Drip vs All-at-Once vs Challenge: Which Format Maximizes MAC Completion (and Is the 70-80% Challenge Number Real)?

The question

Drip vs all-at-once vs challenge format for MAC: which maximizes completion given email-native delivery, and should MAC be reframed as a paid challenge (70-80% completion) rather than a self-paced course (3-5%)? Context: derived from the 2026-06-13 failure-mode brief, where catastrophic non-consumption (3-5% completion) was named the silent compounding killer of low-ticket products — completion drives testimonials, referrals, avoided refunds, and backend conversion.

What we already know (from the vault)

• MAC is an executable course, not a passive one — which already breaks the 3-5% framing. Per [[2026-05-04-mac-product-shape-decisions]], the buyer installs a Claude plugin and runs the 8 phases against their own warehouse, producing real artifacts (data model, tests, docs, deployment plan). "Do something, not watch something" is the exact mechanism the challenge literature credits for completion lift — MAC has it natively. Price band is $300-2k (founder leaning higher), above the question's stated $50-200.
• A free 5-day email drip already exists in the funnel — but as acquisition, not the product. [[2026-05-04-mac-product-shape-decisions]] specifies a "give-give-give" Day 1-5 drip ("5 days of mid-market analytics gotchas") that nurtures toward the paid install. This is a ready-made challenge skeleton: it is already daily, already time-boxed, already email-native.
• Non-consumption is the named killer, and email-native delivery is the prescribed fix. [[2026-06-13-low-ticket-digital-product-failure-modes]] flags 3-5% self-paced completion as failure mode #5 and notes the root cause is delivery friction ("log into a platform you never visit"), not motivation — and that drip can raise completion "only if delivery friction is low." It explicitly leaves this exact drip-vs-challenge question as an open follow-up.
• MAC should be the front rung of a ladder, and completion is what powers the ascension. [[2026-06-13-low-ticket-digital-product-failure-modes]] and [[2026-05-18-low-ticket-dev-tool-launch-friction]] both argue the backend (cohort/templates/done-with-you) is where profit lives; completion is the gate to that upgrade path (the "83% of challenge completers buy the premium offer" claim).

What the web says

• The low baseline is the well-measured part. MOOC completion averaged ~3% in the Reich & Ruiperez-Valiente MIT study; even Harvard/MIT/Stanford top out around 20% (Kizilcec et al.); Udemy's best instructors see ~8% (learnopoly.com). The 3-5% self-paced figure is real and peer-reviewed.
• Challenge completion is claimed at 70-80% (~14x baseline), driven by four mechanisms: fixed start/end deadline (urgency), cohort moving together (visibility/accountability), a daily output requirement (active vs passive), and low-friction native delivery — with re-engagement when students go quiet (communipass.com challenge funnel, communipass.com completion problem).
• Cohort-based course (CBC) completion is claimed at 85-95%+, but the credible anchors are narrower: altMBA 96% (cited via HBR), Section4 ~70%+ (TechCrunch). Most company-specific 90-100% figures (Esme, Disco) are vendor self-reports (learnopoly.com, maven.com/a16z).
• The one apples-to-apples within-platform comparison is far more modest: Ruzuku reports scheduled cohort courses at 64.2% vs 48.2% for open self-paced — a real lift, but ~1.3x, not 14x (learnopoly.com).
• Drip beats all-at-once for retention, but it's advisory, not measured. Course-platform sources agree all-at-once "overwhelms students → lower completion," while drip prevents overload and creates return-visits — but none cite a controlled completion-delta. Best-approach-depends-on-audience is the consistent caveat (learndash.com, convology.com).

Convergences and contradictions

• Direction is robust; magnitude is folklore. Web and vault agree: deadline + accountability + daily doing + native delivery raises completion. But the 70-80% / 90%+ headline numbers come overwhelmingly from course-platform vendors selling cohort tooling (communipass, Disco, Esme) — they are marketing claims, not independent measurement. The peer-reviewed evidence only anchors the low baseline (3-20%). The single neutral within-platform comparison (Ruzuku 64% vs 48%) suggests the true scheduled-vs-evergreen lift is meaningful but closer to ~1.3-1.5x than 14x. The "83% of completers buy premium" stat is also vendor-sourced and should be treated as a hypothesis, not a planning input.
• The 3-5% baseline doesn't even apply to MAC. That number is for passive video self-paced courses. MAC is install-and-execute with artifact output — it sits in a different category before any challenge wrapper is added. The honest comparison is "executable course, self-serve" vs "executable course, time-boxed," not "video course (3%)" vs "challenge (75%)."

Synthesis for RDCO

Recommendation: ship MAC as a drip-paced, time-boxed cohort challenge — but treat it as an A/B-able wrapper around the same executable IP, not a rebuild. The format question is mostly a sequencing-and-deadline decision, and all three options share the identical Layer-1 playbook. The lowest-cost, highest-completion path is: keep MAC's 8 phases, gate them to a fixed N-day calendar (one phase per day or per two days), wrap a start/end date and a "ship your artifact by Friday" deadline around it, and deliver the daily prompt + nudge email-natively (Resend, the channel already specced for the free drip). All-at-once is the format to avoid: it maximizes overwhelm and abandonment with zero offsetting benefit for a product whose whole value is doing the work. Pure evergreen self-paced is the second-worst — it strips out the one cheap lever (deadline urgency) RDCO gets for free.

The mechanism that actually drives the lift is not "cohort" — it's deadline + daily output + low-friction delivery, and MAC already has two of the three. MAC's executable nature supplies "daily output" for free (you produce a real artifact each phase). Email-native delivery supplies low friction (no platform login). The only missing ingredient is the deadline/accountability layer, and that is exactly what a challenge wrapper adds at near-zero marginal cost. RDCO should be skeptical that adding live cohort facilitation (the expensive, non-scaling part of CBCs) is worth it at v1 — the altMBA 96% number bundles live human teaching that RDCO can't staff solo. The cheap 80% of the benefit is the deadline and the daily cadence; capture that first, measure, and only add live/cohort accountability if instrumented completion stalls.

Minimum-viable challenge wrapper (solo-operator, no live facilitation): (1) Fixed enrollment window with a named start date — buyers all begin the same Monday (creates the cohort/visibility effect without a community platform). (2) One email per day for ~5-8 days, each unlocking one MAC phase with a single concrete "ship this artifact today" ask. (3) A public-ish completion artifact (the buyer's own test-coverage report / data model) as the daily output that creates self-accountability. (4) A re-engagement nudge if a phase goes uncompleted by the next day. (5) A completion moment ("you shipped X") that triggers the testimonial ask and the backend upgrade offer. This is the existing free 5-day drip's structure, promoted from lead-magnet to paid-product spine. Instrument per-phase completion from day one — it is the metric, per the 2026-06-13 brief, that the entire literature is silent on measuring honestly.

Be honest about the number with the founder: do not put "70-80% completion" or "83% upgrade" in MAC's own planning math or sales copy as fact — they are vendor folklore. The defensible claims are: (a) self-paced passive completion is genuinely 3-20% (peer-reviewed); (b) scheduling + deadlines produce a real, repeatable completion lift (direction is solid); (c) the magnitude of that lift, measured neutrally, looks more like 1.3-1.5x than 14x. MAC's executable format plausibly does better than a typical course because output is forced — but that is a hypothesis to measure on RDCO's own first cohort, not a number to import.

Open follow-ups

• What N-day length fits MAC's 8 phases without forcing more than ~1-2 hrs/day of buyer effort (over-packed daily asks re-introduce the overwhelm that kills completion)?
• Does a fixed-start cohort window hurt acquisition (buyers must wait for the next Monday) more than it helps completion? Test rolling vs fixed cohorts.
• Can RDCO instrument per-phase completion technically — does the plugin phone home (privacy-respecting) on phase completion, or is completion self-reported via email replies?
• At $300-2k, is a paid challenge the right v1, or should the free 5-day drip stay free-as-acquisition and the challenge wrapper apply only to the paid install? (Don't conflate the lead-magnet drip with the product.)
• Validate the "completers convert to backend" assumption against RDCO's own first cohort before designing the ladder economics on the 83% vendor stat.

Related

• [[2026-06-13-low-ticket-digital-product-failure-modes]]
• [[2026-05-04-mac-product-shape-decisions]]
• [[2026-05-18-low-ticket-dev-tool-launch-friction]]
• [[2026-04-21-nicolas-cole-10-lessons-15m-digital-products]]

Sources

• Vault: /Users/ray/rdco-vault/06-reference/research/2026-06-13-low-ticket-digital-product-failure-modes.md
• Vault: /Users/ray/rdco-vault/01-projects/mac/2026-05-04-mac-product-shape-decisions.md
• Vault: /Users/ray/rdco-vault/06-reference/research/2026-05-18-low-ticket-dev-tool-launch-friction.md
• Vault: /Users/ray/rdco-vault/06-reference/2026-04-21-nicolas-cole-10-lessons-15m-digital-products.md
• Web: https://learnopoly.com/cohort-based-learning-statistics/ (aggregates Reich & Ruiperez-Valiente MIT MOOC study, Kizilcec et al., altMBA/HBR, Section4/TechCrunch, Ruzuku 64.2% vs 48.2%)
• Web: https://communipass.com/blog/9-day-challenge-funnel-for-course-creators-2026-conversion-framework/ (challenge 70-80% claim + four mechanisms — vendor source)
• Web: https://communipass.com/blog/course-completion-rate-problem/ (87% drop-out, root causes of low self-paced completion — vendor source)
• Web: https://maven.com/resources/a16z-cohorts-are-king (cohort-based course thesis)
• Web: https://www.learndash.com/blog/why-you-should-use-drip-content-for-your-courses/ (drip vs all-at-once retention argument — advisory)
• Web: https://www.convology.com/dripping-course-content-vs-full-release-which-is-better/ (drip vs full release tradeoffs — advisory)