Maritime & Safety Training NSW has taught accredited maritime tickets out of Newcastle since 1985. Their new site is on staging, mid-build, with no go-live date locked.
The enrolment process was manual. Students rang, staff took the details down and keyed them in. The client's own description of it on 10 July 2026 was that half their problem was “receiving phone calls and wanting to enrol and it's a real manual, convoluted sort of process”. Every hour spent re-typing a form is an hour nobody spends teaching.
The old site did not help. The heading hierarchy is broken, with an H2 sitting where the H1 belongs, and it is still visible in the page source of the live site today. Navigation is dated and hard. Uptime has flapped four times since May 2026. One page, the recreational boat licence page, carried roughly 58% of all site traffic, which is a single point of failure dressed up as a good month.
Then the brief changed. Part-way through the build, sign-off on the client side moved to a different person, and the business decided to go commercial-only: recreational boat licensing and jet ski licensing dropped entirely, every accredited maritime ticket and first aid kept. That was their commercial call, made for their own reasons, not a recommendation from us. It made a chunk of our earlier keyword work and the two-audience architecture we had built redundant inside a fortnight.
So this page is not a results page and it is not a launch announcement. It is an engagement in progress, published while the work is still being done, with the unfinished parts named.
Sales, design, build, a re-scope, and a rebuild of the enrolment architecture on a live screen-share. Every date below is on file, and the last one has not happened yet.
Discovery ran with the owner-side contact and the enrolments lead. A free homepage mockup went up on the first call alongside a live competitor comparison showing a direct rival at roughly double their organic traffic. Deposit invoiced and paid on the second call.
Two sessions locked the information architecture around how the two audiences actually behave: recreational buyers self-serve fast, commercial buyers read, then call, and need career pathway detail before they commit to a ticket.
The client wanted a cart. We put up two options and recommended the simpler one: Book Now buttons linking to enrolment forms hosted in their student management system, rather than a WooCommerce checkout syncing over an API. The argument that carried it was a multi-course checkout we had built for another client that had not been used once in six months.
Recreational and jet ski pages pulled, every commercial course kept, placeholder imagery called out for replacement, and online against face-to-face delivery made explicit on each course page because it was a real source of customer confusion.
Four rounds of staging feedback worked through and returned, each ahead of the client's own deadline. On 17 July three new build requests arrived inside nine minutes and were confirmed in writing 41 minutes later.
We demonstrated the enrolment flow live: click Find Courses, land on the booking calendar. Except there were no courses in it. Five of their courses are self-paced, so a date picker will never have anything to show. The enrolment architecture was rebuilt on the call into two paths.
The next call trains the client on running the site themselves, with go-live shortly after. The migration checklist behind that date, redirect plan, crawl and export, staging noindex, schema and metadata, is dated 1 August and still open.
Delivered: a full redesign in Figma, a WordPress staging build from those designs, a course-led structure organised around career pathways and accredited qualification names, a re-scope from two audiences to one inside two weeks, approval from the client to proceed to full development on 10 July 2026, four rounds of feedback returned early, and two enrolment paths designed against how their courses actually run.
Not delivered: the launch, the enrolment mechanism for the five self-paced courses, the migration checklist, and three policy pages the regulator requires on display that the client sent through on 26 July. Training is booked for 7 August with go-live after it, and that date has already moved several times.
There are no performance results and none can be produced. No Search Console property has ever been connected for this domain and no analytics property is documented, so there is no first-party baseline and no scoreboard. We are not going to claim traffic, rankings or enrolments off third-party estimates, especially when the pages that carried most of the old site's traffic are being deliberately deleted. Expect a real, temporary drop at launch for that reason.
Rockhampton. A sitemap from 2025 was still telling Google where the old site lived.
Perth. Seventy pages to seventeen, and a menu built to argue problems before answers.
Parallel URL structures had the site bidding against itself on its own best categories.
Grab a free 30 minute call with Tommy or Paddy. We will pull your Search Console live on the call and tell you what is actually holding you back.
Book a free intro call →