Campaign intelligence for the next test.
A defined campaign decision needs a defined learning question. NOMAS reviews objectives, audiences, creative claims, spend and response in context to identify which message, audience or channel hypotheses merit the next test.
Scope, authorized access, method and outputs follow the decision your team faces and the evidence available.
What could the next finding change?
The decision
Clarify whether you need to refine a message, review an audience or test a channel. The learning question follows that choice.
Objective, period and audience
Agree the campaign objective, the period under review and how the relevant audience is defined.
Response and conversion
Define what counts as a response or conversion, and the reporting window used. The same label may refer to different events.
Materials and reports
Identify authorized campaign materials, offers, spend and response reports, their sources and what they omit.
Measurement constraints
Record known tracking gaps, attribution rules and limitations before reading apparent differences as performance.
A difference is useful only when its context is understood.
Messages, creative claims, offers, spend and response are reviewed together. Before interpreting a difference, the comparison considers:
| Comparison factor | What to review | Why it matters |
|---|---|---|
| Periods | Timing, duration and reporting windows. | Different periods may contain different conditions. |
| Audiences and objectives | Who was reached and the purpose of the campaign. | A different audience or objective can change the meaning of the response. |
| Messages, offers and spend | The claims, offer terms and recorded spend behind each response. | Several factors may have changed together. |
| Attribution and conversion | Rules, event definitions and reporting windows. | Unlike definitions do not support a like-for-like comparison. |
| Tracking coverage | Missing events, unavailable channels and known gaps. | What was not recorded remains an evidence limit. |
Observations become specific hypotheses, not immediate declarations of a winning message or channel.
An evidence matrix and a decision-led learning agenda.
Depending on the agreed scope, outputs may include:
Campaign evidence matrix
Available observations, their sources, the comparison context and measurement caveats.
Message, audience and channel hypotheses
Specific explanations or options to examine, separated from what was observed.
Test priorities
The questions worth addressing first, considering decision relevance, available evidence and feasibility.
Learning agenda
For each question, the evidence needed and what finding would change the team’s decision.
Agree the next test without assuming a causal answer.
Observational comparisons alone do not isolate causal impact. Estimating that impact would require an appropriate design and sufficient evidence.
A bounded test may be designed when the agreed scope, evidence and feasibility allow. Its question and limits are agreed before interpreting the findings; a causal design is not assumed in every engagement.
NOMAS agrees the method, access and outputs with you. Execution decisions and implementation remain with your team.
Two messages, more than one difference.
Different responses do not establish a winning message
Imagine two messages about the same offer show different responses, but the audiences and periods differ too. The observations do not establish which message caused the difference or which channel is better.
The next research question is how to test the message difference while controlling the other differences where feasible. What can be held comparable, and what evidence is needed, must be agreed before treating a result as an answer.