How the Wattlebird Plan Management page was built

What this page declined to publish

This comes first because on this trade it is the most important part of the build. Everything below was considered, researched, and left off the page on purpose.

Declined — could not be verified at the source during this build

• The plan-management price limit. The NDIS pricing arrangements page links only to a downloadable pricing schedule (PDF and DOCX, effective 1 July 2026). The PDF’s text could not be extracted during this build, so no figure was verified and none is published.
• Any establishment fee. Same reason. Never mentioned on the page.
• What NDIS provider registration requires. The source describes plan management as done by a registered plan manager, but says nothing we could verify about what registration involves — so the page makes no claim about it, and the fictional business claims no registration status. The footer carries a visibly patterned placeholder (0-000-0000) instead.
• Any regional loading or remote-area pricing rule. Declined; the page says only that it works remotely across WA.
• Support-category and line-item codes. Never published. The page says “the right support budget” and stops there.
• Eligibility advice of any kind. Not the business’s job and not this page’s.
• What happens to a budget at the end of a plan. Declined on the page, out loud: the FAQ answers the part we can answer and then says the rest is a question for a planner or the NDIA, and that we are not going to guess at it on a web page.

Verified at ndis.gov.au on 8 September 2026, and stated once

Local SEO we’d implement for the real business

The live page emits only WebPage, CreativeWork, WebSite and BreadcrumbList, because putting a fictional business into a real search index with fictional reviews attached would be dishonest markup. For a real Wattlebird, this is what ships on day one — printed here as a code sample instead.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "FinancialService",
      "@id": "https://wattlebirdplan.com.au/#org",
      "name": "Wattlebird Plan Management",
      "description": "NDIS plan management for participants in Joondalup, Perth's
                      northern corridor and remotely across Western Australia.",
      "url": "https://wattlebirdplan.com.au/",
      "telephone": "+61 8 9000 0433",
      "email": "hello@wattlebirdplan.com.au",
      "priceRange": "Funded through the participant's plan",
      "address": {
        "@type": "PostalAddress",
        "streetAddress": "<the client's real suite>",
        "addressLocality": "Joondalup",
        "addressRegion": "WA",
        "postalCode": "6027",
        "addressCountry": "AU"
      },
      "geo": { "@type": "GeoCoordinates", "latitude": -31.744, "longitude": 115.766 },
      "openingHoursSpecification": [{
        "@type": "OpeningHoursSpecification",
        "dayOfWeek": ["Monday","Tuesday","Wednesday","Thursday","Friday"],
        "opens": "08:30", "closes": "16:30"
      }],
      "areaServed": [
        { "@type": "City", "name": "Joondalup" },
        { "@type": "City", "name": "Wanneroo" },
        { "@type": "City", "name": "Clarkson" },
        { "@type": "State", "name": "Western Australia" }
      ],
      "sameAs": [
        "https://www.google.com/maps/place/?q=place_id:<GBP place id>",
        "https://www.facebook.com/<page>",
        "https://www.linkedin.com/company/<page>"
      ],
      "aggregateRating": {
        "@type": "AggregateRating",
        "ratingValue": "4.9", "reviewCount": "68"
      }
    },
    {
      "@type": "Service",
      "name": "NDIS plan management",
      "provider": { "@id": "https://wattlebirdplan.com.au/#org" },
      "serviceType": "NDIS plan management",
      "areaServed": { "@type": "State", "name": "Western Australia" }
    },
    {
      "@type": "FAQPage",
      "mainEntity": [{
        "@type": "Question",
        "name": "Can I move to you in the middle of a plan?",
        "acceptedAnswer": {
          "@type": "Answer",
          "text": "People move to us mid-plan most weeks. You tell us, we do our part
                   of the paperwork, and from then on your providers send their
                   invoices to us instead."
        }
      }]
    }
  ]
}
</script>

And the rest of it, which is not markup

• Google Business Profile as the primary asset for this trade: category Disability services / Financial consultant, the Joondalup service area listed suburb by suburb, and the review flow wired to a card the office hands over after the first statement.
• A real review feed replacing the demonstration copy on this page, pulled from the profile so the markup above is honest rather than typed.
• One page per intent, not one page per suburb: “switching plan managers mid-plan”, “plan managed vs self managed”, “how a plan manager pays a provider”. Doorway suburb pages get a plan manager penalised and do not convert anyway.
• The areaServed list and the on-page suburb list kept identical, because the two disagreeing is the most common local-SEO defect we find on trade sites.
• A statement-day content beat: one short post on the first business day of each month. It matches what the business already does, so it never becomes a chore nobody does.

Why this converts

Five sentences for the person deciding whether to hire us, not for a designer.

1. The headline answers the only question this audience actually has — when does my provider get paid — in nine words, and the phone number sits next to it as a button rather than as small print in a corner.
2. The proof element asks for two things a tired parent already knows off the top of their head, invoices a month and providers on the plan, and answers in their currency — minutes and hours — because a page about NDIS money that refuses to print NDIS money still has to prove something.
3. The empty price column is the highest-converting element on the page: every competitor either publishes a fee the NDIA actually sets or hides behind “contact us”, and saying plainly we do not set this and we do not add to it, so we have nothing to quote you buys more trust in one sentence than a discount would.
4. At one or two invoices a month the card turns the work away on screen — twelve minutes, keep self-managing, ring us when you add a second provider — and the button underneath ticks a box on the form instead of losing the lead, which is the difference between an honest page and a page that just looks honest.
5. Every conversion mechanic a real enquiry needs is present and driven in the browser before it ships: click-to-call in the masthead and in a sticky bar, a six-field form that validates and confirms locally with nothing faked, reviews written the way a parent and a support coordinator actually speak, a service-area map with named suburbs, and a page that holds together at 375 pixels wide with the headline, the call button and the trust line all above the fold.

The proof element

One slider, one stepper, one drawing, one refusal.

What it computes

There is exactly one number in the model and it belongs to the business, not to the scheme: twelve minutes an invoice — check it against the plan, claim it, file it. The card says on its face that the twelve minutes is our estimate and invites the reader to correct it. Minutes are held as integers and divided once at display (the W47 rule), so a month reads 216 minutes and prints 3.6 hours rather than 3.5999999999999996.

Below three invoices the card prints minutes rather than hours, because that is the reading the oracle licenses at one invoice and because “0.4 hours” is a silly way to tell somebody their month is fine.

The drawing

Thirty invoice slots are drawn once, synchronously, as dashed ghosts; your invoices are a second set of filled slips laid over them, so the ceiling of the scale is visible without being explained. Underneath, a fixed nought-to-six-hour axis — six hours is exactly what thirty invoices costs, so the axis is the reachable maximum rather than an arbitrary frame, and the bar can never run off its own scale. The whole strip is laid out in real CSS pixels: the SVG’s viewBox width tracks its rendered width, so one user unit is one device-independent pixel and the axis type is ten pixels at 375 as well as at 1440. Below 430 pixels the grid re-flows from ten slips a row to six.

Slip visibility is toggled with the hidden attribute via setAttribute / removeAttribute, never the IDL property — hidden is an HTMLElement property and does nothing at all on an SVGElement, which cost a sibling site a whole feature last wave with every state assert green. The [hidden]{display:none!important} guard ships on both pages so the attribute path actually hides.

The three-step payment strip

Provider sends the invoice → we check it against the plan budget → paid within three business days. That promise is ours to make, so it is sayable, and it is the single thing this audience wants to know. It is built in HTML rather than SVG (see Pass 1) with a dot travelling the connector on the wall clock — a six-second cycle that holds for half a second on step two, because the check is the part being sold. It never drops below full opacity, so the seven-second thumbnail catches it alive at any instant, and reduced motion parks it on step two as a settled frame rather than removing it.

The refusal

At one or two invoices a month the card turns apricot and says “12 minutes a month is not worth changing how your plan is managed. Keep self-managing. Ring us when you add a second provider.” The figure in that sentence is computed, so at two invoices it correctly reads 24. A button in the panel ticks the form’s “not yet — ring me when my plan grows” checkbox and moves the reader to the form, and the confirmation changes wording to match. A marker under the slider states the same floor before the reader drags down to find it.

There is a second give-away at the other end: from twenty invoices a month the card suggests asking whether a couple of those providers will bill monthly instead of per visit, because fewer invoices is a better fix than a faster plan manager — advice that costs us the exact thing we are selling.

Palette

Every disability and aged-care site in Perth is mid-blue and mint-teal on white. This one is warm paper and ink-indigo with an apricot chip: administrative calm rather than clinical brightness, and deliberately unlike the homecare, ndis and supportcoordination pages already in the collection. Ratios below are quoted from contrast-w55.log, hex and figure from the same row of the same run.

With both minted tokens in it the sweep declares 26 pairs for this site and fails none, and the wave-wide run is 0 failures. Two forbidden rows changed decisions on this page rather than merely being recorded: paperOnDark on apricot at 1.76:1 is why the chip and the sticky call button always take ink, and burnt on band at 1.86:1 is why the focus ring changes token on every dark surface (Pass 3).

Type

Bricolage Grotesque at 600 for headings: it has warmth without being soft, which is the register this audience needs from a business that handles their money. Public Sans for body, set at 17px on mobile rather than the usual 16, because this reader actually reads the page rather than scanning it. One typographic move only — the three management choices set as a numbered 01 / 02 / 03 row in display weight with the numerals in apricot, so the page’s single moment of scale is spent on the fact that the choice is the reader’s and plan management is only one of three.

The three passes

Each one rendered the page in headless Chrome and drove the controls — the slider across all thirty positions, the providers stepper to both clamps, the form empty, with junk in the phone field and filled, the refusal button, and the sticky bar. The harness finished at 135 of 135 checks across both pages.

What would change for a real client

The page is the easy half. For a working plan manager the integrations are the job.

• Xero — the bureau’s own ledger and the payment run. Invoices arrive, get coded to the right participant and support budget, and go out as a batch; the monthly statement is generated from the same data rather than typed twice.
• Careview — where a lot of WA providers raise their invoices and manage rosters. Taking the invoice feed directly is what makes a three-business-day promise safe to publish rather than aspirational.
• Brevity — the same story for providers running rostering and claims through it; the value is that nobody re-keys a line item.
• Lumary — the Salesforce-based end for larger providers, where an integration is worth building because one connection covers a lot of monthly volume.
• The bank — an ABA batch file rather than a person typing BSBs, because the whole promise on this page is a clock.
• The form — wired to whatever the office already opens in the morning. On a page like this the enquiry is a phone call waiting to happen; it should land as a task with a name and a number, not as an email nobody owns.

Everything visual on both pages is procedural — SVG, CSS and a little canvas-free JavaScript. There is not a single raster image, no library, no build step and no framework: two self-contained files under 120 KB between them.