/decisions · 2026-08-30 · printables · step-1 memo
What exists already. Two packs built and critic-gated today: a 6-page first pack and a 7-page bilingual early-reading pack that went 4 rounds of fresh-eyes critic before proofs. The generation loop is real, not a plan. The wife marked up the first printed pack and returned photos. That feedback channel is real but one cycle old, and it is not unique to us; any parent selling on Etsy has the same loop. What it is, is fast. The household is a live test user.
What this page asks. Three decisions: a brand call, a yes-or-amend on step-1 scope, and a yes-or-amend on pre-registered kill numbers. Two are cheap to reverse. The brand call is the expensive one to undo, and section 6(a) says why. Nothing here commits to a price or to spending money.
The honest frame. This is currently a hobby that produces a real artifact. Step 1 is the cheapest possible test of whether anyone outside the house wants it. The kill criteria below exist so that answer arrives on a date instead of never.
Settled by the founder 2026-08-30 and reproduced here for orientation only. The ordering is not up for debate on this page.
The catalog opens with 12 to 15 activities. Not a trickle, not a hundred. Twelve is roughly the point at which a visitor reads the page as a catalog rather than a personal blog, and fifteen is about the ceiling on what the home loop can produce and critic-gate at quality before the first fall holiday deadline.
meta.yaml files. The existing tag axes (age_grade, activity_types, interests, holiday, skills, language) become the site navigation directly. The research note found that Twinkl and Education.com both organize age-band first, then skill, which is the same two-axis shape we already independently landed on. The taxonomy does not need redesigning. Worth noting the dissenting example: 123 Homeschool 4 Me organizes by themed pack rather than by single-sheet taxonomy, so the two-axis pattern is the majority shape among the sites checked, not a universal one.Two things must be true in the step-1 metadata or rung 2 becomes a rewrite instead of an extension:
Engine-level TODO, not a step-1 blocker: render profiles (full-color, accents, pure black-and-white). The founder's ruling is that the engine designs for quality and ink is a runtime profile per household. Step 1 can ship a single profile and add the others before any outside household prints anything at scale.
Marginal cost is effectively zero. Generation happens on a machine already running. Cloudflare Pages and R2 for a catalog of this size should land in the pennies-per-month range. That is an unpriced estimate, not a rate check, and it is the one number on this page with no source behind it. There is no cost-of-goods argument against trying this; the only real cost is founder attention, which is why section 5 exists.
Data quality warning, read first Every figure below is search-result-snippet level, not live-pricing-page verified. Twinkl, Education.com, and Etsy all render pricing through JavaScript and returned empty bodies or HTTP 403 to automated fetches, and no browser session was available. The URLs are real and are where the figures came from, but none of these numbers were read off the vendor's own live pricing page today. Treat them as the right order of magnitude, not as quotable prices. Before anything here informs an actual price tag, someone should open these three pages in a browser and confirm.
| Competitor | Reported price | Source |
|---|---|---|
| Twinkl (US individual) | Roughly $9 to $10 per month; about $89.98 per year. The $9-versus-$10 discrepancy appears across two of Twinkl's own pages and is unresolved. School and organization pricing is quote-based, no published figure found. | twinkl.zendesk.com subscription-price article; twinkl.com US subscription review post |
| Education.com Premium | About $15.99 per month standard; about $9.99 per month billed annually (roughly $120 per year). Heavily and frequently discounted: one promotion cited at $47.95 per year. Free tier reported as 3 downloads per month. | techjury.net pricing writeup; slickdeals.net promo listing |
| Etsy printable packs | Roughly $1 to $5 for single sheets and mini sets; roughly $25 to $50 for mid-size bundles; $100 to $260 for mega-bundles, nearly always displayed as a discount off a crossed-out higher price. One specific listing price surfaced in search results at $3.81 (a preschool-to-kindergarten music worksheet bundle), unverified against the live listing. | Music Worksheet Bundle, Preschool to Kindergarten. Other listing prices could not be captured; every Etsy listing fetch returned HTTP 403. |
| Teachers Pay Teachers | Individual preschool items commonly $1.25 to $3.25. Bundles marketed around $25 against theatrical "value" claims. | teacherspayteachers.com preschool browse. Aggregate from browse-page snippets; no individual listing verified. |
| Bilingual English/Spanish specifically | No dominant branded competitor surfaced. What exists is scattered individual sellers on Etsy and Teachers Pay Teachers with narrow catalogs. No price point could be captured for any of them. On the free side, Colorín Colorado is the clearest institutional resource, covering literacy guidance rather than a printable curriculum. | Bilingual Beginnings store on Teachers Pay Teachers; Spanish preschool worksheets listing on Etsy; colorincolorado.org |
What the landscape says. Two tiers: subscription platforms in the $90-to-$120-per-year range, and marketplace one-offs at $1 to $5 a sheet with bundles at $25 to $50. The marketplace list prices are visibly anchored, near-permanently "on sale," so the real transaction prices are lower than the numbers above suggest.
On the bilingual wedge. The search did not surface a dominant bilingual competitor, which is consistent with the wedge thesis. But a caution worth stating: absence of evidence from a snippet-level search is weak evidence of absence. It is entirely possible that a well-established bilingual competitor exists and simply did not rank for the queries run. What can be said honestly is that the large English-language incumbents do not visibly serve this niche, and that the niche's paid side appears to be individual sellers rather than platforms.
| Option | Shape | The catch |
|---|---|---|
| Free catalog, paid packs | Everything browsable and a meaningful slice downloadable free; themed and seasonal bundles priced individually. | Every sale is a fresh act of persuasion. No recurring revenue. But it is the only model that works with zero audience on day one, because the free tier is the discovery mechanism. |
| Subscription | Recurring access to the whole catalog, the way the large incumbents sell. | A subscription is a promise of continuous new material. At 2 to 3 activities a week from a home loop, that promise is fragile the first busy month. Note that this argument inherits the cadence uncertainty flagged in section 2: the rate itself is a projection off one day, so the fragility case is directional rather than measured. It also needs an audience before it needs a price. |
| Neither yet | Ship the catalog entirely free at step 1 and instrument it. | Zero revenue, and no evidence about willingness to pay. What it buys is a clean read on whether anyone outside the house shows up at all. |
Recommendation Ship step 1 entirely free and instrumented; do not set a price yet. Reasoning: the step-1 question is discovery, not monetization, and a price tag added on day one would confound the only signal step 1 can produce. Revisit pricing when there is a non-zero number in the kill-criteria dashboard, and revisit it as a free-catalog-plus-paid-packs shape rather than a subscription, because the production cadence cannot honestly underwrite a subscription promise yet. This is a recommendation, not a decision already taken.
Discovery is the whole problem. Not production, not quality, not cost. A new static site with no audience, no backlinks, and no advertising budget is competing for search terms against sites with years of accumulated authority. This is the same wall every content product hits, and being right about the bilingual gap does not get you past it.
The ladder is what mitigates this, and it is worth noticing that the founder's ordering already accounts for it. Rung 1 exists partly to build the surface that search can index at all. Rung 3, Model Context Protocol and agentic access, is explicitly a discovery route in his own framing: if parents increasingly ask an assistant for a worksheet rather than searching for one, being reachable by that assistant is a distribution channel that does not require out-competing incumbents on page rank. That is a bet on how discovery changes, not a guarantee, and it should be held as a bet.
The second risk, and arguably the stronger one: the market largely clears at zero. Education.com gives away three downloads a month, 123 Homeschool 4 Me gives away whole themed packs, and the free supply in this category is effectively unbounded. "Search ranking is hard" is a solvable problem given time. "A parent can get something good enough for free in thirty seconds" is a harder objection to the paid path, and it is a real argument for the recommendation above to ship free and compete on fit rather than on price.
The bilingual angle is the wedge for both risks: a narrower query is a cheaper query to rank for, and it is the place where "good enough for free" is least true. It is an advantage in degree, not a moat.
The point of writing these before launching is that after launching, a founder who has grown fond of a thing will find a reason every number was unfair. All three numbers below are proposals and are exactly what the founder is being asked to amend.
Read at 90 days after launch is declared. A 60-day read is too early for search indexing to mean anything; 90 is the earliest an organic-traffic zero is honest evidence rather than impatience.
| Signal | Proposed kill number | Why this number |
|---|---|---|
| Organic search arrivals | Fewer than 25 organic search impressions per month, and zero organic clicks, in month 3 after launch | Zero clicks with meaningful impressions means a fixable presentation problem. Near-zero impressions means the site is invisible, which is not fixable by trying harder at the same thing. |
| Non-family downloads | Fewer than 10 downloads from outside the household and the founder's own circle, across the entire catalog, in the 90 days after launch | Ten is deliberately a low bar. Failing a low bar is much stronger evidence than missing an ambitious one. |
| Founder time | More than 2 hours per week of the founder's own time, sustained across any 4-week window after launch | This is the only genuinely scarce input. phData is the main bet; a side project that quietly starts costing half a day a week has changed what it is without anyone deciding to. |
"Kill" means demote, not delete. Hitting these means the printables engine goes back to being a household tool that produces excellent pages for one child, which is a genuinely good outcome and was the original point. It does not mean deleting anything. The library keeps every print, per the standing rule.
Three directions. No strong push from me; the arguments are close and a name is cheap to change at rung 1 and expensive at rung 4. Any specific names below are illustrative sketches, not availability-checked. Availability was checked for a concrete shortlist on the studio charter page (2026-08-31); this section states the directions, that page states the facts.
The 12-to-15 opening set in section 2, static Astro, engine local, roughly 2 to 3 new activities a week.
The three proposed numbers in section 5, read at 90 days.
Umbrella, own name, or your own name. A direction is enough; the exact name can come later.
Send brand callYes as written, or say what to change: the set list, the count, or the cadence.
Send scope callYes to the three numbers, or give me your numbers. Amending them is the expected outcome.
Send kill criteriaKill the product frame entirely. The engine stays a household tool and keeps producing packs. One-line reason.
Archive + sendPush this to a date. The home loop keeps running either way; only the public catalog waits.
Defer + send