Indistractable: turn intention into agreements about attention

You open a tab to check one detail in an API proposal and return twenty minutes later without the answer. No notification sounded. The specification was ambiguous, formulating the next question felt difficult, and looking for something easier offered temporary relief. That kind of detour helps explain why Nir Eyal’s Indistractable, written with Julie Li, does not stop at telling readers to put away their phones. It asks what they intended to do, what shifted their attention and what change would make it more likely they act on their own choice.

The book in brief

Published in 2019, Indistractable: How to Control Your Attention and Choose Your Life proposes a four-part model: understand internal triggers, make time for traction, reduce external triggers and prevent distraction with pacts. Eyal argues that technology is one source of interruption, but not the only one; the publisher also highlights discomfort and workplace context (Simon & Schuster / BenBella Books, Bloomsbury).

Why read it: it is useful for people who begin tasks with good intentions and finish the day responding to other people’s cues. The book offers a sequence for examining detours without reducing everything to notifications or willpower. It also helps teams discuss the work system when interruptions are part of the organization’s design.

Traction and distraction depend on intention

Eyal calls actions that move someone toward what they chose traction and actions that pull them away from that intention distraction. The same activity can belong on either side. Reading a message may be a distraction during a review or traction when the planned task is answering the team. The classification therefore depends on a prior intention and a realistic plan; it does not depend on whether an app seems good or bad.

This distinction avoids demonizing technology. Browsing the web can be part of investigating a bug, while opening the same page without a purpose can be an escape. The practical question is: “What had I chosen to do now, and what changed?” That is more useful than counting screen minutes without knowing whether the time supported work, rest or a relationship.

Four steps for dealing with detours

The first step is managing internal triggers. The book observes that boredom, anxiety, uncertainty, fatigue or discomfort may come before a person seeks another activity. Rather than treating an impulse as a moral flaw, readers can notice what happened before it, name the feeling and choose a response. If a task creates insecurity because acceptance criteria are missing, the next step might be to write a question or request context, not block social media.

The second step is making time for traction. Eyal recommends planning a calendar to reflect how people want to spend their time, including work, rest and relationships. The calendar makes conflicts visible that a task list can hide: four hours of focused work and six hours of meetings will not fit into the same morning. Planning does not guarantee execution, but it creates a reference for reviewing choices and noticing when a schedule was designed without enough capacity.

The third step is reducing external triggers. Notifications, inboxes, conversations and objects in the environment can pull attention. Adjusting alerts, creating checking windows and agreeing on urgency signals can help when someone controls the channel. The book need not be read as “turn everything off”: people on call, providing support or working through an operational queue have to distinguish what should interrupt from what can wait. Eyal himself writes that workplace distraction can point to organizational dysfunction, not only a personal problem (Eyal on distraction at work).

The fourth step is using pacts: prior commitments that make a detour harder or clarify an intention. A pact might be an agreement with another person, a technical barrier or a personal rule. A team could agree on a meeting-free window for an architecture review; someone studying might use a blocker for a period they chose. Pacts involving money or social pressure can create coercion, so they need to be voluntary, proportionate and reversible. A penalty that punishes someone for conditions outside their control is not a neutral productivity tool.

An applied case: reviewing an API change

Consider this hypothetical scenario: an engineer needs to review an API proposal used by three clients and identify compatibility failures. During the first attempt, they switch among the proposal, chat and documentation. The task has no bounded next step, and the team treats every customer message as urgent. The engineer records the intention as “compare the new contract with all three consumers and note questions that prevent a safe migration.” That creates a criterion for recognizing traction; it does not label every interruption a failure.

They reserve two 50-minute blocks on different days. Before the first one, they agree with the support owner on coverage for requests that can wait and identify the incident channel that should interrupt the review. The team defines a window for ordinary questions and a contact for compatibility decisions. If no one can cover the service, the engineer shortens the session or chooses a collaborative review instead.

During each block, the engineer notes when they want to switch activities and chooses a broad category: unanswered question, waiting on another person, external interruption or fatigue. They do not record private content or score productivity by absence of breaks. If a question requires domain knowledge, asking someone is the right work. At the end, they leave a note with consumers checked, incompatibilities found and decisions still needed.

After two attempts, the team reviews outcomes and costs: which incompatibilities became clear, how much resume time was needed, which messages required a response and whether coverage created extra work for colleagues. Two blocks are a test choice, not a universal prescription or a promise that the review will finish. If urgent requests were frequent, the issue may be priority or support design. If the task advanced but produced no decision, the next step may be to align contracts, not book more time.