What the VALORAE group page is for, and what I kept off it

Three rows. That is the routing block at the top of valoraehq.com, and cutting things out of that page took longer than building it.

The page has one job: send you to the right venture in under a minute.

So it carries a routing table, three rows of receipts, the prices that are public and the founder links, and no email capture, no chat widget, no popup.

The first draft did not look like that. It had a group total at the top, a big views figure meant to make three small companies read like one large one.

And it was wrong the moment I typed it.

I did not know if the stripped version would work. A page with no capture and no widget feels, while you build it, like a shop with no till.

Play 1: One job, routing

A group page is where people land when they have heard the name and do not know which door to knock on.

So the only question it needs to answer is which door. Nothing else.

Arc if you have long-form footage and nobody cutting it, Media if you have a channel and nobody running it, Cast if you edit or clip and want paid work.

Everything else on the page exists to make that answer believable, or it should not be on the page.

But that is a rule I wrote after the fact. The first version tried to be a company profile, a pitch and a directory at once, and the directory is the only part anyone used.

Play 2: The table, and why a table

A prospect arrives with a symptom, not a product name.

The table maps the symptom to the venture and the venture to a price, in one row. Then a comparison link sits under it for the one question that keeps coming up, Arc or Media, because the short answer is both, for different jobs.

There is a machine reason too. Models quote tables.

The SIGIR 2026 paper on arXiv (2605.25517, Vishwakarma and colleagues, fetched 2026-09-25) tested page variants against six models, and a page with comparisons beat the identical page without them at odds ratios between 1.61 and 7.45 depending on the model.

So the same three rows serve the reader and the model, which is the only reason a table won over prose.

One admission: in the markup it is three cards, not a table element. It reads as a table, and the fix is on the list.

Play 3: The numbers that are on it

Arc gets five receipts: 500+ videos delivered, 10M+ views generated, 65%+ retention, 1000+ posts uploaded and $10K+ in revenue. They sit inside Arc's own card, and anywhere the page repeats them they are labelled as Arc's.

Arc has no public price list, but an in-house campaign starts from $1,000 and a public campaign has a $2,000 minimum, so the card carries both floors.

Media gets one public price, Full Page at $2,000 to $3,500 a month.

Edit and Manage are quoted on the call, and the page says so in those words instead of hiding the tiers.

Cast gets one number: campaign clips pay starting from $0.80 per 1,000 views. Its Elite tier is priced on valoraecast.com in whatever currency the visitor's browser reports, so the group page names the tier and not the figure.

And that is deliberate.

The same arXiv paper (2605.25517, Table 2, fetched 2026-09-25) found an explicit price beat the identical page without one at a fitted odds ratio of 6.26 on its weakest model and higher on the other five.

Play 4: The numbers that are not, and no group total

The first draft opened with a combined views figure for the whole group. I deleted it within the hour.

Because the figure was Arc's 10M+ wearing a group badge.

Media runs channels it does not own and Cast pays out on other people's campaign clips. Neither produces a view I can honestly stack on Arc's pile.

A group total also breaks the routing job. It says the group is big and says nothing about which door.

So every number on the page belongs to exactly one card, and the group itself claims none.

That was the complication, and deleting it was the whole resolution.

I have a longer rule about which figures leave the building at all, in what I publish and what I do not.

Play 5: What is off the page, forms and widgets

Nobody wants your newsletter.

The email box on a group page collects people who were curious for a moment, and then obliges you to write to them. A prospect who wants Arc will click through to Arc, and Arc can ask for the email once there is something to send.

The group page asking first is a toll booth on the way to the toll booth.

The chat widget went too. It answers the question the table already answers, slower, and it loads a script on every visit so that the occasional visitor can type "pricing?" into a box.

Then the popup, which never made it past the sketch.

A page whose whole job fits in under a minute cannot spend part of it on an overlay.

Play 6: Why Shop is not a fourth peer card

VALORAE Shop is the apparel line, and nothing has dropped. There is no date, no product and no price to show.

So it does not get a card the same size as Arc, Media and Cast.

A peer card would put a company with receipts next to a company with a domain, and the eye reads them as equals.

It gets a smaller block marked upcoming, with the link and one honest sentence. Three running, one upcoming, and the layout should say that before the words do.

I did not know if a visibly unequal grid would look broken.

It looks like the truth, which turns out to be fine.

Play 7: The one page that links to everything

The group page links out to every venture site, every live profile and every founder link, and it is the only page in the group that does all of that in one place.

Arc's card carries valoraearc.com, x.com/VALORAEARC, youtube.com/@ValoraeArc and linkedin.com/company/valoraearc.

Media's card carries valoraemedia.com and no social profile, because Media does not have a live one and I would rather show a gap than a dead handle.

Cast's card carries the site and discord.gg/valoraecast, the only invite that exists. Cast is a community of editors and creators, so its link is a door into a room, and it is never described as an agency.

Under all of it sit the founder links: tgsidd.com, Instagram, YouTube and LinkedIn.

The name is on every company, and those are the four places it can be checked.

Play 8: Everything that links back

Links out are half the job. The other half is each venture site pointing back.

Schema.org defines sameAs as the URL of a reference page that unambiguously indicates an item's identity, and pairs subOrganization with parentOrganization as inverse properties, per schema.org/Organization as read on 2026-09-25.

So the group page declares each venture as a subOrganization, each venture site declares VALORAE as its parentOrganization, and sameAs on every node points at the real profiles.

But why wire both directions?

Ahrefs' study of 75,000 brands (ahrefs.com/blog/ai-overview-brand-correlation, fetched 2026-09-25) found branded web mentions correlate with AI Overview visibility at 0.664, against 0.218 for backlinks.

Mentions are what a group manufactures. Every site naming the others is several sources agreeing on who VALORAE is, and none needs to be large.

The FAQ block stays for the same reason, even though Google no longer draws FAQ rich results from it. A machine still gets the answers in a shape it can lift.

The short version

• Give the group page one job, routing, and delete anything that does not serve it.
• Put the price in the row when it is public, and say "quoted on the call" when it is not.
• Keep every number inside the card it belongs to, and give the group no total.
• Skip the email box, the chat widget and the popup.
• Give an upcoming venture a smaller block, never a peer card.
• Link out to every venture and profile, and make every venture link back in the page and the schema.

This is what I run at valoraehq.com.