1. The Twenty-Seven Minute Problem

• 70% of small business owners spend under five hours a week on marketing. That is Fiverr’s May 2025 survey of nearly 6,000 businesses across 25 countries. Their top worries were brand awareness (42%), creating content (40%), and keeping customers (33%). All three are jobs that reward showing up over and over.
• At the same time, the channel count keeps climbing. Taradel’s 2025 survey found 81% of small businesses now run two or more channels, and the share running on a single channel fell from over 24% in 2022 to 11.5%. Most run three to four. One in ten runs six or more.
• Now do the arithmetic. Eleven touchpoints. Five hours. Twenty-seven minutes each per week, which buys you one scheduled post, a glance at a dashboard, and a vague feeling of being behind.
• The menu grows faster than the week. Scott Brinker’s 2025 martech landscape counted 15,384 products in 49 categories, up 9% in a year and roughly 100 times the 150 he mapped in 2011. Your Tuesday is still twenty-four hours long.

2. Almost Everything Worth Doing Compounds

The case for subtraction rests on one property: some mechanisms improve the longer you repeat them, and some reset to zero the second you stop. Eleven channels guarantees that none of them runs long enough to show you which kind it was.

• Content. HubSpot tracked 2,000 compounding and 18,000 decaying posts. Compounding posts were about 10% of output and 38% of all traffic, so one does the work of six. Six months out they pull 2.5 times launch-month traffic; at twenty-two months, 3.4 times. Decaying posts land under one-fifth.
• Email. Litmus puts average return at $36 per $1, rising to $45 in retail and ecommerce and $48 for senders at five to eight emails a month. 41% of marketers name email their most effective channel, against 16% each for social and paid search. Highest-return asset most firms own, and the one they feed the least.
• Referrals. Schmitt, Skiera and Van den Bulte followed roughly 10,000 German bank customers for 33 months. Referred customers carried at least 16% higher lifetime value, were about 18% less likely to leave, and ran roughly 25% more profitable per year. The margin edge faded after about 1,000 days. The loyalty edge never did.
• Share of voice. Binet and Field’s IPA analysis, corroborated by Nielsen across 123 brands, finds every 10 points of excess share of voice buys roughly 0.5 points of market share a year, about 0.7 in B2B. Sustain twenty points for five years and you have moved the company. Sustain it for one quarter and you have moved nothing.

Channel	Improves with repetition?	The number that proves it
Email list	Yes, strongly	$36 per $1 average; $48 at 5 to 8 sends a month (Litmus)
Evergreen content / SEO	Yes, strongly	3.4x launch traffic at 22 months for compounding posts (HubSpot)
Referral program	Yes, in the customer	16% higher LTV, 18% lower churn, gap persists (J. of Marketing, 2011)
Brand / share of voice	Yes, arithmetically	+0.5 pts market share per 10 pts ESOV per year (Binet & Field; Nielsen)
Paid social	Partly, and rented	Reach and cost are set by someone else’s algorithm
One-off events; monthly-posted profiles	No	Interest decays from event day; posting cadence sits below what any algorithm rewards

Table 1. The subtraction test in one grid. Everything in the top four rows punishes interruption. Everything below it survives being dropped.

The chain has to stay unbroken. Orbit Media’s August 2025 survey of 808 marketers: 21% report strong results overall, rising to 37% among those publishing multiple times a week and 39% among those writing 2,000-word pieces. Only 39% publish weekly at all.

3. MTV Launched With 125 Videos and Refused To Add Anything

At 12:01 a.m. on August 1, 1981, MTV went on the air with 125 videos, 13 advertisers, and 2.1 million households, none in New York or Los Angeles. By any 2026 standard: underfunded, single-channel, one format, zero diversification. It became the most profitable idea in cable television.

• The constraint was the product. With 125 clips in twenty-four-hour rotation, “Video Killed the Radio Star” came around often enough that people who were eleven in 1981 can still hum it. Repetition did the work a bigger library would have prevented.
• One message, everywhere, forever. “I Want My MTV” launched March 1, 1982: the same thirty seconds on a loop, with Mick Jagger, Sting, Pete Townshend and Pat Benatar yelling at viewers to go yell at their cable operator. Co-founder Tom Freston said the network was “adding millions of customers a month.”
• It worked because it never varied. Subscribers tripled from 3 million to 9 million during 1982. Revenue went from $27.7 million in 1983 to $42 million in 1984, with $11.9 million in net income and the top ad revenue of any cable programmer that year. By 1992: 112 million homes worldwide, about $400 million.

4. Then MTV Started Adding Channels

MTV launched MTV2 in 1996 specifically to recapture the free-form spirit of early MTV, a remarkable admission that the original had already been diluted. Then came MTV Hits, MTV Jams, mtvU, Tr3s, MTV Live, MTV Classic, MTV Base, Club MTV, MTV Rocks, MTV Dance, MTV 80s, MTV 90s and MTV 00s. Each carved a slice off the audience and off the attention that used to go into one feed.

	1981 to 1992: concentrated	1996 to 2025: dispersed
Channels run	One	A dozen-plus MTV-branded feeds
Format	One, on relentless rotation	Music, reality, retro, genre, regional
Reach	2.1M households to 112M homes	MTV Music + MTV 90s: 2.25M views for all of July 2025
Money	$27.7M to $400M in revenue	Part of a reported $500M to $2B cost-cutting program
Outcome	First profitable cable network	Five music channels went dark Dec 31, 2025

Table 2. On October 12, 2025, Paramount announced MTV Music, MTV 80s, MTV 90s, Club MTV and MTV Live would switch off at year-end, starting in the UK and Ireland. Two of them drew fewer views across all of July 2025 than a popular music video collects in a day.

The subtraction happened either way. The only question was whether MTV chose it or had it chosen for them, and by 2025 the answer was on a press release.

5. The Subtraction Audit: Two Questions, One Score

List every channel you touch. Not the ones in the deck, the ones that consume a human being’s Tuesday. Score each on two questions, zero to two points apiece.

• Question one: does this get materially better if I do it every single week for two years?
• Question two: would I still want to be doing this in three years, assuming it works exactly as designed?

Anything at 2 or below goes. Below, a real-shaped audit for a nine-person professional services firm running eleven things.

Channel	Compounds	3-yr test	Score	Verdict and hours moved
Email newsletter	2	2	4	KEEP. 2.0 hrs/wk, up from 0.4
Evergreen articles	2	2	4	KEEP. 2.0 hrs/wk, up from 0.5
Referral asks	2	2	4	FOLD IN. Moves into delivery, not marketing
LinkedIn posting	1	2	3	KILL as a channel. Repurpose articles only
Paid search	1	1	2	KILL for now. Frees 0.5 hrs/wk
Instagram	1	0	1	KILL. Frees 0.6 hrs/wk
X / Twitter	0	0	0	KILL. Frees 0.4 hrs/wk
TikTok	1	0	1	KILL. Frees 0.5 hrs/wk
Pinterest	0	0	0	KILL. Frees 0.2 hrs/wk
Quarterly webinar	0	1	1	KILL. Frees 0.7 hrs/wk
Trade show booth	0	1	1	KILL. Frees 0.4 hrs/wk

Table 3. Nine subtracted, two kept. Eleven channels at 27 minutes each become two at roughly two and a half hours, with referral asks folded into how work gets delivered. Same five hours, same budget.

Move the hours, not the budget. Money does not context-switch. People do. Gloria Mark’s UC Irvine work put recovery from an interruption at roughly 23 minutes, the APA’s task-switching review found switch costs can consume up to 40% of productive time, and Sophie Leroy’s attention residue research shows quality degrades before the clock finishes running. An eleven-channel week manufactures interruptions and calls them strategy.

6. The Honest Complication: Concentration Is Real Risk

Anyone selling you focus without saying this part out loud is selling you something. The firm that bet everything on one platform in 2015 has a story, and it deserves a hearing before you pick up the scissors.

The bet	What changed	The damage
Facebook Page reach, 2012 to 2018	Algorithm reweighted toward friends and family	16% organic reach fell to about 6%, then 2% for pages over 500k likes
Publisher reach on Facebook, 2016	News Feed retuned mid-year	Organic reach down 42% Jan to May, 52% by July (SocialFlow, 3,000+ posts)
Google organic, Sept 2023	Helpful Content Update, then folded into core ranking	Of ~400 obliterated sites (Glenn Gabe), only 22% regained 20%+ by Aug 2024. Of Lily Ray’s 130 worst hit, 129 kept falling
HouseFresh	Same update	Lost 95% of search traffic. Recovered Oct 11, 2025, two years and one month later
Zynga on Facebook	Platform terms and player migration to mobile	~94% of bookings from one platform. Revenue $1.3B (2012) to $873M (2013), down 32%. Stock $14.69 to $3.05

Table 4. The casualty file. Every one of these firms was doing the right thing, seriously, on one channel, right up until somebody else changed the rules without asking.

7. How To Concentrate Without Betting the Firm

• Run two, not one, and never eleven. One owned channel you control outright (your list, your site, your customers) plus one rented channel where the audience already lives. Two at two hours beats eleven at twenty-seven minutes, and it survives a platform going hostile.
• Own the asset underneath the channel. Meta halved reach overnight, twice. Google reclassified hundreds of publishers at once. Nobody can reclassify your database. Every rented channel gets one job: move people onto something you own.
• Write the tripwire now, in numbers. “If reach here falls under X for two straight months, I start channel three on the first of the next month.” A pre-written trigger costs nothing and does the job of a permanent 4%-effort hedge.
• Deprecate rather than delete. Leave killed profiles standing with a pinned post pointing to the two that matter. Twenty minutes a quarter keeps the handle and the search result. Re-audit annually: fund one killed channel properly for a month and see if it beats what you dropped it for.

The three-year test does most of the work by itself. Count how many answers arrive as some version of “well, everybody does it.” MTV spent eleven years doing one thing until it reached 112 million homes, then thirty years adding channels until Paramount switched five off in one announcement. Your firm will not fail because you left Pinterest. It will fail because nothing you do ever got enough consecutive repetitions to become anything.

Sources: Fiverr SMB Survey (May 2025, n≈6,000). Taradel 2025. Brinker & Riemersma, 2025 Martech Landscape. HubSpot compounding-post study. Litmus. Schmitt, Skiera & Van den Bulte, J. of Marketing (2011). Binet & Field, IPA; Nielsen ESOV. Orbit Media 2025 (n=808). Social@Ogilvy “Facebook Zero”; Edgerank Checker; SocialFlow. Gabe and Ray HCU cohorts. Mark, UC Irvine (CHI 2005/2008); Leroy (2009); APA. Britannica and EBSCO on MTV; NYT (Feb 1985); Forbes; Paramount, Oct 12, 2025. SEC filings on Zynga.