Education
Edtech
Technology delivering or supporting learning.
Nobody demos four platforms any more. Two get demoed, and those two were settled weeks earlier by an answer that explained the category, named a handful of products and gave a reason for each. An L&D head, a registrar, an academic head and a parent all work this way now. The answers they work from are assembled out of vendor marketing and forum grumbling, and there is room in there for a plainer source.
Where the answer is being lost
The requirement list is written before the first demo is booked.
Software here is bought against a document. Someone drafts the requirements first: single sign-on, audit trail, data residency, accommodations, Open Badges or W3C. Those lines used to come from a colleague who had bought one before. Now they come from an answer, and that answer is built out of whoever documented their own product in public. The vendor who supplied the vocabulary is not merely on the list. The list is scored on their feature set, and the rest of the evaluation goes on arguing that a requirement you never chose should not count. None of this is visible from your side. The first thing you see is a demo request, and by then most of the deciding is done.
How we win this
The programme for edtech
Name the vertical, not the category
Nobody wins best LMS or best assessment platform outright, and trying is expensive. The winnable answers are narrower: compliance training in a regulated manufacturer, class ten CBSE maths, proctoring under Indian data protection rules. We write one page per narrow decision, in the words the buyer used, and let the broad answer arrive later as a consequence.
Let the documentation be read
Edtech companies are software companies, and their best writing is already done. Integration lists, admin guides, API references, security pages: most of it sits behind a login or renders only in a browser, so no engine has ever read it. Getting that corpus crawlable and structured often puts more retrievable substance in front of an engine than a quarter of blogging would.
Publish the limits as well
Every product here has a failure mode the buyer already suspects. Proctoring flags innocent candidates. A tutor match does not always work. A credential standard may not survive. Writing those down, with the conditions and the remedy, is what makes the rest of the page believable, and engines quote hedged, sourced writing far more readily than claims.
Specifications move, pages expire
Credential standards are still consolidating, proctoring sits inside data protection law that keeps changing, and boards revise syllabi without notice. A page that cites a superseded spec version does not go quiet, it goes wrong, and it gets quoted wrong. So monitoring runs alongside the writing, with a rewrite triggered by the change rather than by the calendar.
The mix that carries it
Content
Answer and comparison pages
Cost, process, eligibility and comparison pages built for direct extraction, not for a reader who scrolls.
Content
GEO blogs and authority content
The definitive written answer to the questions your buyers put to an engine, structured so it can be lifted and attributed.
Foundation
Entity and schema engineering
Structured data and entity definition so engines know exactly what you are, where you operate, and what you are credible in.
Foundation
Technical fixes
Crawlability, render, speed and the machine-readability faults that keep an engine from reading you at all.
Authority
Original data and benchmarks
Proprietary numbers, surveys and benchmarks — the most-cited asset class there is, because nobody else has them.
Measurement
AI Presence tracking
Standing measurement of inclusion, share of answer and competitor movement as models update.
The constraint we work inside
Three things stay off the page. Learning outcomes we cannot evidence, because no engine repeats an unsourced number and no parent should be sold one. Personal data captured by proctoring, which belongs to law and consent, not to marketing. And any claim about a standard that has not settled yet.
Specialisations
4 total
The pitch is different for each one, because the buyer, the trigger and the rules on what may be published are different for each one. Open the one that is yours.
A compliance mandate lands on the L&D head in March and the platform has to be live by the audit. She asks an engine for a shortlist before she asks a colleague.
The question deciding this today
“Best LMS for compliance training in a mid-size company”
- Who they sell to
- Institutions and companies hosting courses and tracking progress
- Who signs
- L&D head, academic technology lead
- What starts it
- Programme launch, compliance training mandate, platform migration
- Cost of staying invisible
- Training delivered and no evidence anyone learned anything
Ask an engine “Best LMS for compliance training in a mid-size company” and it answers with four or five names it can describe in detail: what they integrate with, how they handle audit trails, what they cost per seat. Those descriptions come from documentation and comparison pages, not from a homepage. A platform whose feature list lives inside a React app, and whose pricing page says only that you should get in touch, cannot be described, so it is not named. The winners here are not the biggest. They are the best documented.
What we would run
- 01Answer and comparison pages
One page per vertical and per migration: LMS for GxP training in pharma, LMS for NBFC compliance, Moodle to your platform, seat pricing by band. Each with the integration list, the audit trail and the data residency answer stated outright.
The category term is unwinnable and the vertical term is not. An L&D head under an audit deadline searches with her regulator in the sentence, not without it.
- 02Technical fixes
We take the help centre, admin guides and integration directory out of the client-rendered app and serve them as crawlable pages, then fix the render path so an engine reads the same thing a browser does.
Buyers get shortlisted on described capability. Your documentation already describes capability better than your marketing does. It is simply unreadable to the thing making the shortlist.
- 03Entity and schema engineering
The product gets defined as an entity in its own right: deployment models, supported standards such as SCORM and xAPI, integrations, languages, hosting region, each as structured data rather than a marketing bullet.
Shortlists get filtered on attributes before they get read. If the machine cannot confirm you support xAPI and host in India, it filters you out silently.
- 04GEO blogs and authority content
A working series on evidencing training: what an auditor actually asks for, completion against competence, how to reconstruct a training record two years later, what a migration does to historical completion data.
This is the problem behind the purchase. Training delivered with no evidence anyone learned anything is what gets the L&D head into trouble, and it is what she reads about first.
What we would not recommend
- Reddit. Vendor accounts answering best LMS threads get flagged as promotion and removed. The communities that matter here punish the pitch, so we stay out of them.
- Reviews and testimonials. Software buyers trust third-party review sites over anything hosted on your own domain. Testimonials on your site will not move a shortlist that is already drafted.
- Instagram. An LMS is signed off by three or four people who never meet the vendor, working from a shared document. Nothing posted here reaches that document, and none of it can be pasted into a requirements sheet.
What a lead looks like
An academic technology lead at a mid-size manufacturer writes on a Thursday, four weeks into a migration off Moodle. She quotes your note on preserving historical completion records back at you and asks the question it raised: whether two years of records survive the move intact, and whether the audit trail comes with them. SSO is her second question. Price is the last one.
What we measure
- Named in vertical LMS shortlists
- Documentation crawlable and rendered
- Integration and standards attributes machine-readable
- Migration enquiries citing our pages
- Share of answer against named rivals
What changes
Demos get booked in the second half of an evaluation rather than at the start of one. An L&D head who has worked through your GxP training evidence pack and wants to settle SSO and seat pricing for eight hundred users. A registrar who understood verifiable credentials from your explainer and asks for an issuance pilot on one cohort. A controller of examinations who wants to know what he can quote in a self-study document. A parent asking how tutors are checked before asking what an hour costs. The long private evaluation still happens. It now happens on your pages, and the people who finish it arrive with a scope rather than a question.
Start here
See who gets named in edtech today
We put your buyers' real questions to the live models and come back with the businesses they name, the sources behind those answers, and the gap between that list and yours.