right comparison
Ask follow-up questions - "Why did Brand X's familiarity drop?" or "Break this down by age group"
Use natural language - IAN understands conversational queries, not just keywords
Request specific brands - If you're only interested in certain brands, mention them
Example Conversation
You: What are the main changes in brand familiarity this year compared to last year?
IAN: Runs a full-year comparison on familiarity metrics, shows significant changes by brand
You: Interesting. Can you break down the Brand X changes by age group?
IAN: Runs a crosstab of Brand X familiarity by age
You: What about just comparing the last quarter?
IAN: Runs the last quarter's comparison on the same metrics
Disclaimer
As with all AI systems, very occasionally IAN may not quite summarise things in exactly the expected way, so it will remain important to include a checking process for any output that will be put in front of clients.