Anna Hannemann (Picnic) identifies the core reason roadmaps 'lie': cross-team dependencies are left implicit. Two failure patterns recur — a team that is ready to start gets told by another team 'come back in two weeks', or a tech lead attends an alignment meeting without real bandwidth, nods along, and only raises a blocking disagreement once the dependent team is already halfway through development.
The fix is a disciplined, time-boxed pre-alignment cycle run over roughly four weeks in three stages:
Development continues in parallel throughout — the dependency-uncovering work itself is only a couple of time-boxed meetings, not a four-week freeze. The intended output is a transparent per-team delivery plan mapped across weeks, with every cross-team dependency explicitly accepted, renegotiated, or escalated, critical task groups visible to leads, and no hidden delivery-impacting assumptions.
Alignment isn't a one-time gate: once development starts, gaps (e.g. a missing data point) still surface, so Picnic runs weekly check-ins between the corresponding tech leads on both sides of a dependency to keep realigning through the quarter. This continuous-loop stance echoes Continuous Strategy Feedback Loop and Checkpoints vs. Performative Rituals (QBR Theater) — the check-ins are meant as real work sessions, not status theater. See also Explicit Trade-off Prioritization for how competing asks get resolved before they turn into hidden dependencies in the first place.
Picnic runs this pre-alignment as a roughly four-week, three-touchpoint cadence ahead of big or cross-team initiatives:
The cadence is deliberately timeboxed to about two meetings in four weeks so that surfacing dependencies doesn't itself become a second roadmap-killing time sink.
Every cross-team dependency found in that session is forced into one of three buckets — accept, renegotiate, or escalate — so nothing is left as an assumed "it'll work out." The goal: no delivery-impacting assumption remains hidden.
Alignment doesn't stop at kickoff: teams keep a weekly check-in between corresponding tech leads running through delivery, since "you align, you prepare, but then life happens" — gaps (e.g. a missing data point) that pre-alignment missed still surface mid-execution.
The reason this process carries so much weight: the costliest dependency failures aren't a team openly refusing to help, but a tech lead nodding along in a planning meeting without real capacity to deliver — the conflict then surfaces only once the other team is already halfway through development. Damage scales with how late the hidden disagreement surfaces, which argues for early, forced resolution over polite ambiguity.
See Explicit Trade-off Prioritization for how competing asks are made concrete once surfaced, and Build In-House for Differentiation, Buy for Commodity Functions for why Picnic's build-everything-in-house strategy is what generates this dependency load in the first place.
Из тем: Transformation in Practice