Marketing performance and impact analytics system
The team compares the marketing data with the customer's sales and other agreed-upon results. The report clearly explains what the chosen measurement method allows for.
Advertising platforms, websites, CRMs, sales and customer data are combined with periodic and uneven rules. The team notices problems too late, and the comparison of channels and campaigns becomes unreliable.
How the solution works
- Agree the client's goal and proper result criteria
- Obtain compatible spending, channel and result data
- The analyst verifies duplicate records, periods and missing channel data.
- Evaluate attribution or result of a properly planned study
- Capture the limits of the conclusion, the solution, and the lesson useful for the next job
Key challenges
- Campaign data split between channels and reports
- Campaign results are difficult to relate to the impact of a client's business
- Campaigning and creative learning do not return to other projects
Solution capabilities
Definition of the result
Query, proper potential customer, sale and return retain different meanings. The purpose of communication may require audience understanding or behavioral verification.
Data reconciliation
Tags, currency, period and source of actual result are combined before comparison. Conversions assigned to themselves by different platforms are not aggregated as unique sales.
Data included in the analysis and its boundaries
Missing consent, non-transferred sales and modelled result are noted. Data shortages are not reported as zero channel effects.
The Impact Method
The result shown by the attribution rule is separated from the causal effect. The experiment has a pre-agreed comparison, sufficient scope and uncertainty assessment.
Decision Memory
The hypothesis, costs, audience, method used and the limits of the result are preserved. Insufficient proof is also recorded, instead of being turned into a winning campaign.
Business context
- Campaign data split between channels and reports
- Advertising platforms, websites, CRMs, sales and customer data are combined with periodic and uneven rules. The team notices problems too late, and the comparison of channels and campaigns becomes unreliable.
- Campaign results are difficult to relate to the impact of a client's business
- Reports are limited to reach, clicks, or platform conversions regardless of sales, queries, or customer quality. The client has difficulty justifying the investment, and the agency optimizes based on interim, not final indicators.
- Campaigning and creative learning do not return to other projects
- Hypotheses, test results, audience insights and inoperable solutions remain in the presentations or in the memory of individual employees. The same experiments are repeated, and the organization's knowledge grows slower than the amount of data being generated.
- The result of the campaign is associated with customer decisions
- It is important for the customer to understand what marketing data says about his sales and other agreed goals. Agreed sources and measurement limits allow to distinguish the observed relationship from reasonable impact. Such analysis helps to plan the next campaign and to argue for proposed budget changes.
Core features
- Definition of the result
- Data reconciliation
- Data included in the analysis and its boundaries
- The Impact Method
- Decision Memory
Key integrations
- Advertising and Publishing Channels
- Costs, campaigns, real attribution rules and model data tags.
- Client result source
- For the agreed purpose, proper sale, quality of the request or result of an audience inquiry.
Potential impact (%)
The ranges indicate an illustrative relative change in the metric under the stated assumptions. Results depend on the starting position and actual use of the solution. Percentages for different metrics must not be added together.
Report Data Matching Work
16–42%Decreasing
This illustrative scenario assumes that 40-70% of information searches and repeated cross-checks can be addressed. That share is assumed to fall by 40-60%. Company data is needed to verify both the addressable workload and the resulting change.
Measure the analyst's hours of work to collate data from a comparable client period and prepare a conclusion.
Misstatement of measurement altered budget conclusion
12–42%Decreasing
This illustrative scenario assumes that 30-60% of errors can be addressed through the data and rule checks described. That share is assumed to fall by 40-70%. Company data is needed to verify both the addressable share and the resulting change.
Compute budget recommendations already made, corrected for data error or submission of attribution as a proven additional effect.
Sales revenue after marketing budget change
1–5%Increasing
In the example scenario, 15-30% of the original indicator value is associated with the change in campaign budget allocation implemented. This share is projected to grow by 5-15% without other conditions changing.
After the budget change implemented, the actual sales revenue and margin are compared. The method of attribution or comparison used is indicated; the same purchase is not duplicated for channels several times.
Conditional calculation scenarios. The assumptions have not been validated against client measurements.
When this solution is relevant
- Advertising channels attribute the same sales to themselves, so the overall score in the reports is duplicated.
- Campaign data is difficult to link to customer orders, revenue, and other agreed-upon business indicators.
Project scope and implementation
The analysis is based on an agreed-upon source of sales or other business result. If such data is not collected, there is first a consensus on how to capture and link them to marketing. Sales attributed by advertising platforms are not aggregated without verification of duplication.
Further development options
- Controlled experiments or longer-term models only after proper design, data scope and inference uncertainty has been assessed.