Delivery
Dependency management
How to identify and manage dependencies between teams, systems, and suppliers: owners, timing, risks, and solutions.
Addiction becomes a problem not because of the fact of the connection, but when it has no owner, deadline and backup scenario. Identify critical dependencies before the date is promised, make them visible to both parties, and if possible change the architecture of the work so that the next cycle requires less coordination.
Definition and role of dependency management
What is Dependency Management
Dependency management is the practice of identifying, planning, monitoring, and minimizing the impact of dependencies between tasks, teams, and external counterparties on the course of product delivery. Addiction is anything without which a task cannot be completed on time and with the right quality: third-party services, features of other teams, external APIs, infrastructure, resources or approvals.
Why Dependency Management is Critical for Delivery
Any dependence is a risk to the time, cost and quality of delivery. Ignoring dependencies leads to paralysis of development, team conflicts, cascading delays, deterioration of the quality of releases. The more complex the product, the more points where dependence can manifest itself.
Example: The team wants to roll out a new feature, but is waiting for the backend from the next team. Without a clear fix of this blocker and control of its status, supply would be at risk.
Classification and Model Examples of Dependencies
Types of dependency
Depending on the source and influence, there are several types:
- Internal: between tasks, teams, modules within the company.
- External: third-party services, contractors, partners, regulators.
- Blocking: Complete inability to move on.
- Conditional: you can start work in parallel, but complete - only after fulfilling the condition.
A classic example for Delivery
The mobile development team depends on the API update the backend is working on. Internal planning doesn’t take this lag into account: the mobile team is out of work, testing is shifted, the release is disrupted.
Dependency management tools and processes
How to fix and track dependencies
Good governance implies transparency for all participants. The following approaches are used:
- Maintaining a dependency log or special boards (Jira, Trello).
- Explicit labeling of dependencies in tasks (Blocked by, Depends on).
- Regular withdrawal of dependencies on team syncs.
In good processes, all critical dependencies are visible at the level of roadmaps and release plans. At the planning stage, the team tries to identify the blockers as much as possible in advance.
Change management procedures
You can’t just fix the dependence – it is important to constantly update the status, recheck the risks and look for workarounds. Example: The external API is delayed, the team can temporarily use the mock or prepare a foulback.
Mistakes and anti-patterns in dependency management
What’s not working?
- Silent dependencies: The team feels that external tasks will be solved by themselves, and no one reports them.
- Acceptance of deadlines: Planning is built on promises without status verification.
- No prioritization: all dependencies are in one row; critical ones are not highlighted.
What to do differently: practical cases
In one of the cases, the integration team weekly checked the status of dependencies through a separate channel in Slack and updated the team tracker. This made it possible to see shifts in advance and quickly find temporary replacements or change the order of work.
Integrating Dependency Management into the Supply Process
How to build practice in day-to-day delivery
- At the planning stage, make dependent tasks for separate discussion.
- Evaluate the degree of risk for each dependency: the presence of a backup option, the impact on the deadline.
- Assign responsibility for each key addiction and track progress on it separately.
Example: the SaaS product team prepared a release that depended on the certification of a new component by a third party. To minimize the risk, the development was carried out in two versions at once: with and without a component, which helped to release the product on time.
Metrics and retrospective analysis
The percentage of tasks blocked due to an external blocker, the proportion of opaque dependencies, the average unlock time are estimated. These indicators are a good material for retrospective and process improvement.
It is recommended to look at benchmarks and approaches of leading companies:
FAQ
*What is a dependency in a food team? This is all that you can’t complete or start a task without – internal and external systems, services, solutions, people or documents.
**What are the dangers of non-obvious dependencies in the project? They lead to unexpected blockers, wind up deadlines and can spoil the quality of the release.
**What role do metrics play in dependency management? Metrics help to understand how often and strongly addiction disrupts deadlines and where bottlenecks are in the processes.
**Is it possible to get rid of all addictions? No, but you can make them transparent and minimize harm.
**What tools do you use to manage dependencies? Jira (linked issues), Confluence (dependency log), manual boards and Blocked/Depends tags.
*How do you train your team to work with dependencies? Debrief past errors, regularly discuss planning, and encourage explicit labeling of blockers.