Platform Engineering — Health & Wellness Coaching
A five-minute Health Index, adaptive programs and a 900-item content library turn one platform into a different journey for every member — built for retention, not prescription.
Project Overview
DreamBodyClub set out to help women rebuild their health around real life — work, family, and everything in between. The market offered two extremes: rigid programs that ignore individual circumstances, or content libraries so large they paralyse. Our brief was a platform that meets each woman where she is and moves at her pace. That meant personalisation at every interaction, not just at onboarding.
Exploration & Discovery
We interviewed women about why previous fitness platforms hadn’t stuck. The pattern was consistent: programs assumed time, equipment and energy that real life rarely provides. A missed week meant falling behind a fixed schedule, and falling behind meant quitting. That finding made adaptability — not content volume — the design priority.
Auditing the existing content base surfaced hundreds of high-quality workouts and recipes that members never found. Search assumed members knew what to look for; browsing assumed they had time to look. We concluded every item needed structured tags — duration, equipment, intensity, dietary needs — so the platform could do the finding. That decision shaped the entire data model.
You guys absolutely freaking nailed it for us. You guys are the best, and I want to keep working with you forever. Thank you!
Movement, nutrition, mindset and lifestyle data existed in separate silos with no shared score. Testing a combined four-dimension index with prospective members changed the conversation immediately: a single number gave them a starting point and a direction. It became the front door of the platform — a five-minute assessment that drives every recommendation afterwards.
Process & Engineering
Early wireframes treated assessment as a questionnaire, and test users abandoned it. We rebuilt onboarding as a conversational flow that scores the four dimensions as it goes, and completion jumped. The score-first design gives members an immediate reason to return, and the same pattern now drives every program recommendation.
Workouts and recipes lived in different structures, which would have meant two filtering systems. We designed a single tagging model — duration, intensity, equipment, dietary needs, body focus — applied across both libraries. The trade-off was a heavier upfront cataloguing effort; the payoff is one recommendation engine serving every content type. New content now slots in without new code.
Personalisation usually costs latency — every page has to ask a database who you are. We run the logic on Cloudflare’s edge with member data partitioned in D1, so personalised pages render as fast as static ones. Stripe handles flexible billing, Stream serves workout video, and Claude AI drafts coaching prompts from Health Index movement. The stack stays small enough for a small team to own.
The Final Solution
The finished platform opens with the Health Index — five minutes, four dimensions, one score — and everything after it is shaped by the result. Programs adapt to available time and fitness level rather than forcing a fixed schedule, and the workout and recipe libraries surface what fits her constraints instead of asking her to search.
Cohort-based community replaces performative feeds with structured accountability, and the PWA keeps workouts available offline wherever she trains. Every completed workout updates the Index, and every Index update sharpens the next recommendation — personalisation stopped being a feature and became the operating model.
Quantifiable Impact
workouts, each tagged for duration, equipment, intensity and body focus — findable in seconds, not scrolls.
recipes with full nutrition data, matched to dietary needs, prep time and cooking skill.
wellness dimensions scored in a five-minute assessment that drives every recommendation on the platform.
of member journeys are unique — no two members follow the same program.
Retrospective
Not everything survived contact with members. Early gamification — points and leaderboards — felt performative and was cut in favour of cohort accountability. The unified tagging model cost weeks of cataloguing we hadn’t planned for. And we underestimated how much the assessment, not the content, would carry the product. The lesson: in wellness, the feeling of being understood beats the size of the library.
Personalised programs, not generic advice. Tailored to how you actually live.
OH MY GOODNESS!!! It’s a million times better!!! The new site looks amazing so far! Things have been going really well and we’re still absolutely loving the site. Not having to worry about tech has freed us up so much to actually figure out our funnel.
You guys absolutely freaking nailed it for us. You guys are the best, and I want to keep working with you forever. Thank you!
The results I got from Motivation proved that nothing is impossible when you are partnered with the right company.
Working with Motivation has been outstanding. They delivered exactly what was promised, on time, and with quality that exceeded our expectations.
★★★★★
You guys absolutely freaking nailed it for us. You guys are the best, and I want to keep working with you forever. Thank you!
★★★★★
The results I got from Motivation proved that nothing is impossible when you are partnered with the right company.
★★★★★
Working with Motivation has been outstanding. They delivered exactly what was promised, on time, and with quality that exceeded our expectations.
We design and build personalised health, fitness, and wellness platforms — from assessment engines to content libraries.
From diagnosis to daily habit — engineered to fit real lives.