Every idea captured, evaluated, and ranked. Four dimensions: the idea's quality, its credibility (evidence / precedent), the time it requires, and the effort to build. Not a wishlist — a structured pipeline.
How ideas are scored
Every idea is scored across four dimensions. Scores are 1–5. High credibility + low effort = prioritise. High credibility + high effort = plan carefully. Low credibility + any effort = validate first.
Ideas pipeline
Click any idea to expand the full evaluation. Ideas are grouped by category with jump links below. Add new ideas by appending to this document.
Replace all word-based and number-based emotion scales with five bespoke illustrated animals that each embody a regulation state. Critically, these are not just "happy/sad" — they map to the polyvagal regulation states Bloom's AI already tracks. The same five animals used by a 3-year-old in the Early Years companion are the ones a 16-year-old uses. Continuity across the entire product journey.
Every child's Bloom companion begins as a seed. Over time — through attending sessions, trying things, coming back after difficult days — it grows. It never wilts when the child is absent. It never grows faster when they answer more correctly. It grows because they are here and trying. The creature is the product's emotional centre and its most powerful differentiator from every gamified EdTech on the market.
The growth journal is not a portfolio and not a certificate collection. It is a living record of the child's journey — moments, discoveries, things they tried, things they found hard, things they loved. Chronological. Calm. Theirs. The child sees themselves accumulating experience and capability over time — not performing for an audience.
The case for keeping the community internal: Bloom controls the safeguarding, the moderation tone, and the thread back to the product experience. Parents who find community in Bloom stay in Bloom. External communities (Facebook groups, Reddit) are valuable for awareness but create no retention and risk Bloom's reputation if something goes wrong there.
Tonies (Germany) and Yoto (UK) are audio content platforms for children that deliver content through physical figurines or card systems. Both have established trust with families of young children, work offline, and are already used by many neurodivergent children who prefer audio. A Bloom collaboration could deliver Early Years parent guidance podcasts, narrated play activity guides, and pre-school learning stories through these channels.
An advisory panel of 5–7 specialists who review the product, validate the curriculum and AI architecture, and lend their credibility to Bloom publicly. This is standard practice for credible health and EdTech — and for Bloom it is non-optional. Parents of vulnerable children will not trust a product without clinical and educational credibility behind it.
Made By Dyslexia (founded by Kate Griggs, with Sir Richard Branson as a champion), ADHD UK, and Ambitious About Autism each have established parent communities, credible research partnerships, and strong brand recognition. Co-creating content with them — or simply getting their validation stamp on Bloom's condition-specific curriculum — would significantly accelerate parent trust.
The EHCP process in England is deliberately exhausting. Most parents don't know their rights. The generic guidance online is useless. Bloom will build the definitive resource — written by parents who have been through it, reviewed by SEN lawyers, and updated every six months. Not corporate. Not clinical. Real.
Time is one of the most concrete areas where neurodivergent children struggle — and one of the least well-taught. ADHD affects time perception directly. Autism frequently involves difficulty with time transitions and duration estimation. Anxiety compresses and distorts time experience. Yet curricula teach telling the time in short steps across Years 1 to 3. Bloom treats it as a thread running throughout the maths curriculum — from concrete manipulation at age 6 to executive function coaching at 14.
Safety checks must be structural, not aspirational. It is too easy to deprioritise them when things are busy. The monthly protocol locks this into governance — named people, specific checklists, documented outcomes, and a standing agenda item at board level. Over-engineered by design: the cost of missing a check is too high.
No EdTech platform serves the 0–6 window. Parents in this stage are the most desperate, least supported, and most likely to become lifelong advocates for any product that genuinely helps them. The Early Years companion is not AI tutoring — it is parent guidance, diagnosis navigation, play-based learning support, and school transition preparation. The child arrives at Bloom's main platform at 6 having already grown up with it.
The US is not a better market than the UK — it is a different market with different strengths. IDEA's legal framework gives disabled children enforceable rights. North America is the largest regional EdTech market. The architecture, pedagogy, and experience design transfer entirely. Localisation is bounded and plannable. Low scores on time and effort reflect that this is a Phase 4 decision — not that it's unimportant.