Comparing Performance across Time Periods

Rockerbox can be leveraged to understand how your marketing performance and user behavior varied during two different time periods.

Examples of when you may want to do analysis across time periods include:

  • New product launches
  • Spend Heavy ups
  • Change in your marketing mix
  • Testing offline channels

Below we outline four analyses that you can leverage in comparing across time periods:

1. Impact to CPA/ROAS

2. Impact to Channel Mix

3. Impact to Top and Bottom funnel channels

4. Impact to Path to Conversion and Time to Convert


Priority Questions to Answer


Marketing Performance

  • Impact to CPA/ROAS- how did CPA and/or ROAS change between time periods
  • Impact to CPA - if spend heavy up - how did CPA change due to heavy discounting coupled with increased spend (and mitigating against diminishing returns)
  • Impact of discounting- if used different discounts across channels, did that change the improvement on CPA of each channel?

User Behavior

  • Changes to the path to conversion- were users taking more vs less time to convert and interacting with different marketing touchpoints?
  • Impact to marketing mix- were users converting from different channels?
  • If launching a new offline channel or doing a branding heavy up - did this result in users entering the funnel from different channels (first touch) vs converting from different channels (last touch)?

📘

Choosing a control period

Before you begin any analysis you need to first choose your control period.

This is the time period you will compare performance during your current time period against.

When choosing your control period consider the following:

  • Choosing a time period without any significant changes to your marketing mix ( heavy-ups, new channel launched)
  • For sales with a longer ramp time (i.e. BFCM) - you may want to choose two control time periods (one pre-marketing heavy up and one during heavy up)

1. Impact to CPA/ROAS


Priority Questions to answer:

  • Impact to CPA/ROAS- how did CPA and/or ROAS change between time periods
  • Impact to CPA - if spend heavy up - how did CPA change due to heavy discounting coupled with increased spend (and mitigating against diminishing returns)
  • Impact of discounting- if used different discounts across channels, did that change the improvement on CPA of each channel?

Methodology:


Resulting Metrics:

  • % change in CPA/ROAS by channel
  • % change in CPA/ROAS relative to % change in spend

2. Impact to Channel Mix


Priority Questions to Answer

  • What was the impact to your marketing mix- were users converting from different channels?

Methodology:

  • Using buckets breakdown to look at % of overall normalized conversions
  • Identify any changes at the channel level

See Marketing Performance Report (Buckets Breakdown) for more details on this report type.


Resulting Metrics:

  • % of conversions by channel

3. Impact to Top and Bottom funnel channels


Priority Questions to Answer

  • Did offline channel launch or branding heavy up result in users entering the funnel from different channels (first touch) vs converting from different channels (last touch)?

Methodology:

  • Using paths view in UI to see conversions by first vs last touchpoint
  • Step-by-step guide at Path to Conversion to replicate analysis

Resulting Metrics:

  • % of conversions by channel for last touch vs first touch
  • % change in first-touch conversions relative to % change in spend

4. Impact to Path to Conversion and Time to Convert


Priority Questions to Answer

  • How did our users' path to conversion change- were users taking more vs less time to convert and interacting with different marketing touchpoints?

Methodology:

  • Use paths view in UI to see average time to conversion and look at any changes
  • Be sure to filter for new vs existing customers

Go to Path to Conversion (doc:paths) for more detail


Resulting Metrics:

  • Avg # of days to convert
  • Avg # of marketing touchpoints
  • (Both overall and by channel position)

Did this page help you?