effort estimation for implementing ServiceNow chat with Virtual Agent.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
08-29-2022 06:12 AM
Hi All,
I wanted to reach out to find out if anyone can provide effort estimation for implementing ServiceNow chat with Virtual Agent.
Appreciate your inputs.
Thanks
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
08-29-2022 06:56 AM
"How long is a rope?"
Answer: Depends what you're going to use it for.
Implementing Virtual Agent is much like implementing flow (managing conversation flow).
There are very VERY simple VA flows.
There are very VERY complicated VA flows.
The more detailed the use case you provide the better the estimate. With no context, I'd say "weeks".
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
08-30-2022 12:24 AM
Hi Robert,
Thanks for the response. This is helpful.
However One New Scenario: If we need to implement ServiceNow Agent Chat without virtual agent, then in that case what will be the efforts. Plugins are already installed and have plans to keep the implementation OOB as much as we can.
Thanks
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
08-31-2022 04:32 AM
"what does OOB as we can mean"?
I don't ask that flippantly either. Think about it. If its OOB then you just turn it on. You don't change or manipulate it. You can do it right now.
But if you need to build a scope, or have internal pressure to understand LOE, here's how I'd break it down:
1) Gather Requirements: Do the stakeholders understand how this app works enough that they can state what they want out of it? If not, how many meetings do you think it will take to get them there?
2) Build: This *could* be as simple as "turn it on and watch it go", or you could have work from (1) built in. If you're flying blind and you'll be doing the work (and ONLY that work), I'd say give it 1 week.
3) UAT: I hate calling it that, because I feel its more "does this work the way we thought it would for you?" Need to give people time to do that. Will they drop what they're doing and do it instantly?:
4) Conditional Build: You should plan for stuff coming out of UAT. Will you address it then or roll it into some kind of backlog
5) Training & OCM: People are going to need to know this feature exists and how to use it. Beyond just "make a video to show how" you'll need to market this to the consumers. How many of them are there? Multiple languages?
6) Documentation: You are documenting your solutions, right?
After you go through that exercise you may find the estimate was larger than you'd hoped, BUT, being certain takes time!