Rupert one of our DevOps Delivery Managers attended Agile in the City Bristol, a conference that typically attracts industry practitioners who want to improve their success with agile and lean methods.
Here’s 5 key learning points Rupert took away from his three days at the conference:
1. Use data analytics in lean and agile ecosystems:
Alan and Jan from Dataswarm demonstrated how it’s more important than ever to build data models into product experiments and to be ready to iterate or pivot thinking based on the outcomes. The idea of adaptive processes was also reviewed, in the context of what the data tells us. We should be asking:
- When should we be agile in our processes? Data suggests early phases in a product or organisation
- What’s the crossing point for adapting a more “clean” lean manufacturing point? Commoditisation of product and maturity of processes may offer a better value proposition, so automation and tight feedback loops are more important than ever at this point
2. Retrospectives for distributed teams can work well:
A great session from John focusing on the techniques used to lead successful retrospectives (and other meetings!). The focus was on teams where there’s a large proportion of split site or remote attendees. Of course, there was a plethora of tools to choose from (and a lot of those are free to use) but John also worked through the human factors which are often what really makes the difference in a successful retro.
Here’s some of John’s top tips:
- Preparation is key: Good meetings generally don’t just happen, they require significant upfront effort. Plan for this, ideally leaving 2 -3 times the meetings length as a guide to how much planning time is needed
- Icebreakers make a huge difference: Getting everyone talking within the first 5 minutes sets the scene for the next 55…
- ROTI – Not the delicious staple of Indian cuisine, return on time invested. This is a simple method to gauge the team’s perception of the value of the meeting and all you need is a 1 to 5 rating system (e.g. a hand!). Ask the team to come up with their own simple scale (for example where 1 may be “I’d rather sandblast my eyes than be here” to 5 being “I never want this meeting to end” – use whatever works for the team). At the end of each meeting, get a rating. This will give you all a feel for how the meetings are working. Leave enough time to react to this though: if you’re seeing low scores, dig in a bit and agree what needs to change for next time.
3. Are you being agile or doing agile?
Joanne led us through an examination of modern enterprises and how they approach getting things done. The pressure to “be productive” gets lost in the compelling need to be busy. This is all too often reflected in the growth of an enterprise which looks like a relatively calm sea; some peaks, representing break-throughs, some troughs, representing lowering return. Fundamentally running to stand still.
An enterprise that is focused on growing and learning has peaks and troughs too, but the peaks tend to last longer and go higher, leading to a net increase in value or return over time.
Joanne explained that what has proven to make a difference is leaving behind command and control management practices and adopting leadership practices. Leaders create and embrace a culture that is built on experimentation and learning. Leaders focus on outcomes, not outputs or data proxies used as substitutes for productivity. Joanne went on to explain how she uses the “Lean Value Tree” approach, which articulates growth and purpose in terms of Vision, Goals, Bets and Initiatives.
In summarising, Joanne cautioned against the one size fits all approach, as this does not work with people or within enterprises. Be lean with your Agile; experiment, review and adapt!
4. Practical, team focused operability techniques for distributed systems session:
Mathew whizzed through something close to the heart of every DevOps practitioner: Operability, coining a simple, yet inclusive definition:
“Software operability is the measure of how well a software system works when operating ‘live’ in production”
Matthew then went through a series of techniques and practices that will help any organisation improve their visibility and therefore capability to react to changes in state or behaviour. Critically, whilst these are best considered at the design stage of any service or product, with the right mix of skills, many can be applied to production environments retrospectively.
5. Cross the river by feeling the stones
In this keynote session from Simon, he explained the power of maps and how enterprises can use the component parts maps to navigate their own journeys and succeed. I found Simon’s narrative captivating, with the logic and conclusions startlingly simple, but with such obvious potential. The fusion of mapping fundamentals, lean learning and value chains, set in the context of the evolution of product offerings was expertly weaved into a funny, engaging and uplifting, horizon expanding experience.