How to build a site list that matches the portfolio you actually run

The short answer

Make a list of the public sites you operate, remove the ones you are not willing to act on, and add the rest to DashCanopy as separate tiles. Match the count to a plan cap before you start: 5 on Starter, 15 on Pro, unlimited on Agency. Update the list the day you launch, move, or retire a host. The morning check is only as complete as this list. A beautiful dashboard that is missing the store will report a calm morning while the store is down.

This is an inventory, not a branding exercise. Hostnames belong on it. Taglines do not.

Start from what can break in public

Open a blank note and write every hostname a customer, reader, or client can type. Use the canonical form, the one you print on a business card or a footer. Group them by the property they belong to, then flatten the group into lines. A publisher’s flagship, its store, and a niche site are three lines. A software product’s marketing site, app, docs, and status page are four, which the SaaS split explains. An agency writes lines per client host, not per client logo, as in the agency dashboard.

Include the boring ones. Campaign subdomains, a help center on a different host, an old blog you still rank for, a checkout on a subdomain. These are the URLs that expire on a quiet Sunday because nobody put them in the same list as the homepage. If a certificate or a DNS record can lapse without your main site noticing, it is a candidate for a tile.

Include a site you just launched, the same day. The checklist for the launch is not done when the deploy succeeds. It is done when the next person on the morning pass would see that host. Tomorrow’s you is that person.

Leave off anything that is not yours to monitor. A partner’s site, a social profile, a payment provider’s dashboard, and a host’s status page are someone else’s uptime. You can visit them when a tile points at a problem. You do not add them as if you operated them. Also leave off preview deploys and one-day test hostnames. They guarantee red tiles you will ignore.

Mark what you will do about each line

Next to each hostname, write one of three words: repair, report, or skip.

Repair means you can open the host and change something. These lines should be tiles.

Report means you will tell someone else, a client or a teammate, when the check fails. These lines should be tiles too, and the description should name who gets told.

Skip means you will not repair it and you will not report it. Do not add it. A skip that stays on the dashboard becomes noise. If skipping feels uncomfortable, the line is probably a report, and the discomfort is the lack of an owner. Assign one, or accept the skip in writing.

This mark is the cure for a hoarded bookmark list. Bookmarks are everything you have ever touched. The site list is everything you still stand behind.

Count, then pick the plan, then add

Count the repair and report lines. If the count is 5 or fewer and email alerts are enough, Starter at $19 per month is the published fit, with checks every 10 minutes. If the count is 6 to 15, or you need webhooks or SMS, Pro at $69, with checks every 5 minutes. If the count is above 15, or you need more than one seat, the API, or white-label, Agency at $149, with checks every minute and up to 5 seats. Annual prices are $190, $690, and $1,490. The reasoning is expanded in choosing a plan. Do the count before checkout so the trial is on the tier that can hold the list.

Add sites one at a time, or in a short batch, and read the first health check before you paste the next ten. The flow is name, public URL, optional description. Adding a site covers the first result. A first offline on a URL you have not opened is a listing error until the browser agrees the site is down.

Connect stats where a number would change your weekly decision. Not every tile needs visitors. A status page often needs only health. A store needs revenue if you can send it without a research project. The pipes are in how numbers get onto a tile. An inventory that waits for perfect analytics never gets the health checks, so health comes first and stats come on a second pass in the same week.

Keep the list in one place

The dashboard is the list once it is complete. Until it is complete, the note is the list, and you should be able to point at any line and say “added” or “not yet.” Delete lines from the note as the tiles exist, so the note does not become a second, drifting inventory.

When you retire a site, remove the tile after the site is really gone, or keep the tile for a short time if you want the next check to prove the URL no longer answers. Then remove it. A retired tile that stays red for a year teaches everyone to scroll past red. That habit will hide the next real failure.

When you rename a hostname, update the tile the same day. Monitoring the old name after DNS moves is monitoring a memory. The offline sequence starts with the saved URL. If the saved URL is stale, the sequence starts wrong.

Once a week, compare the note you no longer keep — your head, the footer, the client roster — to the tiles. The weekly review already asks which sites you did not review because they are not in the dashboard. That question is the inventory, repeated so it survives a busy month.

A worked shape, without fake numbers

Here is an illustrative example, not a customer story. A publisher operates a magazine, a shop on a subdomain, and a podcast site. That is three repair lines. They also still own a launch subdomain from a spring campaign that they do not intend to renew. That line is a skip, or a repair if the campaign URL still takes ads. They decide it is a skip, let the certificate lapse on purpose, and do not add it. The three live hosts go on Starter, which has room for two more. When they open a documentation wiki for contributors, they add it that afternoon as site four, while the cap still fits. They do not wait for a quarterly cleanup. The cap is a constraint they can see. The forgotten wiki would have been an outage they could not see.

Your list will be different. The shape is the part to copy: public hosts, a verb for each, a cap you fit, an add on the day the host starts to matter.

What the list will not solve

It will not watch a site you refused to add. It will not shrink to fit a cheaper plan by itself; you have to remove tiles or pay for the tier that holds them. It will not tell you whether a site is profitable. Revenue stats, if connected, tell you what was reported. The decision to keep a site is still yours.

If the honest count is above the plan you wanted to buy, the list is doing its job by being awkward. An awkward list is cheaper than a calm dashboard with a hole in it.

Questions about the inventory

Should www and non-www both be tiles?

No, unless both hostnames serve different sites on purpose. Pick the canonical URL, make sure the other redirects, and monitor the canonical one. Two tiles for one site spend two slots and can disagree for no operational reason.

How do I handle a site I only have for a month?

Add it, and put the end date in the description. Remove the tile when the month ends. Short projects are the ones left running, and monitoring, by accident. A dated note makes the removal obvious in the weekly review.

What if I find more sites than the plan allows halfway through?

Stop adding when you hit the cap. Either move to the plan that fits the real count, or mark the remaining lines skip with your eyes open. Do not create a second personal account to hide the overflow. You will not check both. One list is the point.