Overhead, not downtrodden Penny Wyatt The Lead Developer UK 2016

penny wyatt This is a topic that many people are looking for. star-trek-voyager.net is a channel providing useful information about learning, life, digital marketing and online courses …. it will help you have an overview and solid multi-faceted knowledge . Today, star-trek-voyager.net would like to introduce to you Overhead, not downtrodden Penny Wyatt The Lead Developer UK 2016 . Following along are instructions in the video below:

“U003e thank you very much meri. I m an engineering nmanager. But once upon a a time a very very long time ago. I was just a humble software ntester the way i process the way our process worked went like this so.

The developers. Nwould write their features they would poke a bit around the ui write automated tests. Nmake sure they were green and then they would declare it done then i and another tester nwould. Pick up the stories and have another little go at testing them.

What a surprise nwe would find issues in fact we found so many issues that out of nine months of development nwork. We generally created another three months of rework to actually get that stuff shippable nyes. That s a very long release cycle. This was a very long time ago now testing is nnot really very glamorous profession and so there were times.

When i questioned my worth nin the company when i did that i would look at that three months of work that i created nknowing that it was improving the customer experience in the end and say i felt pretty proud about that so i m here ntoday to talk about overhead teams. I don t mean overhead. I mean financial overhead nthe roles in a company that are really important crucial. But they don t directly bring in nthe cash.

If somebody wants to make more money they re not going to say the first thing. I nwill do is high some more testers. So some overhead roles are at the timing. Overhead nsupport bug maintenance tech.

Writing and leadership and sensible companies try to nreduce their overhead as much as is possible and so it often means that when you re in nan overhead role it can feel like you get defensive or you get victimised people are ntrying to cut us. They don t understand the value that we add that can often happen because npeople in overhead roles can look at just the small picture and not understand. Where nthey fit into the broader picture of the organisation so back in my testing days. That was definitely nthe case.

All i cared about was the quality of the product. Don t get me wrong. It s pretty nimportant. But it doesn t mean.

It said the only thing that is important the work we nwere doing was adding an extra three month delay to getting the software out the door nand other people coming in looking at the broader picture were saying. It takes three nmonths to get from code complete to done that is atrocious. I was looking at that in nthe meantime..


Saying of work. I m awesome view. He held me. And my colleague accountable nnot.

Just for the quality of the software delivered. But also for the speed of the development nprocess. So what do you do when given that kind of accountability especially when it nfelt like it conflicted well if you are like me stubborn the first thing you do nis try to dig yourself deeper. I said the code that we are getting from upstream nis really poor quality it takes three months to get that to production quality quality nis non negotiable.

One of our company values is don t f. The customer therefore. There nis nothing we can do to reduce that three months end of story. So thankfully.

My boss helped me to see it na different way and so we started again with that thought process okay. It takes three nmonths because upstream code quality is poor. Why s that it is because the developers aren t ntesting. Very well why is that first of all they don t know how to secondly.

They don t nreally care. Very much and thirdly they re not being held accountable for the quality nof. What they are shipping. So now we re no longer at a dead end we have three problems nthat.

We need so solve so that s what we did we changed our role entirely. We stopped being na testing team and we became what we called a quality assistance team and our full time nrole was to educate developers in really good testing techniques really efficient testing ntechniques secondly to build the customer empathy in the team help the devs realise nthis isn t just a bug android issue tracker. This is something real that s going to impact nreal people thirdly. We worked with the dev team leads and dev managers dev to hold nthem accountable for what is doing it worked i won t say it is easy it took several years nbut.

We went from 100 stories needing rework to four per cent. We got rid of that three month. Nrework process entirely in fact these days. The team shifts multiple times a weeks to nproduction.

But most importantly. The quality of the product actually increased because nall that customer empathy and care about what the customer was really doing was baked in nat. The start of the process and not pushed into the chocolate chip cookie later on to nuse heidi s example..


So that s the benefit of looking at the broader picture. If we had nonly looked at quality quality quality. We don t care about speed. Then we would have ncontinue to cost a lot more to the company.

So it is all they nice when you re in an overhead nteam and you can address the root causes. Like that but often you can t so let s take another nexample. Which is bug fix so earlier this year. I was working as a dev manager.

And i ntook on ownership of the jira bug fix team they don t fix new bugs. But the legacy of nold bugs signature. There in our sitting there in our public issue tracker. They re nnot serious only affect a small proportion of customers yet the total of the bugs adds nup to an amount of customer pain.

And an amount of support cost so when i took on this team ni spoke to the previous bug. If i can managers and asked them me the same thing. They said you will just nend up in this endless battle with the head of engineering. Trying to get more developers nbecause you can never get through that bug fix backlog enough really to make a difference.

Ni. Said is an important quality thing quality is nnon negotiable. Don t f the customer et cetera. Let me see if i can put together a compelling nbusiness case to get more devs on bug fix the way.

The system. Works is there is no one npermanently assigned to bug. Fix each iteration. A certain number of developers are taken off ntheir feature teams and put into the bug fix team at the time i inherited it it was four ndevelopers each iteration.

I said okay quality s really important but first of nall let s look at the broader view a bug s really the aspect of quality that our customers ncare the most about i looked at a bunch of met ricks. We had that aggregated all the different ncustomer feedback. We had coming in and there were different types of met ricks coming from n metrics end users. Cancelled evaluation and people who cancelled their subscriptions.

Nand. They all told. Me the same story taking the cancellations as an example of the people nthat cancelled their subscriptions ten per cent of them..


Said that they did so. Because nthey found the user experience. Too confusing too complicated and to be fair that can nbe true eight perseid that it was because they found the performance too slow 04. Perseid nit.

Was bugfix doesn t seem like the best place to nachieve that particularly since increasing bug fix investment meant. I was pulling those ndevs off the feature teams that were rewriting the confusing ui and fixing the performance nso. It no longer. I was no longer thinking team so we are working sustainably up with a way to measure the amount of bug ndebt.

We had in terms of customer impact. Then i looked back through the history of the bug fix nteams and look how many devs. We had had on the team. At each iteration and correlated nhow to change that metric the number that we came up with in the end was in order to nkeep that debt.

Stable we needed 25. Developers on bug fix. I think what is possibly the first ntime in history. I went to the head of engineering.

And i said to have those developers working on the things nthat customers care about rather than being overhead in this bug. Fix team. So there are nfour lessons. I would like you to get out of this the first one is that if you re in an noverhead role or you re leading an overhead team please don t deny it as but embrace nit recognise you re not the ones making the money and try to figure out how to redirect nthe company s resources.

So they can be secondly broaden your view. Don t just work ntowards your team s goals work. Towards the company and the customer s goals. And you may nend up finding.

Then you find better ways to achieve your own goals. There is an opportunity ncost to everything you ask for for your team understand what that is before you go out nasking for more thirdly try not to mistake your costs for your value as a tester. The nthree months of working. I was adding was not my value my value was the improved quality nthat was the cost that i was adding that was my overhead similarly with managers.

Nyour value is not the number of people that report to you it is the amount of work that nthat team is able to produce if you can get the same work. Done with half the number of npeople you re twice as valuable finally. Please..


Remember that yes leaders nare overhead. Too pretty much in this room is overhead. So. Look out i m coming for nyou next thank.

You very much applause going to take a selfie if i may laughter ncool thank you u003e u003e great job. One of the topics that we had nrequested a loot last year after the conference was to have more about remote working. We nhad a number of really cool and interesting submissions on that topic and this is the none that we chose out of those submissions. So we have laura coming to talk to us about nhow to deal with the culture divide and internal advocacy in a distributed team.

She s a developing community nmanager at couchbase and i m really excited to hear her thoughts on how to help our remote nteams have a consistent culture and to work with well together welcome to the stage nlaura applause u003e u003e. Hello. Everybody i hope you can hear me nokay yes you re not seeing what i m seeing shoot. I m sorry.

This was working best laid nplans and all that u003e u003e. I threatened to sing. But none of us deserved nthat u003e u003e. How.

Is everybody today all awake after nlunch no come on wakey wakey applause. okay we are going to do this currently hating nmy operating system right now it is not a mac. It s not windows either n. U003e u003e and i can t blame this on the tv trucks u003e u003e.

It did work it is because of the slide n. U003e. U003e right everybody. Up super hero poses laughter .

Come on like the middle row. Here. Nare. Awesome.

And the rest of you are just hanging out turn to the person next to you nand teach them the most weird word you ve ” ..

Thank you for watching all the articles on the topic Overhead, not downtrodden Penny Wyatt The Lead Developer UK 2016 . All shares of star-trek-voyager.net are very good. We hope you are satisfied with the article. For any questions, please leave a comment below. Hopefully you guys support our website even more.

“There are many super-important roles in software teams that are also financial overhead, e.g. QA, Support and Bugfix. They re necessary, but they require people that would otherwise be working directly on business goals. For leaders of these overhead teams, it can be frustrating, having to justify keeping them well-staffed. We often point to the problems in the business that they address: We need a big bugfix team, look how many bugs we have! nnBut what if we flipped our thinking? Instead of denying that we re overhead; we embrace it. Instead of using the size of the problem to justify the team size; we aim to reduce both. Let s halve our bugfix team by halving the bugs we ship in the first place . In this talk, using personal examples from JIRA QA and Bugfix, Penny Wyatt will describe how this flip in leadership mindset can lead to happier engineers, teams, stakeholders and customers.nnABOUT PENNY WYATTnnPenny is a QA and Dev Manager at Atlassian (an Australian software company), working on an issue tracker named JIRA. Her role is to challenge everything that makes sense, and to come up with crazy ideas that improve Atlassian s quality and devspeed.nnSuccessful experiments include: improving quality by reducing testing; getting better QA engineers by hiring non-testers; and improving devspeed by increasing the scope of dev work.nnIn previous roles, Penny s been in dev, testing, dev-in-testing, IT support, and bellringing for weddings. Her favourite hobbies are eating and sleeping, with travelling a close third.”,


Leave a Comment