Ruby Castellanos
The reason demos happen on Fridays. Can read a roadmap like a weather map.
- Role
- Producer
- Based
- Sydney
- In the journal
- 28 articles
Pre-orders are a trust exercise, not a checkout variant
Pre-orders and crowdfund launches fail on expectation design, not payment plumbing. Charge timing, delay scripts and the trust ledger that carries a launch.
Human-in-the-loop: designing the review queue
Human-in-the-loop AI fails at the queue, not the model. Triage design, keyboard-first reviewer ergonomics, feedback loops into evals, and honest staffing maths.
Prompt libraries that survive the team
A well-structured prompt library still dies if nobody owns it. The team side: ownership, review rituals, discoverability, drift control and reuse metrics.
Status updates clients actually read
The one-screen weekly note format that kills the 40-slide deck: shipped, next, blocked, decisions needed — with working demos instead of jargon and metrics in context.
Article indexes people actually browse
Most blog indexes are a reverse-chronological conveyor belt. Here's how we design journal indexes that invite real browsing — lead stories, rhythm, and honest pagination.
The first two weeks with an agency: a client's field guide
What actually happens in the first fortnight of an agency engagement: access checklists, stakeholder mapping, decision calendars, and how week-one demos become possible.
Reading agency contracts: IP, liability and the licence traps
A plain-English tour of the contract clauses that matter in creative engagements: IP assignment vs licence, portfolio rights, liability caps, kill fees, warranty windows.
Agent UX: designing for software that acts on your behalf
When software acts instead of suggests, the interface becomes a control system: permission scopes, visible progress, real undo, and calibrated trust.
Careers pages that filter people in, not out
A careers page is a self-selection device, not a résumé bucket. Real work, honest salary bands, humane application flows — and no ping-pong clichés.
Handover without the hangover: documentation your team will read
A clean agency-to-client handover in practice: documentation people actually read, runbooks that run, training that sticks, and who owns the keys afterwards.
Estimating software honestly: ranges, risks and why we cap scope
How we estimate software honestly: signal ranges per phase, explicit risk buffers, fixed-scope sprints — and how to read any estimate a studio gives you.
Build times are a team-health metric
Build time as a team-health metric: profiling CI honestly, cache strategies, test sharding, and the cultural rules that keep pipelines under five minutes.
How to read an agency SOW before you sign it
Most SOW disputes were visible in the document. How to read scope boundaries, assumptions, change control, acceptance criteria and warranties before signing.
Lifecycle email is a product: building programmes, not campaigns
Lifecycle email is a product, not a campaign calendar: state-based programmes, plain-text craft, deliverability engineering, and metrics beyond open rates.
The launch-day QA checklist we run on every site
Launch week is preparation; launch morning is verification. Our day-of QA sweep: redirects, 404s, OG tags, forms, analytics parity, and the 72-hour war room.
Prioritisation frameworks after RICE stops working
RICE works until the spreadsheet becomes the decision. What comes next: strategy filters, evidence maps, learning-first sequencing, meetings that decide.
Subscription models that retain: design for the second order
Subscription retention is decided between order one and order two. Design the second order, save failed payments without resentment, and price subscriptions honestly.
Gift flows: the seasonal feature that deserves year-round craft
Gifting is a revenue line with a calendar. Multi-address checkout, date-promise logic, gift-card UX, the recipient experience, and a peak-season runbook.
The first 30 days: onboarding a client engagement
How the first 30 days of an agency engagement should run: kickoffs that surface real decision-makers, baseline metrics, access week, and artefacts that prevent surprises.
After launch: the first 90 days done properly
Launch day is the start, not the finish. How we run the first 90 days after a site ships: analytics baselines, bug etiquette, and the next engagement.
Product photography that actually sells
Product photography that sells: shot lists per category, a consistent lighting language, scale cues that cut returns, video's real ROI, and the shoot brief.
Error messages that de-escalate
An error message is a conversation at someone's worst moment with your product. The structure, tone rules and support hooks that make errors actually help.
The kickoff workshop agenda we run before every build
Our actual half-day kickoff agenda for a 12-week build: the hour-one decision log, the risk room, role mapping and the 'what would make this fail' round.
How to write a project brief that gets better proposals
A buyer-side playbook for briefing agencies: problem over solution, useful constraints, budget honesty, evaluation criteria, and a template you can steal.
Subscription UX that retains without trapping
Ethical subscription design — skip-and-pause controls, dosage guidance, honest renewal reminders, dignified cancellation — and the LTV case against traps.
The compounding returns of refreshing old content
Old content decays silently. Here's our refresh system: decay detection in Search Console, refresh-vs-rewrite rules, redirect discipline, and a cadence that compounds.
Handovers that don't decay: the living handover playbook
Most agency handovers are a folder of PDFs and a prayer. Here's the playbook we run instead: documentation that gets used, pairing windows, and a 30-60-90 support taper.
Discovery sprints: two weeks that de-risk six months
Our discovery sprint format, published: inputs, the ten days, the decision memo output, who must attend, and when discovery is a stall dressed as diligence.