This one is for the content strategists, the SEO leads, and anyone who has published diligently for three years and ended up with a library so large, so tangled, and so quietly out of date that nobody can say which half of it is still working.
What: Using Breeze Assistant to run a decay audit across your content library and hand every page one of four decisions: keep it, refresh it, consolidate it, or retire it, judged not on vanity traffic but on the leads and pipeline each page actually contributes. The point is to stop pouring effort into new posts while the ones you already own quietly rot, and to fix the library you have before you add another word to it.
Prompt of the week:
Every content library tells the same story if you leave it long enough. You publish diligently, month after month, year after year, and the library grows, but it does not grow tidily. It grows into a thicket. Old posts fall out of date and start contradicting your current positioning. Three different pages half-answer the same question and quietly compete with each other for the same search. Conversion paths wander off in different directions. And your reporting shows plenty of traffic while telling you almost nothing about which of those hundreds of pages is genuinely earning its place.
The reflex, faced with flat results, is to publish more, which in 2026 is very often the wrong move. The clearest message running through content strategy this year is that publishing net-new for the sake of volume has stopped working, and the highest-leverage growth is hiding in the content you already own. HubSpot has long reported that around three-quarters of blog traffic comes from older posts, which means the pages you wrote a year or two ago, not the ones you publish next week, are carrying your results, and most of them are decaying quietly while nobody is watching.
Content decay is slow and cumulative, and it rarely announces itself. A page slips from the top of the results to the middle. A competitor starts answering the question better. Search intent drifts towards a comparison or a checklist that your old explainer does not provide. The traffic bleeds away a little each month until, one day, the page ranks for nothing at all. The fix is not to rewrite everything, which is simply a different kind of waste. It is to look at each page honestly and sort it into one of four fates: keep it as it is, refresh it, fold it into a stronger page, or retire it.
Here is the part most content audits get wrong: they sort by traffic. Traffic is the vanity metric that flatters the wrong pages, because a high-traffic post that converts nobody is worth less than a quiet one that steadily feeds your pipeline. The audit that actually matters ranks pages by what they contribute in leads and pipeline, not by pageviews, and that connection between a page and the revenue it influences is precisely the thing HubSpot can see that a standalone SEO tool cannot. This prompt builds that audit: a revenue-aware decision for every page, scoped tightly enough that it does not collapse into a spreadsheet nobody ever opens again.
Prompt structure
Paste this into Breeze Assistant and make sure CRM data access is enabled in your AI settings so Breeze can reference your pages, traffic trends, engagement, conversions, and the contacts and deals each page has influenced:
Role: You are a content strategist who specialises in content audits
and decay. You know that in 2026 publishing more is rarely the
answer, that most traffic comes from older pages, and that the
highest-value work is usually fixing what already exists. You judge a
page by the leads and pipeline it contributes, not by vanity traffic,
and you sort ruthlessly rather than rewriting everything in sight
Task: Run a decay audit across the section of content I name below
and give every page in scope one of four decisions: keep, refresh,
consolidate, or retire. Rank the work by business value and
recoverability, flag overlapping pages that compete with each other,
and specify the redirect discipline for anything we consolidate or
retire. The goal is a prioritised, revenue-aware action plan, not a
rewrite of the whole library.
Context:
- Company: [COMPANY NAME]
- Industry: [INDUSTRY]
- HubSpot tier: [Content Hub / Marketing Hub edition, since traffic
analytics and the topic-cluster tools depend on it]
- The section to audit FIRST: [blog / resource centre / product
pages / one specific topic cluster]. Do not audit the whole site
at once
- Rough size of that section: [NUMBER of pages]
- What a conversion looks like for us: [demo request / form fill /
trial / contact], so pages are judged on leads, not just traffic
- Known problem areas: [outdated posts / repetitive topics / thin
pages / pages that rank but never convert]
- Business priorities right now: [the topics or products that
matter most]
Run the following:
1. PAGE-BY-PAGE TRIAGE
For every page in the named section:
- Pull the last 12 to 24 months of signal: traffic trend (rising,
stable, or decaying), engagement, and, crucially, the leads and
pipeline the page has influenced
- Sort each page into one of four buckets: KEEP (performing and
current), REFRESH (decaying but recoverable and worth it),
CONSOLIDATE (overlaps a stronger page), RETIRE (no traffic, no
leads, no strategic value)
- The bucket decides the action. Be explicit about which, and why
2. THE REFRESH SHORTLIST
Of the REFRESH pages, rank them by business value and
recoverability, NOT by age:
- A page that has slipped from the top of the results on a
valuable query is a better bet than one that never performed
- For each, name the most likely cause of decay: outdated
information, intent drift (the query now wants a comparison or a
checklist), weak structure, or lost internal links
- Say what specifically to change, not just "refresh it"
3. CONSOLIDATION & CANNIBALISATION
Find the pages quietly competing with each other:
- Identify pages that overlap and chase the same search, splitting
authority between them
- For each overlap, recommend which page becomes the canonical
survivor and which folds into it
- Distinguish a true merge (combine the content and redirect) from
a canonical signal (keep both live, point search at one)
4. RETIREMENT & REDIRECTS
Decide what to remove, and do it safely:
- List the pages to retire: no traffic, no leads, no strategic
role, or actively contradicting current positioning
- For every consolidation and every retirement, specify the
redirect: a 301 to the most relevant surviving page, never a
bare delete that leaks the accumulated authority and breaks
inbound links
- Flag any page carrying valuable backlinks that must be
preserved through the redirect
5. INTERNAL LINKING & CLUSTERS
- For the pages worth keeping and refreshing, check they are
supported by internal links from related content, since
orphaned pages decay fastest
- Recommend where pillar pages should gain links from their
cluster, so authority flows towards the priorities
6. REFRESH CADENCE
- A refresh cycle for the valuable pages (pillar pages roughly
every 12 to 18 months) so the library does not silently decay
back to this state
- How steady refresh work should sit alongside new production,
rather than the team only ever publishing new
Constraints:
- Judge every page by leads and pipeline influence, not vanity
traffic. A high-traffic page that converts nobody is a candidate
for scrutiny, not celebration
- Scope to the one section provided. Do NOT attempt the whole site in
a single pass. A focused audit that gets acted on beats a complete
one that becomes an unused spreadsheet
- Never recommend deleting a page without a 301 redirect to a
relevant survivor. A bare delete leaks accumulated authority and
breaks every inbound link pointing at it
- Do not treat retirement as the default. Prefer refresh or
consolidation for anything with residual value; retire only what
genuinely earns nothing and serves no strategic purpose
- Prioritise by business value and recoverability, not by age. Old is
not the same as dead
- If traffic, conversion, or pipeline data for a page is not visible
from the current context, state:
"SIGNAL MISSING: [what needs checking manually]"
Output format:
### I. LIBRARY HEALTH SUMMARY
{3-sentence overview: the state of this section, the single biggest
opportunity hiding in it, and a headline count of how many pages fall
into keep / refresh / consolidate / retire}
### II. PAGE TRIAGE
| Page | Traffic Trend | Leads / Pipeline | Decision | Why |
### III. REFRESH SHORTLIST (ranked by value and recoverability)
| Rank | Page | Likely Cause of Decay | What to Change |
### IV. CONSOLIDATION MAP
| Overlapping Pages | Canonical Survivor | Merge or Canonical | Redirect |
### V. RETIREMENT LIST
| Page | Reason | 301 Redirect Target | Backlinks to Preserve? |
### VI. REFRESH CADENCE
{The ongoing cycle, and how it sits alongside new production}
Why this prompt works, and how to adapt it
The whole premise here runs against a deep marketing instinct, which is that progress means publishing. It does not, or at least not only. In a mature library, the pages you already own are the asset, and letting them decay while you chase new ones is like buying more rooms for a house whose foundations are quietly cracking. This prompt reframes the work from production to portfolio management: not what should we write next, but what have we already got, what is it worth, and what should we do about each piece of it.
A few things to note about how it is constructed:
It judges pages by pipeline, not by pageviews. The oldest mistake in content auditing is to sort by traffic and protect whatever sits at the top. But a page pulling ten thousand visits that converts no one is worth less than a quiet page that reliably feeds your sales team, and traffic alone cannot tell them apart. The prompt insists on ranking by leads and pipeline influence, which is exactly the page-to-revenue connection HubSpot can make from its own CRM and a standalone SEO crawler simply cannot see.
Refresh beats rewrite beats delete. The four buckets exist to stop two opposite over-reactions: rewriting everything, which burns the team out for marginal gain, and deleting aggressively, which throws away accumulated authority. Most of the value lives in the middle, in refreshing decaying-but-valuable pages, and the prompt is built to find that middle rather than lurching to either extreme.
It hunts cannibalisation. One of the least visible problems in a big library is two or three pages competing for the same query, splitting the authority that should sit behind one. The prompt actively looks for these overlaps and forces a decision: a true merge with a redirect, or a canonical signal that keeps both live but tells search which one counts. Left alone, cannibalisation quietly caps how well any of the competing pages can rank.
Delete is never the default, and never without a redirect. Retiring a page feels clean and decisive, and done carelessly it is one of the most expensive things you can do, because a bare delete leaks every bit of authority the page had earned and breaks every inbound link and bookmark pointing at it. The prompt treats retirement as the last resort and hard-requires a 301 redirect to a relevant survivor for anything removed, so the equity flows somewhere useful instead of into a 404.
It is scoped on purpose. The fastest way to kill a content audit is to try to do the whole site at once, which produces a heroic spreadsheet that everyone admires and nobody ever acts on. The prompt deliberately takes one section at a time, a single blog or cluster, because a small audit that gets executed is worth infinitely more than a complete one that becomes a monument to good intentions.
“SIGNAL MISSING” keeps it grounded. Breeze can read your pages and, through the CRM, the contacts and deals they have touched, but it cannot always see clean historical traffic for an older page, or the backlinks pointing at one from across the web. Rather than guess at a number and let a page be retired on a hunch, the flag marks the gap for a human to fill, because retiring the wrong page on bad data is the one mistake in this whole exercise that is genuinely hard to undo.
Adapting it for your portal:
Drowning in blog posts specifically? If the blog is the mess, point the audit straight at it: “Focus only on the blog. Group the posts by topic first, then find the clusters where we have five thin posts that should be two strong ones, and tell me which to merge into pillar pages.” The output leans into consolidation, which is usually where an overgrown blog leaks the most value.
Planning a migration or replatform? A move is the worst time to carry dead weight, and the best time to prune. Add: “We are about to migrate the site. Audit the content first so we do not carry decayed pages across, and give me the full redirect map for everything we consolidate or retire before the move, not after.” The plan front-loads the redirect discipline so the migration preserves equity instead of shedding it.
On Content Hub with topic clusters set up? If you have the cluster tools, put them to work: “Use our topic clusters. Show me which pillar pages are under-supported by their clusters, which cluster pages are orphaned, and where a content gap is weakening a cluster we care about.” The audit shifts towards cluster structure and internal linking rather than page-by-page triage alone.
Nervous about losing rankings on a merge? If a consolidation feels risky, ask for the cautious version: “Be conservative with consolidation. Where a merge is risky, prefer a canonical signal over combining and redirecting, always preserve backlinks, and tell me the exact redirect and the check I should run afterwards to confirm nothing dropped.” The plan tilts towards reversibility and careful verification.
Thin on analytics history? If the data does not go back far, say so and let the prompt adapt: “We only have a few months of reliable analytics. Tell me what to capture from now on so a proper decay audit is possible in six months, and make the decisions you can with the data we do have, flagging the rest.” You get a usable first pass plus the instrumentation to do it properly next time.
Want a quarterly cadence? Save the output and re-run it 90 days later with: “Compare against the output from [DATE] and report on which refreshed pages recovered, which are still decaying, whether any consolidations worked, and which section we should audit next.” That turns a one-off clear-out into an ongoing maintenance habit.
Beyond the prompt:
The plan the prompt produces is a list of decisions. Turning it into recovered traffic and pipeline is about doing them in the right order and resisting the urge to do all of them at once.
Start with the refresh shortlist, not the full triage. The decaying-but-valuable pages at the top of that ranked list are where the fastest return lives, because they already have authority and history and simply need bringing back to relevance. Work the top few properly before touching anything else; a handful of well-executed refreshes will usually move more than a hundred rushed edits.
Refresh properly, and treat it like a release. A real refresh is not a date change, it is updated information, a structure that matches what the query now wants, and stronger internal links. Update the modified date when the work is substantial, log what you changed, and then watch the page at fourteen, thirty, and sixty days, the same way you would track any product change, so you learn what a refresh is actually worth.
Consolidate carefully, with the redirects mapped first. Merging pages is where equity is either preserved or lost, so decide the canonical survivor, move the worthwhile content into it, and put the 301s in place as part of the same job, never as a tidy-up you will get to later. The gap between merging and redirecting is exactly where rankings fall through the floor.
Retire last, and sparingly. Once the keeps, refreshes and consolidations are done, what genuinely remains with no traffic, no leads and no purpose can go, each with its redirect to the nearest relevant survivor. Retirement is the smallest bucket for a reason: it is the end of the process, not the point of it.
Then hold a refresh cadence so you never have to do this the hard way again. The reason libraries end up as thickets is that refresh is always the thing that waits until next quarter. Put a standing slot in the calendar to revisit the valuable pages on a cycle, pair it with new production rather than letting new production crowd it out, and the decay never gets the chance to compound.
A content library is not a pile you keep adding to. It is an asset you maintain, and the teams that pull ahead in 2026 are not the ones publishing the most, they are the ones whose best pages never quietly died. Fix what you already have, and everything you publish next lands on solid ground instead of on top of a graveyard.
