Score an item, check a session, or read how other teams estimate and prioritise — written for people doing the work.
Calculators and checks — no account. The full set is on the tools page.
Score Reach, Impact, Confidence, and Effort with Intercom’s scales.
Open toolAvailable hours from your people, leave, ceremonies, and focus — no invented velocity.
Open toolCost of Delay over Job Size with SAFe’s formula and closed scales.
Open tool
A shared number is the point of the conversation, not a promise to the calendar.
Browse guides
Choosing what to build is a decision about whose input counts, not only which framework you pick.
Browse guides
A backlog item is ready when someone else can build it without guessing.
Browse guidesA backlog item is ready for a number when the team can name the fact that would change it. A checklist that demands finished screens first is a gate, not a refinement.
GitHub's Copilot app now carries the Jira issue you choose into the pull request. Four lines — outcome, exclusions, proof, repository — decide whether you assign Copilot.
Coding agents can turn CI green without doing the change you meant. A guide to tickets that name what must change, what must stay, what must disappear, and what evidence would prove it.
Strategic debt is repaid if the bet works. Toxic debt costs you even if you stop. A prioritisation guide to when the bill comes due — not another estimation method.
Distributed teams lose the cues that make estimation work — and replace them with marathon video calls and anchoring bias. Three structural fixes: async voting, refinement before planning, and evidence-based sizing.
Teams answer 'when will this be delivered?' with a metric that was never designed for it. A practical guide to velocity, throughput, cycle time, Monte Carlo, and delivery-system metrics — and when to use each.
Teams don't struggle with technical debt because it's unquantifiable — they struggle because they estimate the wrong thing. A layered guide to choosing the right method for the decision you're actually trying to make.
Most teams have OKRs. Few have a backlog that reflects them. Three structural approaches — OKR-tagged scoring, scoped stakeholder input, and outcome-driven quarterly reviews — and a decision guide for choosing the right one.
Teams don't miss sprint goals because the work is unpredictable — they miss them because the planning session ignored who was actually available. A practical guide to velocity-based, available-hours, and hybrid capacity planning.
AI-first disrupters, developer-centric platforms, and enterprise tools are all fighting for the same space. A strategic breakdown of when to choose each — and why your tracker is only half the story.
The number at the end matters. The conversation that produces it matters more. A step-by-step guide to running an honest, structured, and genuinely useful first session.
Two decades of this argument, and teams are still asking the wrong question. A practical guide to when each method helps, when it hurts, and what actually makes estimation work.
AI hasn't broken agile planning overnight — but the pressure it introduces is making familiar practices feel brittle. A reflection on what's changing and where teams are adapting.
When to use RICE, WSJF, MoSCoW, MaxDiff, or Buy a Feature. A practical guide to choosing the right framework — and defending the decision.
Decisions made in isolation miss crucial data and lose buy-in. Learn how inclusive decision-making balances diverse perspectives with clear host accountability.
Compare Fibonacci, T-shirt sizing, story points, and #NoEstimates. Learn which technique fits your team and why the conversation matters more than the number.
14-day free trial. No credit card required.