River
Y CombinatorBacked by Y Combinator
FREE TEMPLATE

Newsletter Editorial Process Template

Three documents and three sheets that fill a newsletter's idea bank from your archive and reporting pipeline, ranked by what readers already opened.

Free download  ·  No account needed

Every guide to newsletter sustainability gives the same advice: keep a running idea bank instead of starting from a blank page each week. None of them says where the bank comes from, so it becomes one more list an editor has to remember to update, and it goes empty in exactly the week that matters. This pack fills the bank from two sweeps instead of a memory: the archive for what a published story left behind, and the pipeline for what is already reported but too small to run alone.

The Halstead Ledger's Friday newsletter needs five items an issue across thirteen issues, sixty-five slots for the quarter. Sweeping a quarter's 168 published stories found 143 leftover angles, and the story pipeline held 24 more items that needed no new reporting. After removing 41 duplicates and stale entries, 126 usable candidates remained against 65 slots, a 1.94x bank depth. Ranked against what readers had actually opened before, the reader-mailbag and development-watch tags, the two best-performing, alone covered 47 of the 65 slots.

Open the pack in River and the agent reads your archive and pipeline before asking for a single new idea, or download the blank sheets and docs and run the sweep yourself. Either way, the Idea Bank holds every candidate with its source and its rank, the Issue Register tracks what filled each slot, and the Production Checklist gates the send date on bank depth instead of on how the week feels.

168 published stories, 24 pipeline items, and a 1.94x bank depth before a single new idea gets asked for

Every count is computed from the two sweeps beside it. Every tag rank comes from the send history in Performance by Issue.

Idea Bank

The Halstead Ledger, an invented paper. "The Friday Brief," 13 issues, 5 items each, 65 slots for the quarter. 10 of 126 usable rows shown.

IDSourceTagThe itemStatus
AR-01ArchiveDevelopment watchParking garage debt: a 2023 report flagged the same drainage failure the January story missedScheduled
AR-03ArchiveExplainerThe sheriff's overtime cap is a budget guideline, not a binding policyScheduled
AR-06ArchiveReader mailbagDo community college credits transfer to the two flagship state universitiesScheduled
AR-09ArchiveExplainerThe landfill permit's comment window reopens if the footprint grows past 10 percentAvailable
AR-10ArchiveReader mailbagA denied charity-care patient's 30-day appeal window is not on the billing statementScheduled
AR-12ArchiveHistorical callbackThis year's appeal volume is the highest since the 2019 reassessment spikeAvailable
AR-14ArchiveReader mailbagThe state's foster placement distance cap is guidance, not an enforceable ruleScheduled
PL-01PipelineDevelopment watchA second landlord cited under January's rental ordinance, confirmed off the city's logScheduled
PL-03PipelineRoundupThe county added a fourth court interpreter this month, up from threeScheduled
PL-05PipelineReader mailbagThe department's public-request form differs from its FOIA form for a car break-in reportAvailable

AR-01 and AR-09 both trace to stories already in the archive; the sweep found them, nobody had to remember them. AR-05, AR-13, AR-04, PL-02, PL-04, PL-06, PL-07 and PL-08 are the rest of the sample the full sheet ships with.

What the sweeps found

Two sweeps, run once against a quarter's worth of published stories and pipeline notes, then joined against send history. None of these four counts comes from brainstorming.

1. The archive sweep

ReadCount
Stories published this quarter168
Stories with at least one leftover angle89 (53.0%)
Distinct leftover angles found143

2. The pipeline sweep

ReadCount
Items tracked across beat notes38
Newsletter-sized, no new reporting needed24

3. The raw pool, after review

StepCount
Leftover angles + pipeline items143 + 24 = 167
Removed: duplicate, too thin, already covered-41
Usable candidates entering the bank126
Slots needed for the quarter65
Bank depth1.94x

4. Ranked by what this audience has actually opened

TagCandidatesOpen rateClick rate
Reader mailbag1963%15%
Development watch2858%12%
Explainer3151%9%
Historical callback1444%7%
Roundup3439%6%

Reader mailbag is both the scarcest tag and the highest-opening one: 19 candidates against 65 slots, at 63 percent versus a 39 percent roundup average. Scheduling it early, rather than saving it, is why reader mailbag plus development watch alone fill 47 of the quarter's 65 slots, 72.3 percent, from the two best-performing tags.

Issue Register

What actually filled each issue's five slots, logged the day the draft is set, not after send. 3 of 13 issues shown.

IssueStatusLeadMailbagSlots filled
Issue 1LockedAR-01 (Development watch)AR-105 of 5
Issue 2LockedAR-03 (Explainer)AR-144 of 5
Issue 3LockedAR-07 (Development watch)AR-065 of 5

The baseline this replaces

Reading, last 52 issuesCount
Issues that dropped a section, reran old ground, or slipped past send time because the bank was empty on Thursday19 of 52 (36.5%)
Reader-question items a typical quarter schedules without the sweepabout 5
Reader-mailbag candidates the sweep actually finds in one quarter19

The gap between 5 and 19 is not a difference in how many good reader questions exist. It is the difference between remembering to look and sweeping systematically for every one that arrived.

What's in the pack

01

Idea Bank, populated by two sweeps

One row per candidate, tagged by source: an archive sweep for what a published story left behind, and a pipeline sweep for what is already reported but too small to stand alone. On the worked quarter the two sweeps produced 126 usable candidates against 65 slots.

02

Performance by Issue

Open and click rate logged per issue and tagged the same way the bank tags candidates, so a tag's rank comes from what this newsletter's own readers have actually opened, not from a guess about what sounds interesting.

03

Format Template

The section order and the word count band for each one, so a reader can skim the same shape every week. Fixes the lead, the short items, the reader mailbag and the sign-off, and nothing else.

04

Editorial Standard

The line an archive-sourced item carries naming the original story, the rule against running the same leftover angle twice, and what happens to a newsletter item when the story behind it is later corrected.

05

Production Checklist, gated on bank depth

A Tuesday check of available candidates against the next two issues' slots, before the send date gets promised. Below 2x triggers a fresh sweep instead of a Thursday night scramble.

06

Issue Register

Which Idea Bank row filled which slot, logged the day the draft is set. This is what lets the next bank-depth check work from real numbers instead of a feeling that the well is running dry.

07

A space rule every prompt reads first

The idea bank is a sweep, not a brainstorm: read the archive and the pipeline before asking anyone to think of anything new, and never let a reused fact skip its disclosure line.

How to use it

  1. 1

    Open it in River, or download it

    Open the pack and let the agent read your archive and pipeline before it asks for a single new idea, or download the blank sheets and docs. The template library holds the rest of the desk's packs.

  2. 2

    Send the archive and the pipeline, not just topics

    Your published stories for the lookback period, whatever tracks beat notes and reporting in progress, and send history with open and click data if this newsletter has run before.

  3. 3

    Sweep before anyone brainstorms

    Every story checked for a cut quote, a smaller fact, or an unanswered reader question. Every pipeline item checked for whether it is confirmed and newsletter-sized rather than a story still being reported.

  4. 4

    Rank the bank, then check its depth before the next send date

    Tag every candidate and rank by what this audience has actually opened. Below 2x the next two issues' slots, sweep again before drafting anything.

Frequently asked questions

Is this template free?

Yes. Download the whole pack as Word documents and CSV sheets, no signup and no credit card. "Edit with AI" is a separate, optional path for anyone who wants the agent to read your archive and pipeline and build the ranked bank. Nothing happens until you send it something to sweep.

How is this different from just keeping a running list of ideas?

A running list only holds what somebody remembered to type into it, which is why it goes empty in exactly the week that matters. This sweeps the published archive and the reporting pipeline systematically, so the bank starts from what the newsroom has already done rather than what an editor recalled at 5pm on a Thursday.

What if we have never sent this newsletter before, so there is no send history to rank against?

The archive and pipeline sweeps still work; a new title just starts every tag at an even rank until a few issues have actually sent. Performance by Issue fills in from the first send onward, and the ranking gets sharper with every issue rather than staying a guess.

Why does the bank need the published archive and not just a topic brainstorm?

Because a leftover angle costs zero additional reporting: it already survived an editor and, on most desks, a fact check. A brainstormed topic still has to be reported and checked before it can run, which is a different amount of work than the newsletter's schedule usually has room for.

Does a fact reused from an old story need to be re-verified?

It needs its disclosure line naming the source story, and if that story is later corrected, the newsletter item drawing on it needs the same correction, independent of whatever ran on the original piece. The correction and accountability pack tracks that chain on the story side; the Editorial Standard doc here tracks it on the newsletter side.

How do you decide what to schedule first when two tags both have candidates available?

By what this audience has actually opened, not by launch order. Open and click rates vary widely by newsletter topic, which is why beehiiv's own 2025 State of Email Newsletters reports a platform average near 38 percent that individual categories run well above or below. Rank your own tags against your own sends, and schedule the scarce, high-performing one first.

Where does this sit in the rest of the desk's work?

It runs on its own weekly rhythm, separate from the editorial calendar that schedules features. A reader's name and location get handled the way the source pack already handles a source's identifying details, and the newsletter's voice can flex further than the desk's style guide without a lower bar on facts.

Fill the bank before you need it

Download the blank Idea Bank, Issue Register and Performance by Issue as Word and CSV files, or open this pack in River and let the agent sweep your own archive and pipeline.

Edit with AI