Design notes: what changed, why it should convert, and what it does not touch
Written for Rob. Internal only.
What this is
Five pages from investmentgrade.com rebuilt with the same content and the same numbers in a different presentation. Nothing on the main site was changed. This subdomain is marked noindex, sits in no sitemap or menu, and its forms do not send anything.
- Home
- Starbucks credit rating and cap rate (tenant page)
- Starbucks NNN for sale (tenant for-sale page)
- Illinois NNN properties (state page)
- Cook County, Illinois (county page)
Two switches sit in the gold bar at the top of every page: Look (A Vault or B Ledger) and Form (Guided or One page). Your choice follows you from page to page.
The idea in one sentence
The brand owns one concept, the line at BBB‑/Baa3, so the design makes that line the signature: the gold rule under the home headline, the rating ladder with real tenants on its rungs, the tenant marker on each credit page, and the gold reference line in every chart.
Type follows the same thought. Headlines and large numbers use Bodoni Moda, the engraved style of bond certificates. Text and data use Hanken Grotesk, which stays readable at small sizes on a phone. Both are served from your own server.
Two looks
| Look | What it is | Strengths | Weaknesses |
|---|---|---|---|
| A Vault | Deep aubergine with your existing gold. Closest to the current brand. | Feels private-bank and confident. Photos and gold carry more weight. Charts read as lit instruments. Smallest brand change. | Long articles are harder to read on a dark page, more so in daylight and for older eyes. Dense tables feel heavier. |
| B Ledger | Cool lilac-white paper, aubergine ink, gold as an accent. Heroes stay dark. | Best for long research pages, tables and printing. Reads as a research publication, which suits the Investment Grade Research byline. | Less drama. Gold has to be darkened for text contrast, so it loses some richness. Further from the current site. |
My recommendation: A Vault as the default, because it keeps the brand and gives the rich feel you asked for. B is worth testing on the longest research pages only if time on page matters more there than first impression.
The capture system
Today the for-sale pages show a wide form with eight to ten fields at once. Here there is one flow, opened from four places:
- The four choices under the home headline.
- The cap rate floor slider under each dot chart. The button counts the listings that clear the visitor's floor.
- The buyer and owner panels on every page.
- Every "Get full property data" link in a listing table.
| Form style | Strengths | Weaknesses |
|---|---|---|
| Guided One question per screen, large type, tap or press a letter. | Each step is easy. Questions change with the answer, so a seller never sees buyer questions. Works well on a phone. Partial answers can be kept even if the visitor stops. | More taps in total. A visitor who wants to see everything first can feel led. Needs step-by-step measurement to tune. |
| One page The same questions stacked, in your current wide style. | Everything is visible up front. Fast for a repeat visitor or a broker who knows what they want. | Looks like more work at first glance. Every visitor sees the full list of questions. |
My recommendation: Guided on a phone and for first-time visitors. That is a hypothesis to test, not a result.
Why these triggers should work
| Trigger | Where you see it | Why it works | How it differs from today |
|---|---|---|---|
| A small first yes | The four choices in the home hero and the first step of the flow. | People who take a small step are more likely to take the next one (the foot-in-the-door effect, Freedman and Fraser, 1966). | Today the first thing asked for is an email address. |
| One decision at a time | Guided form. | Decision time grows with the number of choices shown together (Hick's law). | Today eight to ten fields appear at once. |
| Value before the ask | The floor slider counts matching listings before any contact field appears. The button reads "Send me these 42 listings". | Reciprocity. The visitor has already received something specific to them. | Today the form asks first and gives later. |
| Their own criteria | The visitor sets the floor and the price range. | People value what they helped build (the IKEA effect, Norton, Mochon and Ariely, 2012) and tend to finish what they started. | Today the criteria are dropdowns inside the form, not the reason for it. |
| Visible progress | The gold bar at the top of the flow starts part filled after the first tap. | A head start makes people more likely to finish (the endowed progress effect, Nunes and Dreze, 2006). | Today there is no sense of progress. |
| Real deadlines, not invented urgency | The 1031 step works out the visitor's own day 45 and day 180 from their closing date. | Losses weigh more than gains (Kahneman and Tversky, 1979), and this deadline is real and theirs. | Today the 1031 status is a checkbox. There are no countdown timers or made-up scarcity anywhere, which matters for compliance and for the kind of buyer you want. |
| Specific, dated numbers | Every chart carries its date and source in the caption. | Precision and visible sourcing read as authority. | The figures are the same as today. They are now the first thing seen rather than a paragraph to read. |
| A low-risk last step | Email is the only required field. Phone is optional and says why. The privacy line sits beside the button. | Removes the two common reasons to stop at the last step. | Close to today, with fewer required fields. |
These are established principles, not a promised lift. Nothing has been measured yet. The honest next step is an A/B test on two or three live pages with one metric: captures per visitor.
What protects SEO, AEO and GEO
- Every word, number, table and FAQ is in the HTML the server sends. Nothing needs JavaScript to be read. Charts are drawn on the server and animation only decorates them.
- Every chart has its figures in text too: direct labels, a caption with date and source, and the original tables kept on the page.
- Same copy, same internal links, same FAQ content. No URL or slug changes.
- Light pages: one stylesheet, one small script, three font files, no chart library. Motion is turned off for visitors who ask their device to reduce it.
Four things need a decision before any of this goes near the main site:
- The home H1 changes from "Investment Grade" to the full sentence that begins with "Investment Grade". The keyword stays first, but it is a change to the most important tag on the site.
- Home FAQ answers sit in tap-to-open rows. They are in the HTML and indexable, but it is a change worth watching.
- Schema (FAQPage, FinancialService) and Rank Math output are not carried into this sandbox. A real rollout keeps them exactly as they are.
- Page speed has to be measured with the final hero photos before launch.
Phase two: the agent mirror
Humans get this design. Agents get a plain twin of every page built from the same data functions, so the two can never disagree. Not built here, by design.
- A Markdown version of each page (facts, tables, dated sources) linked from the page as an alternate.
- A small data file per tenant, state and county with the figures behind each chart and their as-of dates.
- An index file that lists the Markdown versions, and a server rule so an agent that asks for Markdown receives it.
- A hand-off path, so an agent can cite the page and route a person who has agreed to be contacted.
What is real and what is not
- Real: every number, chart, table and paragraph of page copy comes from the live site's own data functions and published page content at the time of the build.
- Not real: forms do not send. "Get full property data" opens the form instead of the listing. Hero photos are stand-ins from your own media libraries, converted to black and white. There is no schema and no analytics.
- New copy: chart titles, captions, form questions and the 1031 deadline note were written for this sandbox. None of it is live, and none of it has been through the independent second check that the accuracy rule requires before publishing.
Decisions for you
- Look: A Vault or B Ledger.
- Form: Guided or One page.
- Where to pilot on the live site first. My suggestion is the tenant for-sale pages, because the buying intent is already there.