I’m a business analyst, and a big part of my job involves working with engineers and product managers to gather detailed, in-depth information. For reasons I don’t fully understand (though I have my theories), I often find that engineers, in particular, seem oddly reluctant to share the information I need. This makes the process more challenging than I’d like. Does anyone have tips or tricks for building trust with engineers to encourage them to share information more willingly and quickly?
EDIT: Here’s a summary with more details for those who requested more info: I’m working on optimizing processes related to our in-house file ingestion system, which we’ve been piecing together over time to handle tasks it wasn’t originally designed for. The system works well enough now, but it’s still very much a MacGyver setup—duct tape and dental floss holding things together. We got through crunch time with it, but now the goal is to refine and smooth everything out into a process that’s efficient, clear, and easy for everyone to follow.
Part of this involves getting all the disparate systems and communication silos talking to each other in a unified way—JIRA is going to be the hub for that. My job is to make sure that the entire pipeline—from ticket creation, to file ingestion, to processing and output—is documented thoroughly (but not pedantically) and that all teams involved understand what’s required of them and why.
Where I’m running into challenges is in gathering the nitty-gritty technical details from engineers. I need to understand how their processes work today, how they’ve solved past issues, and what they think would make things better in an ideal world. But I think there’s some hesitation because they’re worried about “incriminating” themselves or having mistakes come back to haunt them.
I’ve tried to make it clear that I’m not interested in punishing anyone for past decisions or mistakes—on the contrary, I want to learn from them to create a better process moving forward. My goal is to collaborate and make their jobs easier, not harder, but I think building trust and comfort will take more time.
If anyone has strategies for improving communication with engineers—especially around getting them to open up about technical details without fear—I am all ears.
Accept “I have no idea” as an answer, and don’t use it as an opportunity to push things in the direction you want.
learn to account for people being wrong, and don’t punish them for it.Engineers want to be accurate. They don’t want to give answers that they’re unsure about or just speculating.
Early in their careers they’re often willing to, but that gets beaten out of them pretty quickly by people with deadlines. Expressing uncertainty often means the person interprets the answer in the direction they want, and then holds the engineer to that answer.
“It could be anywhere from 2-8 months I think, but we won’t know until we’re further into the design phase” is taken as 2 months, planned around, and then crunch Time starts when it starts to go over. Or revising an estimate once new information or changing requirements are revealed is treated as incompetence, even though more work taking more time is expected.It’s in the self interest of the engineer to be cagey. “I don’t like to give estimates this early” is much harder to turn into a solid commitment than an earnest best estimate given the current known state of the project.
Similar for resources required or processes. Anything you don’t say is unlikely to be held against you.
This is brilliant. I often suspected they did not want to “incriminate” themselves, and I have tried assuring them that that is in no way what I am about. I am looking to optimize processes, and I am very eager for their ideas on what would work better than what we’ve been doing.
Verbal assurances mean little to many. At least put it in an email. Otherwise CYA dominates.
And remember the docco kicks around forever. Indemnification until retirement is impossible to ensure.
And, as our RedHat TAM rediscovers, we can bring that shit out of the archive to prove a point … or to heckle about overblown systemd promises, but that’s PTSD for another venue.
I’m an engineer for past 20 years ,and the moment I get a whiff of some biz person fluffing bullshit I check out. Not saying you’re like that but something to be mindful of
So be real, honest, straightforward, put in effort into understanding the technicals so youre not just a sales annoyance or engineers will write you off right away IMO
Also I don’t think some people realize how busy or hard engineering can be. I’ve been working my ass off on a ground up new product and a lot of stuff just falls to the wayside out of lack of time
There’s also !askelectronics@discuss.tchncs.de !engineering@sh.itjust.works among others
A million times this.
That last paragraph hits home, but in a sad way for me. I spent most of the last year working on a new project to streamline one of the biggest time sinks we have, and as we’re coming up on having an MVP ready to start beta testing, my org just dumped the entire team other than 1 guy. So I lost the guy who was my peer/dba on the project, and the dude who knows how to run the driver software.
Going to try to see if we can salvage what we made since it’s still needed, but fuck that wrecks a ton of time and effort. And really sucks cuz my team had to pick up the slack while I was trying to get this working, and we don’t even have anything to show for it…
Thank you!!! In fact I have been emphasizing that I need to know the technicals, and that they should not worry about getting too detailed, because I need a very thorough understanding so I can best come up with a process for how they do file ingestion (which mostly is up to them) but then also how the information gets to them, and how they output the data, in the best, most thoroughly documented (without being uselessly pedantic) way possible. Which is pretty much going to be getting everything into JIRA, and eliminating all the uselessly disparate systems that people are trying to stitch together currently. I need to make sure all the teams along the road of this process are communicating with each other, and at are at least having a basic understanding of the whys behind what is required of the process. And of course that it is efficient and fast and definitely not cumbersome.
In addition to what other people have said, “the technicals” and “getting too detailed” is ridiculously vague. There is always a too detailed. Software engineering is a giant world and engineers specialize, and even other engineers with a slightly different specialty don’t really want to know all your “technicals”.
Be specific about what details you’re interested in and why if you want to build trust. Demonstrate tangible investment in figuring out what your gaps are and ask specific questions, and be clear about what kind of answers you want. “Thorough understanding” is not helpful.
There is always a too detailed. Software engineering is a giant world and engineers specialize, and even other engineers with a slightly different specialty don’t really want to know all your “technicals”.
100%. I regularly have to tell DB or App people that I only need the high level answer because the details are immaterial to the discussion at hand. I might be personally interested, but I got shit to do…
there’s a few things here that trigger red flags for me:
not worry about getting too detailed
oh good! because it’s probably ill-defined and nobody really knows and figuring that out involves a lot of reading other people’s shit code and we have work to do
because I need a very thorough understanding
oh you mean do worry - you want to know exactly how it works… sorry bud, no time, that’s a lot of energy
thoroughly documented
ugghghh documentation is for people that don’t understand that documentation is out of date the second you write it: don’t drag me into your futile attempt to make a static artifact that i’ll need to maintain in the future when i update a living system
eliminating all the uselessly disparate systems that people are trying to stitch together currently
okay but that’s really dismissive… this is work that people have put in - even if it’s shit and everyone knows it’s shit, it’s disheartening to have things thrown out… and what they do now they know how it works, they know the caveats… you’re talking about coming in, getting a cursory understanding (what you think you can understand everyone’s problem when the people that built the thing don’t have the full picture?) and then planning out and telling them what to do
if you want help from engineers, ask them for help to build a new thing: don’t ask them for help to explain something so you can tell them what to build. we’re creative, and we love solving problems and we hate robotically implementing someone else’s vision
Have you stopped to consider that the current solution might be better than an all JIRA one? I can definitely see a lot of “file ingestion” pipelines that would be much better handled by a bunch of different systems intertwined than JIRA, especially for automated file ingestion (which I guess is what you’re doing? Hard to know but hard for it to be something else).
I don’t know what’s the situation there, but if I was an engineer on one such project I would explain to the person why it’s not feasible, but it could be that that got interpreted as not explaining stuff. An easy to understand example would be someone asking what’s the best way to replace a car (that has been cobbled to pieces from separate cars) with a shoe, and then you try to explain to the person that that just doesn’t make sense they say that you’re being uncooperative and not explaining how the car works so you can make the shoe do the same.
Look, I’m not saying this is your case, but it feels like you’re approaching this the wrong way, you have a new solution without even understanding the current one. A better approach would be to gather the engineers and ask them what are THEIR problems with the system, and how would THEY like to fix them. If the Jira thing comes from higher ups tell them that this is a new requirement, but let THEM solve the technical issues, you are unlikely to be able to even if they explain them to you in detail.
Ok, so I AM asking the engineers that. But we need to be angle to track the implementations for each client, and that is why we are using JIRA. I’m very open to alternative ideas, but one of the problems is that teams involved are using salesforce, JIRA, and Azure, and this is causing a serious dearth of communication, as well as there being no way for anyone to get a bird’s eye view of where implementations currently stand. Frankly I have a lot more autonomy and control over this thing than anyone in this thread seems to be used to, so if people have better process ideas, I am all ears.
Ok, so it seems that you’re only talking about managerial stuff, whether to use Jira or Atlassian, engineers don’t usually care about that, and there’s usually no technical reason to use one or the other, so it could be that you’re asking them to explain how a car works to try to figure out the best shoe to wear.
Also no one that’s not involved in the project will have better process ideas because we don’t know the process, and apparently neither do you from what you’re saying, it’s the thing the engineers “refuse” to explain to you. I think at the end of the day you need to sit down and explain what the higher ups want and listen to their ideas on how to get there.
explain what the higher ups want and listen to their ideas on how to get there
absolutely this… engineers want to help you design solutions… if you come to me and ask me to explain something so that you can design a solution that i’m then told to build you’re removing all the fun out of it and i ain’t gonna help with that
JIRA, and Atlassian
Since one makes the other, it’s like saying “Rav4 and Toyota”. I’m assuming you mean Confluence (aka Flatulence; or Confluenza for the stress-based sickness from watching the spinning please-wait-web-loading symbol too much).
using salesforce, JIRA, and [Confluence]
Quit it.
No one uses those tools; they get used by the tools. They’re slow, cloud-based, usually under-budgeted for the great gobby java blob-wrangling clown shows they are.
If you’re asking deadline-driven engineers to use a slow cloud app like that, with their caffeine levels, and invite real feedback without fear of penalty, you’re going to get some interesting opinions. Please, try to find something more usable for the non-deadline-helping work-tangents like documentation; which are important, but not on-fire-important like every other bloody thing on the list.
Sorry, meant Azure
If you are saying things like ‘I’m not interested in punishing anyone for past decisions or mistakes’, I think I can see the problem.
I’m not interested in punishing anyone for past decisions or mistakes
right so you think our decisions were mistakes rather than at best suboptimal but reasonable given the information at the time? how arrogant
deleted by creator
deleted by creator
deleted by creator
My experience is you get the best response if they understand why you need the information and at what level of detail. They seem to respond well to clarity, organization and logic (who doesn’t!), so prepare your communications to include the background they need (how does your request help them in the long run), what it is you need from them (and in what format), and when you need it by. Trust is built by demonstrating your value to them. Think about ways you can help them get the info to you (start the work for them, book time on their calendars to focus on the request, sit with them and help them produce the info).
Side note: engineers sometimes offer information that is not executive ready - you will either need to translate or tell the engineer who the audience is for the information.
You have perfectly put into words what I have been attempting, and not really succeeding at. I should present my needs in a clear and logical, documented manner. Brilliant!
I can’t tell if you are being facetious, but if not feel free to message me for more specific ideas.
Yeah, text communications are constantly problematic for these reasons. I was being 100% serious.
❤️
it occurred to me, if you have the ability to communicate with the engineers in person or on a video call, you could (maybe) avoid the pitfalls of text only. Anywho, don’t hesitate to reach out if you would like more detailed support. 😃
Frankly, it’s tiresome trying to describe technical details with business analysts who glaze over something you’re passionate about, treating it like nerdsprak. If the engineer has spent any amount of time producing a solution, you can bet he’s passionate and invested. Give credit where credit is due and don’t sound like an obnoxious condescending douchbag when doing so. People can tell when a disinterested person is giving fake praise. It’s quite different when a crowd of peers is giving recognition of a job well done. And no, you’re probably not as smart as they are in their field of expertise.
Also, listen to their input. They don’t want a product with their name going live with a feature the bean counters want, but the engineers know make the product worse. It’s like a mom watching your daughter to go to prom with a cheap haircut because dad as too cheap to fork out for a perm. You know what I mean.
As an engineer you learn to be very careful about what you say to non engineers.
A trivial example.
What if we make change x?
It’ll make some things harder and some things easier.
One week later.
Why are you having problems? You said doing x would make things easier.
More complicated example.
Can this be used for real time control?
Define real time.
Just answer the question.
I can’t it’s a bad question. I need to know what you are trying to control.
As a dev, this definitely triggered me a little.
Real life mechE here. I’ll tell you how my brain works.
99% of the time when I get an odd request from outside of the department, it goes one of many ways:- the request is literally not in my scope of work and I let them sit for a day or two and then politely deny with a CC to my manager.
- the request is so vaguely worded that I could give a 2 sentence answer or a 20 page pdf answer or a PowerPoint full of flowcharts, and all would be “right”, leaving me in a state of decision paralysis and needing clarification.
- the request is something I can help with but I don’t know your technical capability levels, so I try to keep it very generic and high level as to not simply knock you over with a technical dictionary.
- the request is in my scope of work and very doable, but I do not want to inadvertently share information that I may not be allowed to divulge freely to other parts of the company.
And of course, there’s a lot of CYA reluctance too depending on what’s being asked.
If you’re asking first or second level engineers things like “how does your technical work flow do it’s thing?” you are starting at the wrong level for a documentation project of this massive scope. Engineers have managers whose job is to translate requests into technical terms and figure out who is the best at doing what. That’s what mine does: he takes a super weirdly worded ECR (engineering change request) and translates them into technical steps and clear direction for me. Then I can pick out the details needed to make it happen, confirm them, and document them.
You need to define clear needs out of your request: start with your end goal, the processes you need, the mechanical details of the processes you need to write, how much detail you are comfortable with, and the format in which you want it . and take all of that to the senior or director level of whatever department manages those systems. They may or may not know the exact information you need, but it should be their job to delegate and translate the request such that their reports can collate what you need in the form that you need it. And because it’s the director delegating, the engineers have inherent CYA and will be a lot more comfortable giving you what you need.
I relate to this style more than the other comments in this thread, this seems more typical of a large company.
You need to define clear needs out of your request: start with your end goal, the processes you need, the mechanical details of the processes you need to write, how much detail you are comfortable with, and the format in which you want it . and take all of that to the senior or director level of whatever department manages those systems. They may or may not know the exact information you need, but it should be their job to delegate and translate the request such that their reports can collate what you need in the form that you need it. And because it’s the director delegating, the engineers have inherent CYA and will be a lot more comfortable giving you what you need.
Unfortunately this adds to the bureaucracy, but it really is the most effictive way of translating business needs to engineering needs. It’s not a straightforward process, and accurately defining the steps that need to happen for a job to get done, takes someone with a lot of experience and training.
If you’re in a startup or smaller company, then I think the other comments that prioritize asking and listing to what the engineers recommend, is the best approach.
deleted by creator
This is why documentation of business process and methods is so important. A lot of time, the engineer solves seemingly small problems without oversight, so imagine a decades old collection of many innocuous solutions leading to the whole ‘dunno what this does’. If it’s important enough to commit to a mission critical system, it’s important enough to document.
Also, it’s incredibly frustrating for an engineer to be given a one line brief, work his ass off producing the solution, then have the business analyst take credit for the work, and not bother to even learn how the system works, even at a high level. It sows distrust and disdain.
This is why documentation of business process and methods is so important
absolutely not… if you think that random code from 10 years ago is difficult to figure out if it’s still needed, try that with documentation!
IT systems are living, dynamic beasts… they should be built in such a way that makes them comprehensible with relatively minimal effort, on their own, because the code is the only source of truth and everything else may as well be a lie
Lol. You’re highlighting ops problem exactly. Oversimplification of the issue and delegation of the documentation problem to the engineering department is the exact reason people there feel resentment. It’s simply not their job. As the other commenter posted - the system spans multiple disciplines and workflows, yet it seems only the engineer is tasked with understanding it all, in order to build the system. Consultants register this as a risk, and management assigns this to engineering because ‘only they understand the code’ - is exactly the problem op is facing.
The system is the property of the company. The company’s language should be used to capture it’s design, function and intent (what it does) versus how it is done (it’s expression in code). There’s a reason they call it ‘living documentation’ - it,.and the company’s understanding, should evolve along with the code.
Edit: are you seriously letting ‘random’ segments into your code? I think I found your problem…
Tracking down why all that junk exists and if its still required can take a staggering amount of time. Trying to juggle that with your day to day is…not practical.
Yes. Please deep-dive into it all and then schedule a long, slow sit-down, regarding the Morton mod, but also be prepared to justify why the Penske Project is behind the arbitrary and impossible schedule some DeVry grad has already set for you.
(I suspect OP will find a lot of “in what fucking time?!?” concerns when it comes to knowledge ‘synch’ or documenting, since neither of those are billable endeavours and no one wants a deadline for the next project crushed because of the prep and meetings justifying the time over-run for the last project)
This is…very interesting
After reading your replies, I am on edge.
Please consider the following questions.
What is the power dynamic?
Are there good reasons to stonewall you?What happened to the first few teams you worked with? Did the engineers involved advance in their careers? Do they talk with you still? What about their prior interactions with your team and department? Do those engineers still work at the company?
If you are confident you are there to help then just speaking to them like people. Don’t bullshit them. Push them up in their careers when you can. Get them what resources you can. Support them in their goals. Do a good job and you won’t get them to shut up.
This is a very long story, from my previous role, but both of those engineers have been promoted, and we are good friends.
You’ve probably tried this, but did those engineers have any insight as to how you can communicate with other engineers? Were things easy with them from the start or did you need to “prove yourself”?
This had not even occurred to me. Now On my to do list for Monday.
I was going to say, if you’re good friends how do they not have advice for you?
Yeah, I hadn’t even mentioned it to them. Also they are in Bosnia, and almost all of us work remote, so it’s not like I get to see them around the office. They are not in any of the engineering teams I oversee, and I’m relatively new to this role. I’m very excited to get their take on this though; it’s a great idea.
I’ve worked very closely with engineers and I’m engineering adjacent myself. Most of the highly technical types I know in every field (myself included) struggle to talk to people about their job because they no longer know what normal people do or don’t know and they don’t want to come across as condecendong. Like for me the basic refrigeration cycle feels like something everyone should know but I logically know that actually isn’t the case and at the same time I don’t know where the laymans actual knowledge on the topic begins. Like do I need to start with explaining that boiling liquids remove heat? Do I need to start with what boiling even is? Do normal people even know that things boil at different temps at different pressures? If I start explaining any of this are they jist going to look at me like I’m an ass and say “Of course I know how thermodynamics works”? Eventually I just decide it’s better to not to talk to them.
At the same time though, if you do manage to break the ice with them then you are more likely to sucessfully get a passionate stream of consiousness rant from them because they’re passionate and now they know that you can be trusted not to see them as being condescending when they overexplain. Honestly the best way I’ve found to break the ice with technical types is to get them to start complaining about some part of their job. That also sounds like exactly what you’re looking for if you’re trying to make their jobs easier. But if they start seeing you as someone who it is safe to complain to then they will start seeing ypu as someone it is safe to talk to about other things.
Also as always there is a relevant XKCD.
I am the wife of a mechanical engineer, who’s brothers are mechanical and electrical engineers, who’s parents are electrical engineers, who’s best friends are aerospace engineers.
Basically I married into a family of robots, and I agree with this commenter here.
This is the crux of why senior engineers struggle to talk about work I think, and I find the best way for me to get them talking, is to try to learn something small about their work, enough that I can ask intelligent questions, and then listen carefully to the replies.
After a while they open up and I get to listen to the best rants about “special metals” or “systems architecture” or “braking systems in the railway”. It’s awesome.
It’s how I connect with my husband.
The other wives stand in a circle and roll their eyes about them talking about work because they don’t understand anything. “Oh there they go, talking about work again.”
I decided I didn’t want that to be me, and told myself I would listen when they were talking, listen when my husband was working from home. Learn to ask intelligent questions about his work, and eventually, I knew what he was talking about.
Enough that I now freelance in condition monitoring, giving me yet another way to connect with him.
Ask intelligent questions, get excited about the replies, encourage them so they know you won’t be insulted when they assume you don’t know about <speciality subject> and you will have them opening up in no time.
This is a really sweet comment that’s brightened my day while also being practical advice
You should always listen to your significant other. Of all the people in the world, they chose you to talk to
Of course I know how thermo dynamics works! But uh if you could just explain it for my friend here, gestures in general direction of dog, that would be perfect.
Talk less, listen more.
They’re probably (no offense) nerds, so let them nerd out and listen to them.
Then actually act on what they say, and soon they’re be telling you more shit than you want to know.
My husband is an engineer. Screaming, “what the fuck are talking about?” is probably not the way.
This post is a little too vague to give real advice. You don’t tell us what industry you’re in. You don’t tell us if the engineers are the end users of the software or processes you’re working on, or if they will implement the software or processes you’re working on.
If they’re the end users, they might be concerned that the changes you’re designing are going to make their jobs harder. A lot of changes in the past couple decades aimed at “efficiency” have involved making people take on more work for no additional pay, then firing the administrative staff or other engineers who used to do that work. Even if that isn’t the sort of project you’re working on they are reasonably wary based on past experience. Or maybe it’s not clear to you how this will make their life harder but management will find a way.
If the engineers are writing the software that you are helping design, how are you helping to make their jobs easier and more fulfilling? It’s an unfortunate fact that software engineers are sometimes treated like misbehaving vending machines that will produce software if you force them to. If they are writing the code, there’s a very good chance that they know more about this process than anyone else in the room, but are they treated like they know more than anyone else in the room? Is their expertise valued or are they treated like roadblocks when they give their expert opinions?
Was trying to compose a similar statement on that lack of details. Like, my background is scrum/ agile software development and if a random BA called me up out of the blue for project details, my first response is going to be “I’m busy, talk to my scrum master and/or manager” and failing that it’s likely going to be the minimum amount of information required to get said BA to leave me alone so that I can get back to work. Plus, unless I know that my audience has the technical capacity for low level details, I tend to leave them out (I don’t mind answering questions, but I also don’t have time in my life to spout information that’s going to go in one ear and out the other).
This is extremely insightful. Thank you. To keep it somewhat vague, I am trying to optimize processes surrounding file ingestion. And I am trying to eliminate all the roadblocks caused by siloing of information. We have an in house file ingestion “engine” if you will, and we have really been rebuilding it from the ground up because its original function was not what we are using it for. So there are problems. To date, we have be MacGyvering the fuck out of everything with a pen knife and some dental floss, but we got through crunch time, and now we need it to be smooth, and by the numbers. Easy and clear for everyone.
Well that might explain some things.
Not to throw shade at your company but that process is so backwards that it’s no wonder the engineers are sparse on the details. I saw another comment likening software development to a crossword puzzle, which is a pretty good analogy. To further it, changing software once it’s done is like trying to swap out a clue/ word once the rest of the puzzle is built. It’s theoretically possible, but depending on how the puzzle is designed, it can range from an absurd amount of work to nearly impossible. Given the way you’ve described the state of things, your engineers are probably low on goodwill to boot.
I’ve worked on cobbled-together crunch-time hell-projects and the last thing I’d want after getting free would be a random BA coming to me about details that more than likely packed with the project PTSD and would very much like to forget. Doubly so if it’s issues that I bought up early in the design/ development process (when they would have been comparatively easy to fix) and was dismissed by the powers that be. I can only speak for myself, but I can only take so much “that’s not a priority”, “we don’t have time for that”/ “we’ll see if that becomes a problem in the future and deal with it then” before I throw in the towel, stop keeping track of everything that’s wrong, and just bin the entire project as dumper fire run by people who would rather check boxes than make things better.
Anecdote from my first job (software engineering): New manager wants to know what our team does and how our process and software works. Like, he really really wants to know it!
Okay, I book a timeslot and prepare some slides and an example; we have a meeting. I go over the high level stuff, getting more and more specific. (Each person on our team was responsible for end-to-end developing bootloaders for embedded HW.) When I got to the SW update process and what bit patterns the memory needs to have and how the packets of data are transmitted, he called off the meeting and I’ve never seen him since.
I guess, he didn’t want to know THAT much after all.
Fair enough. But I actually do want to know that stuff, and it’s not over my head.
Keep your promises and tell the truth. If you don’t keep your promises, be the first to acknowledge the failure.
I was an engineer for a long time and among my peers the problem we had with management was often that they had a slippery relationship with the truth.
Also, demonstrate forgiveness within the organization for technical mistakes. If your engineers don’t want to share the bad decisions they’ve made, look for aspects of your company culture that punish people who admit mistakes.
One example would be times when someone spoke about a mistake they made and then was relieved of responsibility because of it. That’s an example of punishing the admission of a technical mistake.
