What a Typing Pilot Must Prove Before Your Board Scales It

A pilot can prove that a typing program is implementable in your district. It cannot prove that the program works, and a single-school pilot never will. Holding those two apart is most of a trustee's job on this vote, and it gets much easier if the board writes down what the pilot has to show before the pilot begins.

The pattern you are guarding against is familiar enough to name. One building runs a program for a semester. In April a clean deck arrives with a growth chart, three student quotes and a recommendation to buy district-wide. Nobody in the room can separate the program from the teacher who ran it.

What a one-school pilot can prove, and what it cannot

Federal law already gives you the vocabulary. The U.S. Department of Education's non-regulatory guidance Using Evidence to Strengthen Education Investments lays out four tiers of evidence: strong, moderate, promising, and demonstrates a rationale. The statutory wording of that fourth tier, as reproduced by Ohio's ESSA evidence clearinghouse, covers an intervention that "demonstrates a rationale based on high-quality research findings or positive evaluation that such activity, strategy or intervention is likely to improve student outcomes or other relevant outcomes." Tier 3, promising, requires at minimum "one well-designed and well-implemented correlational study."

Your pilot has no comparison group, no random assignment and no statistical controls. It lands in Tier 4. That is not a criticism of the pilot. It is the definition of the design, and every vendor who tells you otherwise has told you something you can check. Note also that federal school-improvement funding is held to the top three tiers while most other federal titles will accept a documented rationale, so where the money comes from changes what your evidence has to carry.

Which makes the board's real question narrower, and far more answerable, than "does keyboarding instruction work." It is this: can this district run this program in ordinary buildings, with ordinary teachers, for the students we actually enroll?

Four tests to write into the motion before the pilot starts

Digital Promise's edtech pilot planning guidance asks districts to "draft a statement that describes what success would look like at the end of the pilot and develop specific benchmarks for that success," and to build that statement collaboratively with the educators, leaders and families involved. That happens in the planning step, well before anyone signs in. In practice the board almost never sees it. Ask for it, in writing, attached to the authorizing motion.

Four tests carry most of the weight for a keyboarding program specifically.

Test	Passes if	Fails if
1. Per-student fluency growth	The median student in each participating class shows measurable gain in words per minute with accuracy held steady or improved, reported student by student	Only a class average or the top quartile appears, or speed is reported without accuracy beside it
2. Adoption without coercion	The least enthusiastic teacher in the building used it, and can state what it cost in minutes per week	Usage concentrates in one or two classrooms, or participation was directed from the front office
3. Parent visibility	Families of participating students can say what their child is working on and roughly where they stand	Progress lives only in a teacher dashboard that nobody outside the building has opened
4. Performance across language communities	Growth is disaggregated by the district's multilingual and newcomer student groups, and holds up there	The pilot ran only in the district's English-dominant school

Test 2 is the one trustees underweight, and it is the one that predicts what happens in year two.

Why the language test is a pass line and not a footnote

National Center for Education Statistics data puts English learners at 10.6 percent of US public school students in fall 2021, about 5.3 million children, with state figures spread from 0.8 percent in West Virginia to 20.2 percent in Texas. Districts sit all along that range, and so do buildings inside a single district.

A keyboarding pilot run in the district's most English-dominant elementary school tells you very little about the school where a third of the class arrived in the last two years. Those students often have the largest gap between what they can say and what they can get onto a screen, and they are frequently learning a second keyboard as well as a second language. If the pilot did not include them, it was not run in your district. It was run in the easiest part of it.

The enthusiastic teacher problem

Every district has the teacher who makes anything work. She stays late, she builds her own reward chart, she troubleshoots the login problem herself instead of filing a ticket. Give her a mediocre product and the pilot data will look excellent.

The number that survives her departure is not on the slide. It is minutes per week of teacher labor, reported by someone who did not volunteer.

The market gives you a reason to insist on this. LearnPlatform by Instructure's EdTech Top 40 analysis found that districts accessed an average of 2,982 distinct edtech tools in 2024-25, a nearly 9 percent year-over-year increase, drawn from more than 64 billion interactions across 3.7 million students. The following year's report, based on 12.6 million users inside Canvas, found that individual students and teachers interact with an average of four tools. A district can hold thousands of licenses while a classroom uses four things. Purchase volume and instructional reality have come apart, and the gap is precisely what a scaling vote is supposed to catch.

Where this framework is wrong

The strongest objection to everything above is that it strangles good programs in the crib. A volunteer-led pilot in one willing building is how most worthwhile practice enters a district in the first place. Demand representativeness in year one and you guarantee the district only ever adopts things that were already easy, which is a good description of how districts end up with nothing new.

That objection is largely right, and the answer is not a lower bar but a staged one. Let the first semester be volunteers, and treat its output as a feasibility report rather than evidence. Then run one further semester with an assigned cohort that did not raise its hand, and put the scale vote after that. It costs one additional term. That term is the cheapest insurance a board can buy against a three-year contract that quietly stops being used in month five.

What has to be signed before children log in

Before any of the above matters, your counsel has a prior question: on what legal basis do our students use this at all?

Under FERPA, a vendor can be treated as a school official with a legitimate educational interest, but only on conditions. The Department of Education's student privacy guidance states the outside party must perform "an institutional service or function for which the agency or institution would otherwise use employees," must be "under the direct control of the agency or institution with respect to the use and maintenance of education records," must be subject to the restriction that personally identifiable information "may be used only for the purposes for which the disclosure was made," and must meet the school official criteria in the district's annual FERPA notification. Direct control in practice means a signed contract.

For students under 13, the Federal Trade Commission's COPPA guidance governs the collection of personal information online and sets out the narrow circumstances in which a school may act in place of a parent for consent. State law is often the sharper constraint. Illinois SOPPA, at 105 ILCS 85, requires an operator to "enter into a written agreement with the school, school district, or State Board before the covered information may be transferred," and requires the school to post each executed agreement publicly within 10 business days. New York's 8 NYCRR Part 121, implementing Education Law 2-d, carries a dedicated Bill of Rights for Data Privacy and Security and a separate section of obligations for third-party contractors.

Most districts route all of this through a standard instrument rather than bespoke drafting. The Student Data Privacy Consortium reports that over 222,000 standard Data Privacy Agreements have been executed since 2016, with its National Data Privacy Agreement developed by 28 state alliances. Whether a vendor will sign your state's version, unedited, is frequently the entire decision.

Here is where we have to speak about our own product plainly, because a board should not accept a compliance claim from any vendor's blog. Meta Typing Club collects no age and no date of birth from any account, and a teacher or parent creates the student account directly, so no child email address is required. Those are checkable facts and they matter to a COPPA conversation. What we cannot offer you is a documented data residency position: our privacy policy names no storage location, no sub-processors and no retention period, and account deletion is handled by email request rather than a self-service button. So do not take our word for a compliance posture. Put the requirements in the DPA and make us sign them, exactly as you would with any other vendor.

There is a pilot mode that sidesteps most of this while the paperwork is negotiated. Accounts can be created under neutral class identifiers, with the roster mapping held inside the school, so nothing personally identifying leaves the building. The honest limit: the moment real student names go into those fields, you are creating education records and the full regime applies. That is a conversation for your privacy officer, not for a vendor to settle.

The cost question, answered

Boards get vague answers here, and the vagueness is the signal. So, straight: free covers personal use, meaning an individual learner, a family, or a parent working with their own child. An independent teacher, on their own account, can run one class of up to five students. School, district and organizational use, including a teacher acting for their school, is commercial and priced on request. There is no published per-seat price and no self-serve purchase, so a district figure comes from a quote via our contact page and nowhere else, and a dedicated school route is planned rather than built.

Note what that rules out. The free teacher account is not a side door around the review this post has just told you to demand. The moment your students are on the platform it is school use, it creates education records, and the executed data privacy agreement comes first, free or not. A vendor that tells you to require a DPA and then shows a teacher how to skip it has told on itself.

What that means for your motion is simple enough, and it is a sequence rather than a single step. No trustee can carry a motion to a finance committee with a blank fiscal note, so do not start with the vendor. Start with the move that is genuinely free and genuinely board-sized: a one-page instructional audit of who teaches keyboarding in this district, in which grades, for how many minutes a week, and who owns the outcome. It requires no vendor, no budget line and no signature, and it is defensible in open session. Then have the costed conversation about school access, with the quoted per-student annual price, the term, the renewal terms and the exit terms in the board packet before the scaling vote, from us or from anyone else. A pilot that never produced a price has not tested the decision you are actually being asked to make.

Draft motion language a board can amend

Boards ask for pilots and receive presentations because the authorizing motion asked for nothing specific. This is a starting point, not legal drafting, and your counsel and policy committee should mark it up.

Moved, that the Board authorize a keyboarding pilot at [school] for the [term] semester, subject to a fully executed data privacy agreement on the state template prior to any student account creation; and that no proposal to expand the program be placed before the Board unless the accompanying report contains: (a) per-student words-per-minute and accuracy growth for every participating student, not class averages; (b) weekly usage by classroom, including classrooms whose participation was assigned rather than volunteered; (c) teacher time cost in minutes per week, reported by participating teachers; (d) results disaggregated for multilingual and newcomer students; (e) a quoted per-student annual price with term, renewal and exit provisions; and (f) a written statement of the success benchmarks agreed before the pilot began, with actual results against each.

Item (f) is the one that does the work. A success statement written before the pilot cannot be redrawn around whatever the pilot happened to produce.

Key takeaways

• A one-school pilot sits in ESSA Tier 4, which demonstrates a rationale rather than an effect. Treat any efficacy claim from it as overreach.
• Write the pass/fail benchmarks into the authorizing motion, before the pilot starts, and require them back unchanged.
• Report per-student growth in words per minute with accuracy beside it. Class averages hide the students the program is failing.
• Teacher time cost, reported by a non-volunteer, predicts year-two survival better than any growth chart.
• Districts accessed an average of 2,982 edtech tools in 2024-25 while classrooms interact with about four. Adoption evidence is not optional.
• If the pilot excluded the district's multilingual and newcomer students, it did not test the district.
• Require the executed DPA on your state template before any student account is created, not after the vote.
• Run the free one-page instructional audit first, then hold the costed conversation. Get the quoted per-student annual price, term and exit terms into the packet.
• School and district use of Meta Typing Club is commercial and priced on request. An independent teacher's free account tops out at one class of five students, and it is never a route around the district's privacy review.

Frequently asked questions

What does this cost per student per year?

There is no published figure. Free covers personal use: an individual learner, a family, a parent with their own child. An independent teacher on their own account can run one class of up to five students. School, district and organizational use is commercial and quoted on request, so your business office needs that quote in writing before the vote, and a class of your students on the platform is school use whichever account created it. The free step available to you tonight is not a trial, it is the one-page audit of what keyboarding your district already teaches and who owns it. Any vendor answering "free to start" when you asked for a per-pupil annual price has not answered you.

How long should the pilot run before we vote?

Long enough to cross a full grading period and at least one staffing disruption, which in most calendars means a semester. If the first semester ran on volunteers, add a second with an assigned cohort before scaling. Two terms is inconvenient. Reversing a district-wide rollout is worse.

Who signs, the board or the superintendent?

That depends on your policy and your bid threshold. Typically the board approves the budget line and the contract above threshold, staff select the product, and purchasing runs the process. The board's real power here is the authorizing motion and the evidence it demands, not the product choice.

How much teacher time does this take each week?

Ask the pilot, and require the answer from a teacher who did not volunteer. For a keyboarding program the honest components are initial class setup, occasional account fixes, and whatever review of dashboards the school expects. If nobody in the pilot can give you a number in minutes, that is itself a finding.

What happens to student data if we cancel?

Put deletion, return and timelines in the DPA, because platform defaults vary and ours is not self-serve: Meta Typing Club handles account deletion by email request. Ask any vendor for the deletion mechanism, the completion window, and what survives in backups.

Does this duplicate something we already pay for?

Frequently, and it is worth checking before the vote. Many districts already hold a typing tool inside a rostering suite or a state-provided platform. The questions that separate them are structured lesson progression, per-student reporting your teachers will actually open, and coverage of the languages your families speak, which is where most tools stop.

Our students speak languages the platform may not support. Does that disqualify it?

Check it directly rather than assuming either way. Meta Typing Club runs structured courses in multiple international languages, including Persian, Pashto and Dari, which is unusual in this category. Ask any vendor to demonstrate the lesson sequence in your community's languages, and to show a student progress report in one of them.

If the pilot report in front of you cannot answer test 2 and test 4, the motion to scale is premature, whatever the growth chart says. You can see the lesson structure and per-student reporting a pilot would run on at our help center, and price it through contact.