Been here way too many times. Worse is the story points never seem to match actual days.. cause after asking about story points they ask how many days.
Absolutely true. Also funny if the estinates are not in time but just complexity. Ans then there are Tasks like MANUALLY put in those 10.000 datasets into a database. Well, conplexity? 0? My little Brotterode who is 8 can do that. But time wise if you cant automate it? Probably one to two weeks? Okay. So 0 Story points it is!
@@notHifen But you have a specified number of days in a sprint and you want to plan the sprint in a predictable way, which is tasks not being added mid-sprint and tasks being completed by the end of the sprint. How do story points which do not match days fit in there. If they don't, what are they good for?
@@daladirn5119 Because through a few trials and errors, a teams velocity will become rather consistent. "Oh we can do 40 story points every 2 weeks". The thing is a) tickets aren't the same for everyone, it might take one person 3 days to do something, compared to someone else's 5. Story points are relative to the size of other tickets, so the subjectiveness of individuals are removed. The other point is as you get better, ticket "days" become less, which means despite you're teams velocity increasing over time, there is no metric to measure that since you've always got the same "days" scheduled a sprint. If you have a ticket today that you mark as 3 days. But 5 years later you comeback with experience and relook at that ticket and say "now its 1 day", then your metric is bad because it can just change. Story Points reflects work effort of a ticket, If I see a team has 80 points going through, a PM can estimate the amount of work getting done, he doesn't care about team composition. If you put "15 tickets equaling 80 days of work" that tells them nothing, is it a one man team, and these are all small tickets, or is it a large team and these tickets are huge. Telling me "days" doesn't tell me scope of work.
I think he's got them over a barrel and if he did leave it would be for either less money or he wouldn't be able to get away with half of these things and he has no guarantee the next company won't screw him over in the same way.
As a tech, always estimate double the time since you never know if will have a problem that may take a while to fix. Never give them the exact amount of time. And when you finish earlier you look good or you get time to verify a few times to make sure it works the way you want it to work
they figured out scotty factoring already and planned against it. Managers DEMAND that you fail the estimates otherwise it looks like their the only ones ever fucking up the timeliness. If they can't have the occasional mistake to lord over you then they'd rather not even have you
unfortunately, unless the jareds have a strong hold inside the core company structure, the jareds are mostly the ones kicked off by management via politics.. seen it too many times... that or they the ones who gets off the company like a chad
True tho, as a software developer promoted to "tech-lead / project manager" I've made it my mission in life to protect my developers from management. Holy crap the amount of bullshit meetings and feedback loops I have to go through. When all I really want to do is code. But we have tried "un technical" project managers in the past. And they simply waste everybodys time on getting "estimates" We spent more time estimating than getting work done.
True I always say I need a week for the dev while it's actually a week for dev + QA. It's really great when the tech lead understands the bullshit us devs had to deal with from incompetent PMs and BAs and cooperate with us to always overestimate the dates
Dad's a project manager, but he was online in the 90s and taught science most of my life. When he schedules an hour, it's an hour and when things get unprofessional, he's spent decades wrangling 2 dozen teens at a time.
The task most likely just takes 3 days, but because you have to request data from 2 diffferent sources and have 4 progress check in meeting where you have to have a up to date flowchart and progress report, it will take a week.
You can google it but basically the idea is to estimate in complexity and not in time. It is much easier to say something is a complexity of 5 or 3, and then managers figure out how much of these tasks will be done in the next week, based on how fast the team worked in the past.
So the point of story point etc is to ensure there isnt underestimating time, but if you minimise time perfectly on every task so that every minute is accurate you will just get burnt out employees who will either leave or unable to meet deadlines. Working with people not bots.
I had a Jared in my first IT workplace and he only wasn't fired cause he had some papers that proved he had some disability, which was beneficial to the employer... (I live in eU)
It's so weird when profit driven people try to conceptualize complex tasks. Like, things take as long as they take, do you want the feature or not? Anyone who has ever spent any amount of time programming knows that any given task can suddenly take 10 times longer to code.
"How long will this take" "Well it will take a week so adding in the Bob Tax thats two weeks. If we remove to from the equation we can take it down to one week."
Mangalers like this is why I subscribed to the Scotty theory. If you say it's 5 days, it's 5 days no ifs, no work harder to make the unrealistic timeframe they suddenly decide to force through. And for whatever reason it's "not good enough" If it's 5 days it's going to take 15, minimum. And then they will try to negotiate because sales wants to look good, the customer was told by "someone", Jesus' second coming; the project needs to be done in 8 days...
"Well then, i correct my previous estimate to a XL Tshirt-Size or 10 Story Points and lets meet in the middle at a solid 7 Story-Points this week and a solid 2 Story Points next week for any incoming disturbances like unneccessary PM Meetings"
the terrible thing about this is how it forces people to overestimate. if you think it'll take a week then theyll make it 3 days. so you have to lie and say 2 weeks just so they give you an adequate time
Depends on where you are. Largest was 1 point = 1 days work for one person. Smallest 1 point = 1 hour for one person. You usually account for uncertainty as well. So if you think it will take 2 days, but you are uncertain you give it 4 points.
It's supposed to be a rough indicator if size/effort/complexity of a task for planning reason and not specifically an amount of time. it's supposed to be decided by the team and not by "management". So management says "we need the software to do A,B,C, and D, in the next release" And the developers say "OK, we'll take a look at those" then they have a meeting, break down those features into User Stories. So a story could be "the user can save the file by clicking a save button or pressing Ctrl+S" and the team can say, yeah, look, we've got to sort the menu system out, so that on it's own is a Zero story points, but the way we're handling the menus is going to be "n" You break down all the features and tasks and stuff into small tasks, and give them numbers to show how much effort they are, so that you can decide who does what. Generally Management want numbers like "This will take 48 man hours" but you can't have 48 people crank out that task in one hour. Some tasks can be split or shared, and you could split it into 4 tasks that each take 15 hours. BuT tHaT's 60 hOuRs!!! Yeah, there's an overhead for those people to coordinate on matching up their parts of the task! You have to have meetings and decide how different parts of the code will mesh together. The problem is that management want to treat story points as an indicator that they can just slap onto any task, when by design they are relative to other stories, and are assigned internally by the team during the planning phases of a project. The people who should be estimating who complex a task is, are the people who have the skills to do the task, and are actively doing the tasks. If the dev says "this task is twice as much work as that task" a manager can't really make a solid judgment and say "nah, I'm pretty sure you can bang them out in the same amount of time".
Refactoring shouldn't be on the backlog though. Whether or not you should refactor is a development question, not anything the PO, PM or scrum master should be concerned about.
It is a real thing in the tech world. Depending on which company or organization you are in 1 story point could be 1 hour or 1 working day and you only use fibonacci numbers 1,2,3,5,8 and so on to represent how long a task gonna take. Its kind of dumb tbh but for some reason the tech world project managers/product owners who are graduates Scrum Agile University of brain dead bullshit and can't write a single line of code to save their own life fucking loves this shit.
Yes, unfortunately. The reality of building software is it's not like building a car. You're not doing the same things over and over again and you can't measure how close it is to being done; software is done when it's done. That means you can't say "Oh, one program will take us one week, so three programs will take us three weeks." However, most business leaders can only conceive of projects in terms of "how long will this take?" because the more time something takes, the more it will cost, and the greater the risk that the company will miss out on The Next Big Thing. Story points were originally created to bridge this gap: instead of measuring time, they abstractly measure complexity. Your project manager says "We have a deadline, how long will this take?" and instead of pulling a number out of your ass that stands a high chance of being wrong, you say "Oh, it's X story points", which satisfies the PM's need to have charts and graphs and measurements to demonstrate progress. However, business still only conceive of things in terms of time, so story points get converted back into measures of time, which makes the whole thing useless. Management starts "budgeting" story points to get things to fit into deadlines, which you inevitably fail to meet because _software is done when it's done_. There is an entire cottage industry of "Agile coaches" and "Scrum masters" out there who charge companies hundreds of thousands of dollars to train their staff on "the best process to meet software deadlines", which leads to the kinds of meetings Anthony is demonstrating. In reality, the best process is the one that works for your team and everything else is waste. Given the amount of fighting they do over it, story points look to be the wrong method for this team. Everyone seems pretty independent, so a simple task list ordered by priority would probably work better. Instead of calculating story points, Cindy's job would be to work with both the customer and the developers (Jared and Anthony) to figure out what features are important, what tasks are necessary to deploy those features, and manage the task list so it's always up-to-date with progress and feedback.
