Trend and seasonality analyses provide insight into patterns within historical data. Examples are seasonal demand trends, spend and buying cycles, yearly, and quarterly trends.

PlanIQ calculates quarterly and yearly trend as a moving average performed over a period of time. Example trend intervals:

  • Quarterly trends are from the past 91 days
  • Yearly trends are from the last 365 days

Trend analysis requires populated values for at least 50% of historical data points.

PlanIQ identifies seasonality. It checks for repeated patterns on weekly, monthly, or yearly timescales. 

For example, to gauge a daily trend, PlanIQ looks at the performance data for each Wednesday of the year. 

PlanIQ identifies seasonality patterns based on the historical data timescale. The table below describes the minimum amount of historical data you need for each type of seasonality analysis. 

Time scaleWeekly seasonalityBi-weekly seasonalityMonthly seasonalityQuarterly seasonalityYearly seasonality
Daily5 weeks6 weeks3 months


3 months3 quarters3 years

3 quarters3 years

Even with positive historical values, you can get negative seasonality values. This is due to how the analysis was carried out:

  1. The trendlines are subtracted from the historical data
  2. The seasonality data is calculated from the remaining historical data points, which can produce negative seasonality points. 

If PlanIQ doesn't identify any seasonal patterns, it can return a noisy output.

Tip: The more historical data you provide, the more accurate your seasonality analysis will be.

To make the trend and seasonality results available:

  1. Add these line items to the forecast results module:
  • PlanIQ_Seasonality_Weekly
  • PlanIQ_Seasonality_Bi_weekly
  • PlanIQ_Seasonality_Monthly
  • PlanIQ_Seasonality_Quarterly
  • PlanIQ_Seasonality_Yearly
  • PlanIQ_Trend_Quarterly
  • PlanIQ_Trend_Yearly
  1. Follow the trend and seasonality instructions detailed in the Create an import action section of Stage 2: Create export and import actions. The results are accessible once the relevant forecast action runs successfully.

The trend and seasonality results are meant for analysis through visualization only, as the values aren't meaningful on their own. The results should be visualized in a chart along with historical data. We recommend that you use a secondary axis for ease of interpretation.

This video shows an example of how to create trend and seasonality charts.

Notes about the video:

  • The example in the video shows separate charts for seasonality and trend analysis.
  • In both charts, trend and seasonality are visualized against ‌historical actuals with a secondary axis.
  • To improve interpretability, since the results are only provided for part of the history, we filter values without data. In addition, we only visualize line items that we expect to contain data (see tables above).