Illustrative case study

A fictional business. The figures are a modelled scenario, not a client result.

Prairie Static Radio · Listener-funded public radio · Minneapolis

How a community radio station turned 30 years of archives into a donor pipeline

889092949698100102104106108SIGN-ONCHALLENGEDIAGNOSISSTRATEGYTIMELINERESULTSSIGN-OFFPRAIRIE STATICFM · LISTENER FUNDED · EST. 1994

Headline signal

+2,800

organic member sign-ups over 14 months

Plan: Content Engine · Content & AI search

91.1 FMThe challenge

Thirty years on tape, and nobody could hear it.

Decades of local journalism sat in an unindexed audio archive, and national outlets were the cited sources even for Minnesota stories.

The station had covered Minneapolis since the mid-nineties: council fights, neighbourhood rebuilds, the jazz clubs, the winters. That reporting sat in an audio archive no search engine could read. When people searched for Minnesota history, they found national rewrites of stories Prairie Static had broken, and the AI answers above them credited those rewrites.

Membership, which pays for everything, depended on on-air pledge drives. Search brought almost no one to the donor page. The brief: make the archive findable, make the station the cited source, and give every archive reader a reason to join.

hours of broadcast audio
31,000hours of broadcast audio
pages indexed at month 0
212pages indexed at month 0
local queries citing national outlets
61 / 80local queries citing national outlets
organic sessions reaching pledge
0.2%organic sessions reaching pledge

94.3 FMDiagnosis

The audit: six sources of interference.

  1. 01 · Crawl−18 dB

    31,000 broadcast hours, 212 pages

    Thirty years of segments lived as MP3 files behind a search form. No URLs per story, no transcripts, nothing a crawler could follow.

  2. 02 · Schema−14 dB

    No NewsArticle or PodcastEpisode markup

    Daily news posts shipped without datePublished, author or location, so Google could not tell a 1998 council report from a press release.

  3. 03 · Entity−12 dB

    Minnesota stories credited to national desks

    On 61 of 80 sampled local queries, AI Overviews cited national outlets that had rewritten the station's own reporting.

  4. 04 · Content−9 dB

    No hub pages for the beats it owns

    Neighbourhood, music and civic coverage was scattered across tag pages with 40-character descriptions and no internal links.

  5. 05 · Conversion−7 dB

    Pledge page reachable only from the homepage

    Organic visitors landed deep in story pages and never saw a membership ask; 0.2% of organic sessions reached the pledge form.

  6. 06 · Technical−5 dB

    Duplicate audio players, 6.4 s LCP on mobile

    Three embedded players per page and an uncompressed station logo pushed story pages past every Core Web Vitals threshold.

97.9 FMStrategy

Four channels, one clear signal.

Built on our Content Engine plan (content & ai search): turn the archive into structured, citable journalism, organise it around the beats the station owns, and put a membership ask where organic readers actually land.

CH 1

News article schema

Every story, new and archived, became a NewsArticle with a real dateline, reporter, location and an AudioObject for the original broadcast.

  • dateline + locationCreated on 4,100 pages
  • speakable sections for headlines
  • correction and update history exposed
CH 2

Topic hubs

Twenty-two hubs for the beats the station has covered longer than anyone: neighbourhoods, the Twin Cities music scene, civic life, weather history.

  • 22 hubs, each with an editor-written primer
  • chronological ‘on this beat’ timelines
  • every archive story linked from one hub
CH 3

AI Overviews

Answer-first summaries, clear sourcing and first-hand audio made the station the citable original rather than the story national outlets paraphrased.

  • 40-word answer block above each story
  • quoted primary audio with timestamps
  • author pages with beat history
CH 4

Evergreen archives

Machine transcripts, edited by volunteers, turned decades of tape into readable pages, each ending with a membership ask tied to the story.

  • 4,100 transcribed segments published
  • ‘keep this archive on air’ pledge module
  • anniversary calendar to resurface stories

Before · month 0

1991 halloween blizzard coverage

AI Overview · sources

  • National Daily: ‘The storm that buried the Midwest’
  • Wire service recap
  • Encyclopedia entry

After · month 14

1991 halloween blizzard coverage

AI Overview · sources

  • Prairie Static Radio: original 1991 broadcast + transcript
  • National Daily recap
  • State climatology office

101.3 FMTimeline

The log sheet: 14 months on air.

  1. Month01

    Archive audit: 31,000 hours catalogued, 9,600 segments shortlisted for transcription.

    180 indexed · 0 citations · +20 members

  2. Month02

    Story URL scheme agreed; NewsArticle template built; first 240 segments transcribed.

    420 indexed · 0 citations · +45 members

  3. Month03

    Volunteer transcript-editing desk launched; 400 pages a month leave the queue.

    820 indexed · 1 citations · +70 members

  4. Month04

    First six topic hubs live: Lake Street, North Minneapolis, Twin Cities jazz.

    1,300 indexed · 2 citations · +110 members

  5. Month05

    Audio players consolidated; mobile LCP down from 6.4 s to 2.1 s.

    1,800 indexed · 4 citations · +150 members

  6. Month06

    Pledge module added to every archive story; membership attribution set up.

    2,300 indexed · 7 citations · +190 members

  7. Month07

    Hubs reach 14; weather-history hub picks up the first winter-storm searches.

    2,750 indexed · 10 citations · +230 members

  8. Month08

    Answer-first summaries rolled across the 500 most-searched stories.

    3,100 indexed · 14 citations · +250 members

  9. Month09

    First AI Overview citations on blizzard and election-night queries.

    3,400 indexed · 18 citations · +260 members

  10. Month10

    Author and beat pages published for 38 current and former reporters.

    3,650 indexed · 22 citations · +270 members

  11. Month11

    Anniversary calendar resurfaces 120 stories; peak month for new members.

    3,820 indexed · 26 citations · +280 members

  12. Month12

    All 22 hubs live; internal links audited across the full archive.

    3,950 indexed · 30 citations · +300 members

  13. Month13

    Archive search rebuilt on indexed story pages, retiring the old form.

    4,040 indexed · 33 citations · +310 members

  14. Month14

    Wrap-up: 4,100 indexed pages, 36 AI answer citations, handover to the newsroom.

    4,100 indexed · 36 citations · +315 members

104.7 FMResults

Loud and clear: the results.

+198%Organic sessionsscale 0 → +250%
+2,800Members from organicscale 0 → 3,000
4,100Archive pages indexedscale 0 → 4,500
36 queriesAI answer citationsscale 0 → 40

Modelled scenario for a fictional business — not a client result. Rankings and traffic are never guaranteed; see our Terms.

Signal over 14 monthsSessions indexNew members (2,800)
0100200300M2M4M6M8M10M12M14

Archive & membership queries

Google position for archive and membership queries, month 14
QueryPosition
1991 halloween blizzard radio coverage#1
minneapolis community radio archive#1
minnesota election night 1998 coverage#1
north minneapolis oral history#2
twin cities jazz history#3
lake street rebuild reporting#4
donate to local radio minneapolis#3
minneapolis city council history#2

107.1 FM · Sign-off

Get a plan like this

Archives, newsrooms and member-funded organisations: we will audit what you already own and tune it for search and AI answers.

← Back to all case studies