INTRODUCTION
Drag and drop is one of the few interactions that feels as good in a demo as it does in daily use. Pick up a card, carry it to another column, let go. No menus, no forms. Then someone with a tremor tries it with a trackball, or someone uses a head pointer, or a speech tool that emulates a mouse, and the board that looked effortless becomes the hardest screen in the product.
Since WCAG 2.2, this has a criterion of its own. Success Criterion 2.5.7, Dragging Movements, at level AA, says that “all functionality that uses a dragging movement for operation can be achieved by a single pointer without dragging”, unless dragging is essential. This piece is for founders and product leads whose products include boards, sortable lists, sliders, maps, crop tools or file drop zones. The decision: what the second way to move should be for each drag, and how to design it so it is not an afterthought.
Why dragging is harder than it looks
W3C’s explanation of the criterion breaks a drag into separate actions: press to set a start point, keep pressing while moving the pointer, then release at the end point. The difficult part is the middle. Holding contact while moving precisely is exactly what some people cannot do reliably, and what some input devices make slow and error-prone.
A single-pointer alternative removes the “hold while moving” part. Tap or click, then tap or click again. Nothing has to be held.
Keyboard support does not settle it
The common objection is: “We already support the keyboard, so we are fine.” The W3C guidance addresses this directly. The criterion is separate from keyboard accessibility because people using a touchscreen may not have a physical keyboard, and keys such as Tab or the arrows may not be available to them. Keyboard support is still required by other criteria. It is just not the answer to this one.
Nor is a swipe. An alternative that relies on a path-based gesture, such as swiping or flicking, fails the related criterion 2.5.1, Pointer Gestures. The alternative has to work with plain taps or clicks.
A second path for each kind of drag
The examples in the W3C guidance map neatly onto common product surfaces.
Sortable lists. After selecting an item, show controls to move it up or down.
Boards with columns. Tapping a card opens a menu with “Move to…” and the list of columns.
Sliders. Clicking anywhere on the track moves the thumb there. A text field beside the slider also counts, because the on-screen keyboard is operated by single taps.
Maps and canvases. Buttons to pan up, down, left and right next to the draggable view.
Selection rectangles and crops. A selection mode where you click one corner, then the opposite corner.
Matching or connecting items. Click an item in the first column, then its partner in the second, and the line is drawn for you.
File uploads deserve a mention. A drop zone that only accepts files dragged from the desktop has the same problem; a visible “Choose file” button solves it. The drop zone was never the hard part of an upload anyway, as we argued in Design the upload before asking for the file.
The second path is often the better path
Teams tend to treat the alternative as a compliance chore and bury it. That wastes it. A “Move to…” menu does things a drag cannot. It can move a card to a column that is off screen. It can move twelve selected cards at once. It can say “Move to top” in one action, where dragging means scrolling while holding a card at the edge of the screen. A position field on a ranked list lets someone type “3” instead of nudging an item through twenty rows.
The people who rely on those controls will find them. So will power users, people working one-handed on a phone, and anyone who has ever dropped a card in the wrong column because the list scrolled under their finger.
The serious counterargument
Direct manipulation is fast, and every extra control adds visual weight. A board with up and down arrows on every card looks like a spreadsheet. Some teams fear that the alternative will clutter the interface that made people choose the product.
The answer is placement, not omission. Most alternatives can appear only after selection: tap a card and its menu appears; select a row and the move controls appear beside it. The drag stays exactly as it was. And the exception for “essential” dragging is narrow: W3C defines essential as something that, if removed, would fundamentally change the functionality and cannot be achieved another way. A freehand signature or a drawing tool may qualify; reordering a backlog does not, because the outcome, a new order, can be reached another way.
A roadmap tool, before and after
Hypothetical example. A roadmap tool lets teams rank initiatives by dragging rows and move them between “Now”, “Next” and “Later”. Before: dragging is the only way to do either. After: selecting a row shows a small toolbar with “Move up”, “Move down”, “Move to top” and a “Move to…” menu listing the three horizons, plus a rank field. Dragging still works. A product lead on a train can now reorder the roadmap with one thumb, and a teammate using a speech tool can say “click Move to, click Later” instead of fighting a drag.
What to do this week
List every place in your product where something is dragged: cards, rows, handles, sliders, crop boxes, map views, drop zones. Next to each, write the tap-only route that reaches the same outcome. Where there is no answer, that is the work. Where the answer exists but is hidden in a settings page, move it next to the thing being dragged.
Sources checked on 17 November 2026: W3C, Understanding Success Criterion 2.5.7 Dragging Movements and Success Criterion 2.5.1 Pointer Gestures (WCAG 2.2).