Internal Evidence Plus External Trends
Combine proprietary performance data with outside forces before planning.
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 6
- Confidence
- 98%
Internal Evidence Plus External Trends is a planning method that pairs proprietary performance information with the outside environment. Begin with previous campaigns, channel metrics, audience segments, timing, strategic documents, and known outcomes. Analyze the internal material for patterns, then research conferences, seasonal demand, industry shifts, policy changes, and other forces that could explain or alter those patterns. Link each significant external factor to the relevant internal evidence and mark whether the connection is proven, plausible, or merely speculative. Use the combined view to develop channel priorities, campaign timing, target personas, and monitoring assumptions for the next plan. This avoids treating historical numbers as if they occurred in a vacuum and helps the planner distinguish repeatable performance from results caused by temporary conditions.
Origin
Extracted from Marketing Against The Grain through a Grok 3 test using synthetic marketing campaign data for a hypothetical healthcare software company.
Core principles
- 01Internal results explain what happened within the organization.
- 02External trends reveal conditions that may have influenced those results.
- 03Plans improve when evidence spans performance, seasonality, personas, events, and policy shifts.
- 04Synthetic data can test the method, but real internal data makes it more useful.
How to run it
- 1
Assemble internal evidence
Collect campaign performance, channel results, personas, timing, goals, and prior strategic decisions.
Pro tip Include failed campaigns and weak channels, not only winners.
Watch out Inconsistent definitions across datasets can create false patterns.
- 2
Extract internal patterns
Identify differences by audience, channel, offer, season, and campaign type.
Pro tip Flag outliers separately from recurring trends.
Watch out Do not infer future causality from correlation alone.
- 3
Map external conditions
Research relevant events, seasonality, market movements, policy changes, and industry shifts.
Pro tip Match each factor to the exact period and audience it could affect.
Watch out Broad trend lists are not useful unless connected to the business context.
- 4
Connect evidence and environment
Assess which external factors may explain internal results and state the strength of each connection.
Pro tip Label connections as evidenced, plausible, or speculative.
Watch out Avoid retrofitting a convenient outside explanation to every result.
- 5
Construct the plan
Set future channel, timing, audience, and campaign priorities from the combined analysis.
Pro tip Attach each major recommendation to both an internal signal and an external assumption.
Watch out Do not build a detailed plan around an external condition that cannot be monitored.
- 6
Monitor assumptions
Track whether the expected external factors occur and update the plan when conditions change.
Pro tip Assign leading indicators to the most consequential assumptions.
Watch out A plan becomes stale when its environmental assumptions remain implicit.
In the wild
The hosts tested Grok on synthetic campaign data for software sold to doctors. The model analyzed performance, recognized seasonality and personas, then incorporated healthcare conferences, seasonal factors, industry shifts, and possible policy changes into a future campaign plan.
→ The output connected historical campaign evidence to market conditions that could influence future execution.
A subscription company combines conversion and retention data with holiday timing, competitor promotions, consumer-confidence changes, and a forthcoming regulation. Recommendations are retained only where internal performance and an observable outside condition support them.
→ The company avoids repeating a campaign whose apparent success depended on a temporary seasonal surge.
Common mistakes
Using internal data in isolation
Performance metrics can be misread when seasonality, events, industry changes, or policy shifts are ignored.
Adding generic trend lists
External research creates noise when trends are not mapped to a particular audience, period, channel, or result.
Presenting speculation as causation
A plausible external explanation should remain labeled as a hypothesis until stronger evidence supports it.
Is it for you?
Best for
Marketers creating campaign, channel, or annual plans from historical performance data.
Not ideal for
Organizations without reliable internal measurements or a clearly defined market and audience.
From the transcript
“Everything I pair with external trends.”
“it's picked out conferences, it's peaked, picked out seasonal factors, it's picked at industry shifts.”
“Changes in healthcare policies stay informed on these.”
From the episode
GROK 3 vs GPT-4: The AI War Just Got Real [First Look]