Your archive is where your traffic actually lives#
Most blogs get the overwhelming majority of their traffic from posts published months or years ago, not from whatever went up last week.
The numbers on this are consistent. One HubSpot partner agency reported that 85 percent of their monthly blog traffic came from old posts rather than anything published in the past month. Databox found that three individual posts accounted for 50 percent of their organic traffic in a single month.
Yet almost all content effort goes into producing the next thing. HubSpot's own historical optimization work, where they assigned people specifically to improve existing posts rather than writing new ones, more than doubled monthly leads from those posts and lifted their organic search views by an average of 106 percent. They did not write anything new to get that.
The relevance here is simple. If your archive is where your traffic lives, your archive is also where your repurposing opportunity lives. You are not looking at a blank page. You are looking at forty posts that already proved they were worth reading.
Why this matters more in 2026 than it did in 2023#
Here is the part that changes the calculation, and it is worth being blunt about.
The model of publishing informational content and capturing organic clicks is under real structural pressure. Zero-click searches, where the user gets their answer without visiting any website, are now widely estimated to exceed 65 percent of all queries. When Google displays an AI-generated answer above the results, industry estimates put the drop in organic click-through rates somewhere in the range of 30 to 60 percent. These figures move depending on who is measuring and what vertical they are looking at, so treat them as direction rather than precision. The direction is not in dispute.
What that means practically: the traffic your archive currently earns is not guaranteed to persist, and the mechanism eroding it is outside your control. You cannot out-optimize an AI Overview that answers the question on the results page.
Audio is a channel that mechanism cannot intercept. Nobody's podcast queue gets summarized above the fold. There is no AI Overview sitting between a listener and an episode in their app. When someone subscribes to a feed, the relationship is direct, and it does not route through a search result that may or may not send them onward.
That is not an argument that audio replaces search traffic. It is an argument that a channel you own outright is worth more now than it was three years ago, and your archive is the cheapest possible way to build one, because the content already exists.
Which posts to convert#
Do not convert everything. Most archives contain a lot that should stay written, and picking badly is how people conclude this does not work.
Convert these:
- Posts that already perform. Your top ten by traffic or engagement have already been validated by an audience. That is better evidence than your own opinion about what deserves a second format.
- Evergreen explainers. Anything that will still be true in a year. A repurposed piece published three months from now still works, which is the entire economic argument for doing this.
- Anything argumentative or conceptual. Posts where you take a position, explain a mechanism, or walk through reasoning. These translate to audio naturally because they were already closer to how someone would explain the idea out loud.
- Long posts with high traffic but poor completion. If your analytics show people arriving and leaving before the end, that is a format mismatch. Those readers were interested enough to click and did not have the time or attention to read 2,500 words. Audio addresses exactly that gap, which we cover in more depth in the unread problem.
Leave these written:
- Reference material. Anything people look things up in. Nobody wants to scrub through audio to find a threshold or a setting.
- Anything heavily visual. Screenshot walkthroughs, data-driven pieces where the chart is the argument, design breakdowns.
- Step-by-step tutorials people follow along with. They need to pause, scroll back, and check they are on step four. Audio is hostile to that.
- Posts that are already thin. Converting a 600-word post produces a four-minute episode with not much in it. Better to combine three related short posts into one substantial episode.
- Anything time-sensitive. News commentary and dated announcements will not repay the effort.
The practical rule: convert posts where the value is in the reasoning; keep posts where the value is in the reference.If your blog runs on Substack, hosting and distribution are handled natively, which changes the workflow: see Substack to podcast.
The workflow#
This is a quick win, not a project, provided you do it in the right order.
1. Pull your top posts by traffic. Twenty minutes in analytics. Sort by organic sessions over the last twelve months, then cross-reference against the criteria above. You are looking for evergreen, conceptual, high-traffic posts. Most people find between five and fifteen genuine candidates.
2. Batch, do not drip. Whatever method you use, doing five at once costs less than five times meaningfully doing one, because the setup and the mental context are paid once. Block a morning rather than adding a step to your weekly publishing routine, which is how this ends up abandoned.
3. Convert with the format decision made deliberately. Narration reads your text aloud. Restructuring rewrites it for the ear. For a 2,500-word analytical post, narration will lose listeners around minute three, for reasons covered below. For a short conversational post, narration is fine.
4. Review before publishing. Read the script or listen through once. Blog posts contain things that break in audio: URLs, bracketed asides, "as the chart above shows," "click here," bulleted lists that were visual. Catching those takes minutes and is the difference between something publishable and something obviously auto-generated.
5. Put it in two places. This is the step people miss, and it doubles the return.
Two places, two different jobs#
An audio version of a blog post can do two entirely separate things, and most people only do one.
Embedded on the post page. A player at the top of the article gives every visitor a choice before they start scrolling. This serves the reader who arrived from search, is interested, and does not have ten minutes to read right now. It also serves accessibility, giving people who prefer or need audio a way to consume the piece. This costs you nothing beyond hosting the file.
Published as a podcast feed. This is the part that builds the channel described earlier. Your back catalog becomes an actual show with an RSS feed submitted to Apple Podcasts and Spotify, reaching people in the app where they already listen rather than only people who found your website.
The second one is where the strategic value sits, because it creates a relationship that does not depend on search. Ten converted archive posts is a ten-episode back catalog, which is enough to launch a show without writing a word. If you are setting that up for the first time, the mechanics of hosting and distribution are covered in our guide on how to turn your newsletter into a podcast, and the same three problems apply: making the audio, hosting it with an RSS feed, and distributing that feed.
Why straight narration disappoints#
Worth understanding before you pick a method, because this is the usual reason people try this once and stop.
Blog posts are written for the eyes. They use subheadings, bullet lists, bold text, and links to signal structure, and readers navigate that visually, skimming ahead and glancing back. None of that survives being read aloud. The listener cannot see where they are, cannot skim, and cannot flip back when the thread slips.
There is also a memory constraint. Spoken information is transient: once a sentence has passed, it is gone unless the listener holds it in working memory. Sentences with subordinate clauses and parenthetical asides, which read elegantly, are difficult to follow when they arrive once, at someone else's pace, while the listener is driving.
This is why a narrated post frequently loses people around minute three, even when the writing was strong. The content was fine. The conversion did not happen. Something was read aloud, but nothing was rewritten for listening.
Does this hurt your SEO?#
No, and this comes up often enough to answer directly.
Audio and text are different media, not competing pages, so there is no duplicate content issue in publishing both. Embedding an audio version on the same URL as the written post is simply an additional asset on that page, the same as an image or a video.
The one thing worth getting right: if your podcast host generates a public web page for each episode, those pages can end up competing with your original posts in search. Point the canonical tag on the episode page back to your blog post, or keep episode pages minimal with the full text living only on your site. Your blog post remains the canonical version and the episode links back to it in the show notes.
The broader question of duplication across repurposed formats is covered in our guide to content repurposing.
Where Sprep fits#
We build a document-to-podcast tool, so treat this as an interested party talking.
Sprep converts blog posts you have already published into two-host conversational episodes rather than narrating them. One voice asks the questions a reader would have asked and the other delivers the argument, which is what addresses the transience problem above, since questions create natural checkpoints for a listener whose attention drifted.
For back-catalog work specifically, the relevant properties are that it works from the existing text rather than requiring a recording session, and that every word of the script is editable before audio renders. That editing step is where you strip the things that break in audio: URLs, "as shown above," references to visuals the listener cannot see. It is also what makes the episode publishable under your name rather than something that approximately resembles your writing. An approved script can be generated in over 70 languages if your audience is international.
It produces the audio file. Hosting and distribution are still yours to arrange, which is why the two-places section above matters.
Where it is the wrong tool. If your blog's appeal is your voice and personality, record yourself, and our guide to starting a podcast with no equipment covers doing that with a phone. If your posts are reference material or heavily visual, they should stay written. And if a post was unclear in writing, conversion produces a clear recording of unclear thinking, because the format change is faithful to what you give it.
FAQ#
How do you repurpose blog posts into audio? Select evergreen, high-traffic, conceptual posts from your archive, convert them either by recording yourself or using a conversion tool, review the script to remove things that break in audio such as URLs and references to visuals, then publish in two places: embedded on the original post page and as an episode in a podcast feed.
Which blog posts should you convert to audio? Posts that already perform well are evergreen rather than time-sensitive, and where the value is in reasoning or explanation rather than reference. Leave reference material, visual walkthroughs, and step-by-step tutorials written, since listeners cannot easily scrub back to find a specific detail.
Does converting blog posts to audio hurt SEO? No. Audio and text are different media rather than competing pages, so there is no duplicate content problem. The one thing to manage is podcast host episode pages, which can compete with your originals. Set canonical tags pointing to your blog post, or keep episode pages minimal.
Why do most of a blog's visitors come from old posts? Because search traffic accumulates over time while individual posts age. One agency reported 85 percent of monthly blog traffic coming from posts older than a month, and it is common for a small number of posts to account for a large share of total organic traffic. This is why optimizing and repurposing the archive often outperforms writing more.
Is it worth converting old blog posts given AI search is reducing clicks? That is arguably the strongest reason to do it. Zero-click searches are widely estimated to exceed 65 percent of queries, and AI-generated answers reduce organic click-through substantially. A podcast feed is a channel that is not intercepted by search results, so converting the archive builds a direct relationship that does not depend on a click-through.
Should the audio be narration or a restructured conversation? Depends on the post. Short, conversational writing narrates acceptably. Long analytical posts usually lose listeners when read aloud, because written structure such as subheadings and bullet lists does not survive narration and spoken content cannot be re-read the way text can.
Where should you publish the audio version of a blog post? Both on the post page and as a podcast episode. The embedded player serves visitors who arrive from search but do not have time to read. The podcast feed reaches people in Apple Podcasts and Spotify, which is the part that builds an audience independent of search traffic.
How many blog posts do you need to start a podcast? Ten converted archive posts gives you a ten-episode back catalog, which is enough to launch a show without writing anything new. A show with an existing catalog also signals to new listeners that it is worth subscribing, since there is more to listen to immediately.Once that catalog exists, growing it is a separate job, covered in how to grow a podcast audience without ads.
How long does it take to convert a blog post to audio? The conversion itself is fast with a tool, and the meaningful time goes into selection and review. Batching five posts in one session is considerably more efficient than doing one per week, because setup and context are paid once rather than five times.
What should you remove from a blog post before converting it? Anything that only works visually: URLs read aloud as strings, "as the chart above shows", "click here", bracketed asides, and bullet lists that depended on being seen rather than heard. A review pass catching these takes minutes and is what separates a publishable episode from an obviously automated one.
Start with your best-performing post#
The free plan converts one post you have already published into a full episode of up to 15 minutes, with the script editor included so you can strip the URLs and visual references before any audio is generated.
See it in action