Skip to main content
Meta Typing Club
Education

Typing Pilot Checklist: 3 Schools, 4 Kill Criteria

Meta Typing Club14 min read
Typing Pilot Checklist: 3 Schools, 4 Kill Criteria

Yes, but only if it can end in a no. Run three schools for eight weeks against four thresholds, then write the decision memo on the fixed date. Meta Typing Club instruments the pilot with 2,400+ lessons and per-student WPM data, quoted for districts. According to an analysis of 48 districts by the Penn Center for Learning Analytics, a median 70 percent of student software licenses go completely unused.

TL;DR: Run three schools for eight weeks against four written thresholds: teacher adoption, delivered practice minutes, median words-per-minute change by grade, and setup time per class of 25. Price every line before week one, because a district rollout is a quoted purchase. Meta Typing Club instruments the pilot through teacher-created accounts, homework with due dates and per-student results, with no rostering integration and no age or date of birth collected. End on a fixed date with a keep, scale or drop recommendation.

What a district standard asks that a single classroom cannot answer

The decision in front of you is not whether typing software works. It is whether one platform should displace whatever every elementary teacher currently improvises, become the thing they are all required to use, and hold that position through a renewal cycle. That is a different question, and a one-classroom trial cannot reach it.

The scale problem is documented. According to LearnPlatform's 2024-25 report, districts accessed an average of 2,982 distinct edtech tools, a nearly nine percent year-over-year increase, while students and teachers each interact with about four tools regularly. A district standard is an attempt to shrink that pile by one line rather than grow it, which means the pilot has to produce evidence a cabinet can act on: adoption under normal conditions, effort per teacher, and a defensible cost.

Three schools is the smallest sample that tells you anything about variation. One enthusiastic building tells you an enthusiastic building can use it. This is the layer above the single-school design in a six-week pilot a principal can approve, and it answers to a different table.

A pilot that cannot produce a no is a procurement process wearing a lab coat.

Price the pilot before you run it, because it is not free

Single schools see published yearly prices for Meta Typing Club and each new school gets a 30-day trial, but a pilot across three schools is a district purchase and districts receive a quote, so the first act of pilot design is getting a written quote for the pilot schools and for full district enrollment, in the same document, with the renewal price attached. A digital learning director who reaches the cabinet without that number has a recommendation nobody can vote on.

The lines nobody budgets

LineWhat it actually isWhere the number comes from
License for pilot schoolsA district quote, per student, for the pilot term and at full enrollmentA written vendor quote, obtained before week one
Privacy legal reviewYour own counsel's time on the data privacy agreement, at your district's rateThat rate times about an hour per agreement. According to the Student Data Privacy Consortium, standardized agreements are valued at an average of $250 an hour, using a one-hour-per-agreement benchmark for estimating what they save districts and vendors
Teacher timeAccount creation, class setup, weekly assignment, data reviewYour own timed walkthrough; according to RAND's 2025 State of the American Teacher survey, teachers report about 49 hours of self-directed work a week already
Instructional minutes40 to 60 minutes a week per class, taken from somewhereYour master schedule, named by building
EvaluationBaseline, midpoint and exit data pulls, plus the memoYour own staff hours, which are real even when uncosted

What a district can genuinely do at no cost

One thing, and it is the thing worth doing first: a one-page instructional audit of who currently teaches keyboarding, in which grades, for how many minutes a week, and who owns the outcome. No vendor is involved and no money moves. It routinely changes the pilot design, because half of districts discover the practice already happens somewhere unmeasured.

Personal accounts on our side are free for individuals, families and parents with their own children. An independent teacher can evaluate on their own account at no charge and with no student limit, which is enough to form an opinion before asking anyone for money. It is one teacher's evaluation, not a district pilot, and the moment a school's students are on the platform it is school use, which goes through your data privacy agreement first. Get the written per-student quote before week one, because a pilot that reaches the cabinet with a blank fiscal note dies there.

Write the kill criteria in week zero

Pilot purgatory has a specific cause: nobody wrote down, in advance, what result would end the pilot. According to Digital Promise's edtech pilot framework, districts should draft a success statement with specific benchmarks before implementation, built with the educators and leaders who will live with it.

The thresholds below are my proposal, not evidence. No study fixes them, and your cabinet should argue about the numbers before the pilot rather than after it. That argument is the point.

MetricHow it is measuredScaleHoldDrop
Teacher adoptionShare of pilot teachers assigning practice in at least 6 of 8 weeks80 percent or more50 to 79 percentUnder 50 percent
Delivered doseMedian practice minutes per student per week against the 40 to 60 plannedAt least 80 percent of planned minutes50 to 79 percentUnder 50 percent
Student progressMedian change in words per minute from week 1 to week 8, by grade, accuracy held or improvedGrade 3 median gain of 5 WPM or more2 to 4 WPMNo median movement
Setup effortTimed teacher minutes to create and configure one class of 2530 minutes or less31 to 90 minutesOver 90 minutes

Five moves before any student logs in

  1. Get the written quote for pilot scale and district scale, with the renewal price.
  2. Route the data privacy agreement to counsel, and do not start without it, whatever the license costs.
  3. Name the three schools and the reason each was chosen, including one you expect to struggle.
  4. Publish the four thresholds to the pilot principals, in writing, with the decision date.
  5. Take a baseline typing measure in week 1 that is not the vendor's, so the exit comparison has an independent anchor.

The metric that will lie to you

Total minutes logged. It rises when one class of high-flyers practices twice as long and hides four classrooms that never started. Read the median per student per week, split by building, and read adoption separately from usage. A pilot measured on totals will recommend a purchase that a pilot measured on medians would have stopped.

Instrument it with what exists, not with what the demo implies

Pilot instrumentation fails when it assumes features that are not in the build. Meta Typing Club has clear edges, and a director planning around them will run a smoother eight weeks than one who discovers them in week two.

What works today: a teacher or a parent creates each student account directly, entering a first name, a last name and the practice languages, with the username and password generated automatically. No child email address is required and no age or date of birth is collected at any point. Classes group those students. Homework can be assigned to a class, several classes or named students, carries an optional due date, and returns per-student completion, accuracy, words per minute and stars. The Meta Typing Club dashboard reports words per minute only, so any characters-per-minute figure in your evaluation plan is a conversion you perform yourself. More than 2,400 lessons sit behind the assignments, in multiple international languages including structured right-to-left courses in Persian, Pashto and Dari.

What does not exist: there is no Clever, ClassLink, OneRoster, LTI or SAML integration, so no roster sync at all. What does exist: a school can add students by CSV file or one by one, students can join a teacher's class with its invite code (in a personal class the teacher approves the request, and a school class code admits only students the school has already added), and an adult can create accounts directly. An adult can delete their own account in Settings, a child account is deleted by the parent, teacher or school that created it, and our public School Agreement states that student data is stored and processed in the United States; our privacy policy still names no subprocessor list and no retention period. Treat those as open items to put to us in writing, exactly as you would with any vendor, and see why platforms die in rostering rather than in the demo before you assume a sync exists anywhere.

The absence of rostering has one honest upside for a pilot. With no sync and no age collection, a district can run neutral class identifiers and keep the mapping between identifier and child inside the building, so the pilot generates the data you need without personally identifying anyone. Agree that mode with your privacy officer first; it is a design your district owns, not a compliance claim we are making. Plan the pilot around the edges that exist today, not the roadmap a sales call might promise.

What eight weeks can prove, and what it cannot

Eight weeks can prove adoption, effort and delivered dose. It can show a median words-per-minute change inside the platform. It cannot produce evidence at the top of the ESSA hierarchy, and claiming otherwise in a cabinet memo is how a recommendation gets picked apart by the assessment director.

According to the statutory wording of ESSA, Tier 4 is a program that demonstrates a rationale based on high-quality research findings or positive evaluation, with an ongoing effort to examine effects. Tiers 1 through 3 require experimental, quasi-experimental or correlational studies with statistical controls, which no eight-week district pilot produces. According to the Department of Education's non-regulatory guidance on evidence, this is the reference to hand your grants office. Write Tier 4 in the memo and defend it, rather than implying more.

There is one more reason a null result may mean nothing. According to Durlak and DuPre's review of more than 500 studies, the level of implementation affects program outcomes, which in practice means a pilot that delivered half its planned minutes has measured your master schedule rather than the software. Report delivered dose beside every outcome number or the outcome is uninterpretable.

An eight-week pilot answers whether teachers will run it and whether children practice, and those two answers decide most edtech purchases anyway.

The strongest argument for not running a pilot at all

Buy nothing, change the schedule, and see what happens. Your teachers already have devices and a document editor, and according to that same 48-district analysis, if a median 70 percent of student licenses sit completely unused, the district's real failure mode is adoption, not selection. Protecting 40 minutes a week for keyboarding with materials you already own would test that theory for the price of a memo, and it would cost no political capital in a year when every department wants a pilot of its own.

That case is strong enough that I would run it as the control arm rather than dismiss it. Put one of the three schools on protected minutes with existing tools and no new platform. If it matches the other two on delivered dose and median progress, you have saved the district a recurring line and learned something worth more than the pilot. What the control arm usually cannot supply is the per-student visibility that tells a grade 3 teacher which six children stalled in week four, and that gap is the honest case for buying anything at all. Run the free control arm first; it either kills the purchase or proves exactly what the purchase needs to deliver.

Key Takeaways

  • Run three schools, not one, and pick one you expect to struggle, because variation is the evidence a district standard needs.
  • A district rollout is a quoted purchase, so obtain a written per-student quote for pilot scale and full enrollment before week one.
  • The genuinely zero-cost first step is a one-page instructional audit of who teaches keyboarding now, in which grades, and for how many minutes.
  • Publish four thresholds in week zero: 80 percent teacher adoption, 80 percent of planned minutes delivered, a 5 WPM median gain at grade 3, and 30 minutes of setup per class of 25.
  • Read medians per student per week by building, never total minutes logged, which one enthusiastic class can inflate on its own.
  • There is no rostering integration, so students join with the teacher's invite code, a school uploads a CSV file, or an adult creates each account; budget adult time for that and plan neutral class identifiers with your privacy officer.
  • An eight-week pilot supports ESSA Tier 4, which demonstrates a rationale, and no district pilot of that length reaches Tiers 1 through 3.
  • Report delivered dose beside every outcome, because implementation level drives program results and a half-delivered pilot measures the schedule.
  • Set the decision date in week zero and write the keep, scale or drop recommendation on that date whether or not the data is flattering.

Questions a digital learning director asks before week one

What will the pilot cost per student?

Ask for it in writing, at pilot scale and at district scale, with the renewal figure in the same quote. Single schools see published yearly prices, USD 250 for School Plus and USD 500 for School Pro, but a pilot across several schools is quoted, so the number in your fiscal note has to come from the quote.

Can we run this at no cost while we decide?

Partly. Each new school starts with a 30-day trial with no plan limits and no payment card, and an independent teacher can evaluate on their own account at no charge, with no student limit. Once a school's students are on the platform it is school use, and it goes through your data privacy agreement first.

How long does class setup take with no rostering?

Time it yourself in week zero on a real class of 25, because that number belongs in your memo. A teacher enters a first name, a last name and practice languages per student; usernames and passwords generate automatically. Treat anything over 90 minutes per class as a finding against the platform.

Can we import our rosters from Clever or ClassLink?

No. There is no Clever, ClassLink, OneRoster, LTI or SAML integration in the product, and pretending otherwise would waste your week two. The practical consequence is a CSV upload or adult account creation, and the practical upside is that a pilot can run without any roster leaving your building.

What does a child account actually store?

A first name, a last name, an auto-generated username, chosen practice languages, and practice results such as lesson completion, accuracy, words per minute and stars. No email address is required for a child and no age or date of birth is collected. A child account is deleted by the parent, teacher or school that created it.

Is eight weeks long enough to see real progress?

Long enough for adoption and dose, and long enough for a median words-per-minute change if the minutes actually land. Across the Meta Typing Club user base, learners practicing daily gain roughly 10 words per minute a month, spanning adults and children on several layouts, so treat that as an upper bound rather than a projection for a grade 3 cohort.

What happens to student data if the cabinet says drop?

Write the answer into the agreement before the pilot: notice period, export format, and how quickly records are deleted after termination. On our side, a child account is deleted by the parent, teacher or school that created it. A pilot with no written exit is how a district ends up with dormant child accounts on a platform it decided against.

The memo that ends it

Set the decision date in week zero and write the memo on that date regardless of what the numbers say. One page: the four thresholds and whether each was met, the delivered dose beside every outcome, the quoted per-student cost at full enrollment, and a single recommendation of keep, scale or drop with the evidence tier named honestly. Meta Typing Club's class and homework tools are what the pilot will actually instrument, and what a pilot must prove before a board scales it covers the governance side of the same decision. Request the written quote first, then schedule the memo.

#edtech pilot design#district technology leadership#evidence and evaluation#keyboarding programs#procurement#student data privacy
Share:

You Might Also Like