Every guide answering "how often should agencies send a newsletter" hands you the same number. Weekly. Occasionally bi-weekly. Monthly if you are stuck. The number is the wrong answer to the wrong question.
An agency's list is not one audience. It is two lists on the same sheet. The pipeline: prospects at three levels of warmth, most of them six to eighteen months away from an RFP. And the book: current retainer clients who see your face on a Zoom call every Wednesday and get pinged in Slack twice a day. The same weekly send to both teaches one of them to unsubscribe and the other one to skim.
What the SERP says, and why it under-specifies
Consensus across the top ten results is the same: weekly or bi-weekly, monthly as the floor, quarterly is death. Campaign Monitor, beehiiv, GetANewsletter, and Madison Miles Media all converge. The advice is not wrong. It is written for a company selling widgets to strangers, where every subscriber is a stranger and no subscriber has your calendar link.
An agency does not have that list. A large chunk of your subscribers are people who already spend $10K to $50K a month with you. Another chunk is the buying committee that might. Consensus cadence, weekly, is the right answer for exactly one segment of the pipeline. It is loud noise to the retainer book and quiet death to the alumni. Sending the same edition to all three trains one of them to hit archive on sight.
Why one cadence breaks for an agency specifically
An agency is already inside its clients' inbox every day. Status updates. Weekly reports. Slack pings. The strategy deck. A separate weekly newsletter from the same firm they already have four standing meetings with is not new information. It is a fifth touchpoint on a subscriber who did not ask for the extra channel.
MailerLite's 2025 benchmarks report ranks marketing and advertising as one of the lowest-performing industries by click rate, well below the 2.09% platform average, per MailerLite's 2025 industry benchmarks. That is not because agency writing is worse. It is because agency lists are stacked with two populations who need very different pacing, and most agencies pick one cadence for both.
The pipeline sits at the other extreme. The B2B buying committee for a services engagement now runs six to ten stakeholders, climbing to 11.2 for deals above $50K in ACV, per Databox's 2026 cycle-length research. Those eleven people are not on your list yet. One of them is. Your job is to keep sending that one subscriber something worth forwarding, on a rhythm the committee can absorb without pattern-matching to spam.
The matrix: three tags in the pipeline, one on the book
Same three tags as the agencies welcome sequence, plus the retainer clients who need their own frequency.
- Curious (pipeline, cold). Read a case study, subscribed to hear more thinking. Six to eighteen months out from any procurement conversation. This is 60 to 75% of an agency list. Bi-weekly. Weekly reads as pushy for people who did not ask you to sell to them yet. Bi-weekly gives you 26 sends a year to stay present without pattern-matching as "agency looking for work."
- Evaluation (pipeline, warm). Has a defined problem, and someone in the room has been told to explore agencies. Three to nine months from a signed contract. On a shortlist of two to four firms. Weekly, because they are actively comparing you to competitors and want proof you keep showing up.
- RFP-active (pipeline, hot). Scope of work is being drafted. Three to six weeks from a decision. Weekly, and every send has a specific link to a piece of thinking the committee should read before the next stakeholder meeting.
- Retainer clients (the book). Monthly at most. Every send is a specific thing they might want next: a new engagement type, a workshop invitation, an alumni-network invite for their team. Not a general update. They are watching you deliver work in real time. The newsletter is not for delivery; it is for expansion or referral.
You do not have to write four newsletters. You write one and let the tag decide who gets it this week.
The math on why the matrix beats the number
An agency running a single weekly send to a mixed list of 2,000 subscribers looks like it is doing marketing. Do the math per segment and it is doing something else.
Say 1,300 curious, 300 evaluation, 150 RFP-active, 250 retainer clients on the same sheet.
Send weekly to all four:
- 1,300 curious × 15% opens × 52 sends ≈ 10,140 opens per year, and a rising unsubscribe rate because most of them wanted the case study, not fifty-two updates.
- 300 evaluation × 40% opens × 52 sends ≈ 6,240 opens per year, the segment doing most of the pipeline work.
- 150 RFP-active × 55% opens × 52 sends ≈ 4,290 opens per year, most of them the same three inbox neighbors before a decision meeting.
- 250 retainer clients × 20% opens × 52 sends ≈ 2,600 opens per year, plus a slow drift of client contacts silently marking your firm as noise inside their inbox.
Send by the matrix (bi-weekly curious, weekly evaluation and RFP, monthly retainer):
- 1,300 curious × 20% opens × 26 sends ≈ 6,760 opens per year, but a healthier unsubscribe rate and a warmer pool moving to evaluation next quarter.
- 300 evaluation × 40% opens × 52 sends ≈ 6,240 opens per year (unchanged, this is who weekly is for).
- 150 RFP-active × 55% opens × 52 sends ≈ 4,290 opens per year (unchanged).
- 250 retainer clients × 30% opens × 12 sends ≈ 900 opens per year, with each open reading a specific expansion offer instead of an industry roundup.
Total opens drop from 23,270 to 18,190. On the number, it looks worse. On the outcomes an agency actually cares about, retainer expansion and pipeline health, it is the trade you want. You bought back client goodwill, protected sender reputation on your best segment, and stopped burning your cold pipeline with sends they were not ready for.
Replies, not opens, are the point for an agency. A weekly send to a mixed list produces a reply rate the tool reports as one average that hides four different signals. Segmenting the cadence improves the ratio of "sent to somebody ready to reply" without writing more. The evaluation segment carries the pipeline. The retainer segment carries the expansion. The curious segment carries next quarter's pipeline. Weekly to all three treats those three jobs as one job.
The tooling implication: the tag is the send
The matrix requires two things most stacks handle poorly.
A tag written the moment the subscriber comes in. The case-study download writes "curious." The RFP-response request writes "evaluation." The retainer-onboarding import comes in tagged "client." No manual sort. No nightly Zapier reconciliation. The tag is the cadence lever, and if it arrives late, the automation the tag was supposed to trigger has already missed the window.
One list where every send filters on tag. Not four lists on four calendars. One list, one send calendar, four filters running on the same subscriber column. The Wednesday send to evaluation goes out without touching the retainer book. The monthly retainer send fires from the same editor, filtered on the client tag.
This is why Nashra tags at the page. The Magic Link that hosts the case study writes the tag on submit. The subscriber list holds the tag as a column. The automation that runs the welcome sequence keys on the tag from the first email onward. Four cadences, one send calendar, no stitching. For the deeper argument on why the tag beats the send button as the unit of planning, see three tags, two jobs.
The answer, in one line
For most agencies, the honest cadence is: bi-weekly to curious prospects, weekly to evaluation and RFP-active, monthly to current retainer clients, quarterly at most to alumni. One send calendar. Four cadences. The list decides which subscribers hear which edition.
The number every SERP result gives you, weekly, is not wrong. It is the right answer for one of the four segments on your list. Three of the other four are where an agency's next retainer, next expansion, and next referral live. Cadence per tag is how you serve all four without writing four newsletters.
The consultants version and the coaches version of this piece run the same matrix on different segments. The agencies version differs on one point: the retainer book. It is the only segment where too much of you is a real risk, and the only one where the newsletter is competing with your own Slack messages for attention. For the deeper reasoning on why an owned subscriber list is worth the segmentation work in the first place, see the cornerstone piece on subscribers vs followers. Same argument, one level up: the tag on the subscriber decides what you can do with them, and the follower graph never gave you the tag.
