Developing
The scenario is illustrative and gives no numbers; the author recommends holding a proposed increase and testing.
In 60 seconds
- The example: sales are up, first-time buyers increased, search and retail media beat targets, media costs rose and margin "barely moved".
- The author says to break sales growth into new, occasional and regular buyers before approving more media spend.
- Repeat purchases do not always signal loyalty, because habit and convenience can look similar.
- Recommended: hold the proposed increase, keep prospecting at current levels, test a smaller discount separately and run a geo holdout with success thresholds agreed in advance.
Impact Layer
- What changed
- A practitioner shows how to reconcile marketing and finance views of growth with simple evidence.
- Why it matters
- Disagreement between CMOs and CFOs often comes from different definitions of growth, and a shared test can settle it.
- Who is affected
- CMOs, CFOs and revenue operations leaders.
- Martech stack area
- CRM, Analytics
- Stack impact
- Agree test thresholds with finance before spend increases, and report growth by buyer type, not only total sales.
- Business impact
- Low A contributed opinion column on MarTech (owned by Semrush). It cites no data sources, and the scenario is hypothetical; the author notes tests carry their own costs.
- Related technologies
- What to watch
- Your own results from a holdout test.
Angelina Eng, founder and principal advisor at Enso Horizon, argues in a MarTech column published October 21, 2026 that a CMO can see growth where a CFO sees a problem.
The hypothetical review
Sales are up, first-time buyers increased, search and retail media beat targets, media costs rose and margin “barely moved”. The example gives only directional results and no numbers.
The advice
- Break sales growth into new, occasional and regular buyers before approving more media spend.
- Repeat purchases do not always signal loyalty; habit and convenience can look similar.
- Reach and share of voice show brand presence but cannot prove an investment will pay off.
- Privacy rules and separate platform data make channel reports incomplete for cross-channel journeys.
- Hold the proposed increase, keep prospecting at current levels, test a smaller discount separately and run a geo holdout with success thresholds agreed in advance.
Caveats
The scenario is illustrative, not empirical. Customer lifetime value forecasts are only as reliable as the outcomes of earlier, similar cohorts, and tests carry their own costs. It is contributor opinion.
Our read
The structure is useful: agree on what evidence would change the decision before you look at the results.
Sources
- MarTech: When the CMO sees growth and the CFO sees a problem (opens in a new tab)
Source for the column's argument and caveats.