Your designer catches a spacing bug on a 27-inch monitor. Your client opens the same page on an iPhone in a taxi and never sees it, because the layout collapses to one column before that bug even appears. Two people, two devices, two completely different views of the same build. When both need to leave feedback, the tool you picked decides whether that happens in an afternoon or drags across a week of resent invites.
Why cross-device review breaks in invite-based tools
Tools like Markup.io are built around a project and a seat. You create the project, add the client as a collaborator, and they get an email invitation. That flow assumes the client will do three things: open the invite, create or confirm an account, and figure out the interface. On a desktop with time to spare, fine. On a phone between meetings, each of those steps is a place the client quietly gives up.
Then there is the device split itself. Your designer wants to comment on the desktop layout. Your client is looking at the mobile layout, which is a different arrangement of the same content. If the review tool only knows about one captured state, one of them is commenting on something the other cannot see. You end up mediating: "the client means the mobile menu, not the desktop nav." That translation work is the real cost.
The account requirement compounds it. A client who reviews one project a quarter does not want a login for your tooling. Every credential you ask them to manage is friction that shows up as silence. If you have hit this wall before, the Markup.io alternative that skips the invite is worth a look, because it removes the account step for the person you least want to slow down.
How a no-login public link handles both reviewers
The fix is to stop treating the reviewer as a seat and treat the review as a link. With Cobalt Capture, whoever is looking at the screen captures it themselves. Your designer captures the desktop layout from their browser. Your client, on their phone, opens the published link and reads what is there, or captures their own mobile view if they spot something. No install, no extension, no signup on either end.
Each capture becomes an item with a comment attached. Your designer can crop the still to the exact panel they mean and drop a numbered pin on the misaligned element. Your client can type a comment, or dictate it if they are in Chrome or Edge, which matters more on a phone where typing is slow. On Firefox they type instead, since dictation relies on the browser's own speech support.
The output is one public URL of the form /r/<slug>. Both reviewers point at the same review. Anyone with the link can add a comment to any item, and you, as the owner, mark each one resolved as you work through it. No one had to be granted access first.
Keeping desktop and mobile feedback from colliding
The trick with two devices is to make each comment carry its own device context, so a developer never has to guess. Because every item holds its own captured still, a desktop comment sits on a desktop screenshot and a mobile comment sits on a mobile one. The image itself tells the receiver which layout is broken. That is far cleaner than an annotation drawn on a live page, which shifts the moment the viewport changes. There is a longer argument for why a cropped still beats an annotated live page if you want the reasoning.
A few habits keep the two streams readable:
- Have the client capture from the device they actually use, not a resized desktop window. A real phone screenshot shows the real breakpoint.
- Ask reviewers to name the device in the first line of a comment when it is not obvious from the crop.
- Use numbered pins so a developer reads steps in order when a single screen has several issues.
When you close the review, you have more than a link. Export it as a PDF or Word document for a stakeholder who wants a file, or hand the plain-text markdown at /r/<slug>/markdown to whoever builds the fix. The same review serves the designer, the client, and the person doing the work, without anyone re-entering the feedback somewhere else.
When the split is really about who can reach the URL
Sometimes the desktop-versus-phone problem is not about device at all. It is that your staging site works for you on the office network and not for the client on their phone data plan. That is a different failure, and it is covered in the staging URL works for you, not the client. But once the client can load the page, the review itself should never be the thing that stops them.
If most of your friction is the client learning yet another interface, the pattern in collecting client feedback without making the client learn a tool maps directly onto the phone reviewer. And for the wider set of feedback jobs this covers, the design review workflow shows the full loop from capture to a link you can send.
Next time a designer and a client need to review the same build, open a review, capture your side, publish, and send one link. Let each of them capture what they see from the device in their hand.