One Expert Interview, Multiple Articles, No Thin Pages
To turn one expert interview into multiple articles, split by reader task instead of by synonym: each article gets its own job, query, evidence, and artifact from the transcript.
To turn one expert interview into multiple articles, split the transcript by reader task, not by keyword variation. Each article must earn its own URL with four things the others do not have: a distinct job the reader is trying to finish, a distinct primary query, distinct evidence from the transcript, and a distinct artifact. If a candidate article cannot claim all four, it is a section of an existing article, not a new page.
That rule is the whole method. A good one-hour interview usually supports one to four real articles, most often two, plus a handful of sections and quotes; the interview yield worksheet sets that count for your own transcript. The failure mode is treating the transcript as a keyword mine: spinning "what the expert said" into five near-identical pages that answer the same underlying question with reshuffled paragraphs. Those pages compete with each other, read as filler, and none of them accumulates enough evidence to be the page worth citing.
Who this is for
This is for content leads, editors, and founders who invest real effort in expert interviews and want the yield without the slop. The interview was expensive: someone senior gave you an hour, someone transcribed it, someone cleared the quotes. Getting one article out of it feels wasteful; getting five thin ones is worse.
This guide is not about syndicating an interview across formats (video clips, social posts, newsletters) — that is distribution, and duplication rules barely apply. It is about the search-facing decision: once you know how many articles the transcript can carry, which angles become separate indexed pages and how each one stays distinct.
Is splitting one interview into several articles duplicate content?
Not automatically — it becomes the pattern Google's guidance warns about when the resulting pages answer the same underlying question with reshuffled paragraphs. Two Google positions frame the constraint. The helpful content guidance asks whether a page provides substantial additional value compared with other pages in the search results — including your own. And the spam policies define scaled content abuse as producing many pages that add little value for users, regardless of whether a human, an AI, or a transcript-slicing workflow produced them. A batch of interview spin-offs that each restate the expert's core thesis fits that description uncomfortably well.
There is also a quieter, purely mechanical cost: when several of your pages answer the same query, search engines pick a canonical-ish winner among near-duplicates and the rest split or lose visibility. Google's canonicalization documentation is about URL-level duplicates, but the strategic lesson transfers — if two pages are close enough that you would struggle to say which should rank, you should have published one page.
The test that catches this early: write the primary query for each planned article, then ask whether one searcher typing query A would be fully satisfied by article B. If yes, merge them before drafting, not after both underperform.
How do you decide which interview angles deserve their own article?
An angle deserves its own article only when it wins all four columns of the splitting map: a distinct reader task, a distinct primary query, distinct transcript evidence, and a distinct artifact. Fill this map before commissioning anything. One row per candidate article; the columns are the four things each article must own.
| Candidate article | Reader task (unique?) | Primary query (unique?) | Transcript evidence (unique?) | Artifact (unique?) | Verdict |
|---|---|---|---|---|---|
| How-to on the expert's core workflow | Do the workflow | "how to X" | The anchor story, steps 1–6 | Step checklist from the transcript | Article |
| Mistakes piece from the failure story | Avoid known failures | "X mistakes" | The failed-project story, minutes 30–40 | Failure-mode table | Article, if the failure story has enough distinct detail |
| Tool-selection angle | Choose between approaches | "X vs Y" | The tradeoff discussion, with reasons | Decision criteria table | Article, only if the expert compared options concretely |
| "What [expert] thinks about X trends" | None — it is about the expert | none searched | Scattered opinion quotes | None | Section or cut: opinion garnish belongs inside the other articles |
| Beginner definition piece | Understand X | "what is X" | Nothing the transcript uniquely adds | None | Cut or write separately from primary sources; do not spend interview material on it |
The verdict column has three values: article, section, or cut. Be stingy with "article." A section placed in a strong page compounds that page's value; a weak standalone page just adds another URL that must justify itself.
Two structural notes on the map:
- Evidence rows must not overlap. Each article gets its own quotes and examples. The one exception is a single bridging quote used in two pages with different surrounding analysis — more overlap than that and you are writing the same page twice.
- The artifact column is the tiebreaker. When you cannot decide between section and article, ask what reusable thing (checklist, decision table, template) the page would give the reader. No artifact, no article — the same standard this site applies to every guide.
How do you go from transcript to article set?
1. Tag the transcript by reader task, not by topic
Read it once and mark each passage with the job it serves: "helps someone do X," "helps someone avoid Y," "helps someone choose between A and B." Topic tags ("pricing," "onboarding") produce overlapping articles; task tags produce distinct ones.
2. Cluster the tags and count real candidates
Most one-hour transcripts cluster into one to four tasks with substance, most often two, and several orphan passages. Resist inflating orphans into articles.
3. Fill the splitting map and apply the verdicts
Every row must win all four columns to be an article. Downgrade ruthlessly.
4. Assign each article its own primary query and check them against each other
If two queries would show the same intent in search results, merge the rows. If you maintain a topical map, add these pages to it now — the same no-duplicate rules apply as anywhere else on the site.
5. Sequence the drafts and cross-link by task
Publish the strongest evidence page first; it becomes the hub the siblings link to. Cross-link with task-descriptive anchors ("the failure modes and how to catch them") rather than "read part two."
6. Re-run quote permissions per article
Clearance for one article does not automatically cover a second, differently-framed article — especially when a quote lands in a more critical context. Route each new page through the quote permission and fact-checking workflow before publishing.
7. Schedule a 60-day overlap check
If two siblings end up with materially overlapping impressions for the same queries in Search Console, merge them and redirect. Treat that as normal pruning, not failure.
Mistakes and tradeoffs
- Splitting by synonym. "Expert interview tips," "expert interview best practices," and "how to interview experts" are one page. Query phrasing variants belong in secondaryQueries, not in separate URLs.
- Publishing the raw Q&A as a separate article. A lightly-edited transcript page next to your synthesized articles duplicates all of them at once while being the weakest version of each. If you publish a Q&A at all, make it the single canonical piece or keep it off the index.
- Letting every spin-off restate the full context. Each sibling needs a two-sentence orientation and a link to the hub, not a 400-word repeated intro. Repeated boilerplate is how technically-distinct articles converge back into near-duplicates.
- Spending interview evidence on definitional pages. "What is X" pages rarely benefit from your expert's war stories, and they burn unique material you could concentrate in the how-to. Write definitions from primary sources; spend transcript evidence where first-hand experience is the differentiator.
- Ignoring the freshness asymmetry. The workflow article ages slowly; the tools-and-tradeoffs article ages fast. Splitting means each page needs its own review schedule, which is a real ongoing cost of every additional URL.
- Over-merging. The opposite failure exists: a 6,000-word monolith serving four different reader tasks satisfies none of them cleanly. The splitting map cuts both ways — if a section keeps growing its own task, query, evidence, and artifact, promote it.
Claim ledger
| Claim | Source | Status | Review rule |
|---|---|---|---|
| Google's guidance asks whether content provides substantial additional value compared with other pages in search results. | Google Search Central helpful content guidance | verified on 2026-09-13 | Recheck within 90 days or if guidance changes |
| Google's spam policies describe scaled content abuse as many pages that add little value, regardless of how they were created. | Google Search Essentials spam policies | verified on 2026-09-13 | Recheck within 90 days or if guidance changes |
| For duplicate or near-duplicate URLs, Google selects a canonical version; canonicalization signals include redirects and rel="canonical". | Google canonicalization documentation | verified on 2026-09-13 | Recheck within 180 days |
| A one-hour interview typically yields one to four task-distinct articles, most often two. | Our own editorial rule of thumb, dated 2026-09-13 | disclosed | Present as judgment, not measured data |
| The splitting map and its verdict rules are our own artifacts. | Our own, dated 2026-09-13 | disclosed | Present as workflow advice only |
FAQ
Is repurposing an interview into multiple posts duplicate content?
It depends entirely on whether the resulting pages are distinct by task and evidence. Reworded versions of the same underlying answer are the pattern Google's spam and helpful-content guidance warns about; task-distinct pages with separate evidence are just a content cluster.
Should I publish the full Q&A transcript as its own page?
Only if it is the canonical piece and the spin-offs do not exist, or if it is kept out of the index. A public transcript plus synthesized articles built from it means every article competes with a superset of itself.
Do I need new quote permissions for each derived article?
Re-confirm when the framing changes. A quote cleared for a how-to can read very differently inside a mistakes piece. Cheap insurance: send the expert each new article's quote list with one line of context before it ships.
What if two of the resulting articles end up ranking for the same query?
Merge the weaker into the stronger and redirect. Overlapping impressions in Search Console for the same query set is the signal that the splitting map's verdict was wrong; correcting it consolidates the evidence into one page.
Where this fits in the interview pipeline
Start with the expert interview content brief so the transcript is worth splitting, and run the call with the expert interview run sheet so each candidate article has real evidence. Before committing budget to the split, size it with the interview yield estimate. Clear every derived article through the quote permissions, attribution, and fact-checking workflow, and draft each one with the voice-preservation drafting protocol so every derived article keeps its own share of the expert's specifics. When you add the new pages to your site plan, use the topical map method that avoids duplicate articles, and hold each page to the experience-signal standard in what experience means in a blog post when you are not a famous expert.
Sources and last-reviewed notes
Last reviewed: 2026-09-14. Sources checked: Google Search Central guidance on helpful, people-first content, Google Search Essentials spam policies (scaled content abuse), and Google's canonicalization documentation on duplicate URLs. The splitting map, verdict rules, and the one-to-four-articles rule of thumb are our own editorial artifacts; no example in this guide is a real client case and no traffic outcome is claimed.