A 100-Point RFP Rubric for Early-Grades Typing Tools

Score early-grades typing platforms on a weighted 100-point rubric: curriculum structure 20, teacher workflow 20, progress and adoption evidence 15, data practices and exit terms 15, language coverage 10, rostering and accessibility 10, cost per active student 10. Use it as a committee starting draft. Meta Typing Club, with more than 2,000 structured lessons, built its tools against these criteria.

TL;DR: Typing vendors all demo well and price low, so an unweighted evaluation quietly becomes a price decision. This rubric puts instruction and workflow ahead of cost and states how to verify each answer before the award.

A published RFP shows where the points actually go

Rio Rancho Public Schools scored its 2024 solicitation, Software Related to Instructional Products (RFP No. 2025-010-C&I), out of 100 shortlist points: experience and references 15, service and support and training 20, product features and technical specifications 25, and cost or price 40.

Price was the single largest line at 40 of 100, though the three quality categories total 60 together. That does not neutralize the 40: when quality scores cluster close, as they do among vendors who all demo well, the largest line decides. Rio Rancho is not a bad example; a general software solicitation stays general by design. Nobody scored whether a seven-year-old can find the home row in April. An evaluation that does not weight instruction will not select for instruction.

The rubric, a committee starting draft

Seven criteria, 100 points, one page. Each row states what earns full points and how an evaluator confirms it without taking the vendor's word for it. Route it past your business office and legal counsel before it enters a solicitation.

Criterion	Points	Full points require	How to verify
Curriculum structure	20	A published sequence from single keys to connected text, with lesson counts per stage	Open three consecutive lessons in a trial account
Teacher workflow	20	Class creation, invite-code enrollment, and assignments with due dates, all teacher-run	Time a teacher setting up one class and one assignment
Progress and adoption evidence	15	Speed and accuracy per student over time, plus the share still active at 90 days, backed by exportable logs	Request the session-level export or a live analytics login; a summary figure alone earns half credit
Data practices and exit terms	15	A signed privacy agreement and a written return-and-delete clause with a timeline	Require the executed agreement as an attachment
Language coverage	10	Structured courses, not translated menus, in your students' home languages	Check the vendor's list against enrollment data
Rostering and accessibility	10	Roster sync via a standard such as OneRoster or district SSO, plus a completed accessibility self-assessment	Ask for their 1EdTech accessibility rubric response or VPAT, and test SSO
Cost per active student	10	Price divided by students the vendor projects still practicing at 90 days	Recompute it from their own adoption number

Score each row 0, half, or full. Half means the vendor answered but could not evidence it. A claim nobody can verify scores zero, not partial credit.

Where these criteria came from

Several rows lean on work already done by bodies with more standing than a local committee. SETDA released its EdTech Quality Indicators procurement guide on February 12, 2025, sorting quality into five indicators, Safe, Evidence-Based, Inclusive, Usable, and Interoperable. The Student Data Privacy Consortium's National Data Privacy Agreement (version 2.2, November 19, 2025) is a model contract from 28 state alliances. 1EdTech's TrustEd Apps program keeps open rubrics for privacy, security, and generative AI data, and its accessibility rubric is the source for the rostering and accessibility row.

All three are dated 2025, so check each link for a newer version before citing it in a 2026 solicitation. Treat published quality frameworks as evidence a vendor hands you, not homework for your committee.

Price still has to count

The strongest objection to a 10-point price weight is procedural, and it is fair. Many state codes require cost as a stated, meaningful factor, and a committee scoring price at 10 against a bid half the cost invites a protest it may lose. That objection wins on the law. It loses on the substance.

Hold price at whatever floor policy sets, score cost per active student rather than cost per license, and take the points you cannot take from price out of the experience narrative. A platform nobody uses is not cheap. It is a write-off with a small invoice.

Score us with it too

We built Meta Typing Club's classes, invite codes, homework with due dates, and per-language speed and accuracy tracking because districts score those criteria. Structured courses in multiple international languages, including Pashto and Dari, make language coverage a strength for us. We are a typing-lessons platform, not an assessment vendor or a compliance authority, so make us answer the data and accessibility rows in writing like any bidder.

Key takeaways

• Sequence the scoring: settle curriculum and workflow first, since price and legal review come after.
• Self-reported adoption numbers with no underlying logs are the most common inflated score on this rubric.
• A poor curriculum or workflow score explains stalled adoption better than any satisfaction survey.
• This is a committee starting draft, not a finished solicitation; route it through legal review.
• A cheap platform nobody uses is not a savings. Score cost per active student, not cost per license.

Questions procurement officers ask about this rubric

Can our committee score this without running a pilot first?

Yes, for five of the seven rows: curriculum structure, teacher workflow, data practices and exit terms, language coverage, and rostering and accessibility all score from documents and a demo account. Progress and adoption evidence stays provisional until your own trial produces a number, and cost per active student with it, since it rests on the same figure.

How do we weight price if our rules require cost as a factor?

Set price at the floor your policy requires, then take the remaining points from experience and references rather than from instruction. Firm longevity tells you less about a K-5 rollout than a teacher finishing class setup unaided.

What evidence should we accept from a vendor with no efficacy study?

Most have none, and demanding one pushes vendors toward dressed-up claims. Ask for first-party usage data instead: how many students, over what period, what share active at 90 days, backed by logs rather than a slide. If the purchase draws on Title I or IDEA funds, confirm with your business office whether that source requires ESSA evidence-tier documentation.

Who should score the language coverage row?

Not the technology office alone. Pair that evaluator with whoever owns multilingual or newcomer services, since they can tell a structured course in a language from a translated menu over an English key sequence.

What should the contract say about our data if we do not renew?

Name the format, the recipient, and the clock: progress records in a machine-readable file within 30 days of termination, confirmed deletion within 60. It is also a fast read on how a vendor behaves once it no longer holds the stronger hand.

Putting it in the file

Take the seven rows as a starting draft, route them through policy and legal review, set the points to what your rules allow, and hand evaluators the verification column instead of a blank comment box. Then score the platform you already run: a low curriculum or workflow score explains the stalled adoption you suspected, in language a board can read. Score Meta Typing Club's lessons yourself, check what it costs, which for us is currently nothing, and read our other writing for school leaders.