The lean engineering process is agile and fast-paced, with numerous moving wheels in any single moment. Usually, the technical architecture is in place – the steps have been laid out, sprints are in place, continuous development is the way to go.
But What Happens When Something Goes South?
Short iterations and continuous improvements aren’t easier to match and gauge. Initial analysis is usually trimmed down, since it’s a recurring activity performed every week or two. So, fewer people estimating tasks, less attention in terms of risk management, simple goals to be achieved.
The standard problems which occur with lean engineering teams are:
- Miscalculated estimates
- Buggy features not caught by automated tests or manual QA (due to time constraints)
- Undocumented changes adding up over time
- Miscommunication from tech folks failing to deliver and/or report a problem
- Future sprints not accounting for delayed or buggy features
- Inadequate management of technical debt
- Missing project revisions (every 2–3 months) calibrating the pace and the ongoing goals
For instance, a social network starts small, grows in popularity, keeps pushing changes, and follows their initial process zealously. In the meantime:
- The project grows in scope – several high-scale features are added to the pipeline over the next few months.
- Dozens of team members join the team. Onboarding may be insufficient, thus experienced folks are pulled frequently for guidance, mentoring, code reviews, business revisions.
- Bug reports and customer requests keep piling up. Prioritizing becomes a challenge.
- Major bugs keep interfering with the sprints, delaying other features.
- Testing takes longer. Bugs and new features pushed together, regressing existing functionality.
- Performance is challenged. Speed optimization sprints are planned, in addition to bug fixes and business requirements on the roadmap.
- Technical debt gets pushed back over and over, along with documenting change sets and keeping everyone up to speed.
- Companies often hire scrum masters or other senior managers with different styles of work who try to get everything back on track, often taking a long period of time to rectify anything, stressing everyone out and digging deeper and deeper.
It’s A Fine Line
It’s a fine line balancing bug fixes, new features, performance and security updates, scalability, additional business requirements by partners and investors, code cleanup, and keeping everyone in sync.
Depending on the growth curve, this may happen pretty much out of nowhere, leaving the business with scarce resources, sometimes turnover due to stress, roadmap out of sync, and bugs lurking under the surface.
Service-based companies are most prone to facing this earthquake and failing to get on track without losing several high-profile projects. There are too many stakeholders with arbitrary deadlines and requirements, leading to communication overhead and challenging team scale in-house across different teams.
Of course, this is just as applicable for product companies, reporting to a board of directors, investors, partners, and the public (especially in the event of bad PR).
Senior Management Having An Outline Of The Process
Best-case scenario, senior management would outline the process a couple of years ahead, based on statistics and projected growth and interferences. A sample plan can cover:
- Sprints allocated on performance optimization and cleaning technical debt – probably every 4 months at first, 2–3 months a year after.
- More wiggle room in every sprint – such as booking resources at 70%, preventing layoffs, holidays, sick leaves; planning for onboarding, meeting sessions, and more.
- Carefully planning less important features in the roadmap. Here’s why…
- Scope review every 3 months or so. Pushing back on less critical features if needed, or celebrating deliveries of “bonus” assignments if everything moves as anticipated (best-case scenario). Making sure documentation and onboarding docs are kept up to date – and quickly updated in case they aren’t.
- Proactive communication at all times. Reporting a problem, scope creep, miscalculated estimates ASAP can prevent multiple bottlenecks from piling up later on.
- Preparation for rainy days. Covering possible PR obstacles, competition intervening, lack of talent delaying projected growth, budget management.
- Willingness to turn the process around if and as needed. Flexibility to act outside of the established protocol as soon as it’s clear that the current workflow won’t accommodate the ongoing roadmap.
While being inconclusive, the plan sets some extra buffers and allows for hiring early on, automating critical tasks and projects, allocating enough time for reviews and everything in-between.