Why committing to a few territories compounds
Fifteen pages across one territory build authority. The same fifteen spread across fifteen subjects read as fifteen unrelated posts.
The same principle shows up in five different screens in GrowthOS, and it is the one most likely to be ignored: concentration beats coverage.
Fifteen pages, one territory
Fifteen pages across a single territory build a body of work that search engines and answer engines both read as authority.
The same fifteen pages spread across fifteen subjects read as fifteen unrelated posts. Same effort, same word count, a fraction of the result.
This is why Clusters have priority bands. The bands exist to force the choice rather than let it happen by accident, and why the default path into work is via a cluster rather than an individual opportunity. You can accept a standalone opportunity, and sometimes that is right: a competitor launches something and you respond. But the default is via clusters, because that is what accumulates.
Triage is the point, not the queue length
A research run can produce more opportunities than anyone will write in a quarter. The value isn't in the size of the backlog.
It's in the ordering, and in the confidence to dismiss. A backlog nobody prunes stops being a decision aid and becomes a source of guilt.
Relevance outweighs volume deliberately. A high-volume term your buyers do not search is worth less than a modest one they do. The scoring encodes that, which is why the top of the list is not simply the biggest numbers.
Most opportunities are fixes, not new pages
On a site with real history, the fastest gains come from pages that already rank somewhere. They have links, indexing and measurable behaviour.
A rewrite inherits all of it. A new URL starts from nothing and competes from zero.
Cannibalization is the cheapest version of this. Two of your own pages splitting the same intent means neither wins. Consolidating them usually beats writing a third.
This is why the Rewrite path exists in In Production: to make the better bet the easy choice rather than the disciplined one.
Fewer axes, used consistently
The same logic governs the things you define rather than the things you write.
- Personas. Each one divides attention across the portfolio. A persona earns its place when it would lead a writer to a genuinely different page, not just a different job title in the intro.
- Taxonomy categories. Every category multiplies the ways the portfolio can be sliced, and an axis nobody groups by is overhead on every classification run.
In both cases the instinct to add one more is the instinct to avoid a decision.
Where this shows up
- Clusters is where you commit to a territory.
- Opportunities is where you triage.
- Opportunity types is where fix-versus-new gets decided.
- Personas and Taxonomy are where the same restraint applies to your axes.
Last updated at October 20, 2018
Reading a portfolio
Distribution is the honest read of a site. Averages hide the thing you needed to act on, and grouping is what turns a list into a decision.
Context is the shared source of truth
Every agent in GrowthOS starts from Context. A vague Company Overview produces vague opportunities, vague briefs and generic drafts.