Is there a need for tools for sig campaign management?

[gentlemand]

Not that I will ever manage a campaign, but were I do to so there's no way I could face it without some automation on side.

Some campaigns must have 100 posters doing 30-50 posts a week. That's thousands of posts to count and skim. My eyes would roll back in my head I'd puke my own pelvis up on the first attempt. The idea of having to do it all again the following week would fully break me.

Kudos to those who do do it.

If it were largely automated you could still dip into every few posts manually to make sure they weren't posting complete and utter bilge.

[1715697178]

1715697178

[1715697178]

1715697178

[1715697178]

1715697178

[shield132]

To me, yes we actually need some tools to save a lot of time and efforts managers take while managing a campaign's spreadsheet. There should be a date-specific column setting that must allow the bot to check the posts based on total posts made during a specified date (week) range so to save users from being provided false numbers due to deleted posts and decrease in total number of posts of a user because of the same. I was once a part of a campaign that used bot to determine post number, but it had this issue of starting count with the previous week's last end number and even if a few posts get deleted, bot used to show the same number as starting post count for which users had to make the remaining amount of posts to first make their count reach the count the bot used to show.

I think if someone creates such software and publishes it there, since effort of campaign managers will decrease, their salary will decrease too and possibly there will be less demand on them. But on another hand why to publish it in public? I think orders may come from campaign managers or if any of them knows programming, they'll do it for themselves and lessen their time spent on campaign management activity.
Again, I think it's not a big deal without great algorithm, bitmixer had money and etc to do that but they chose simple bot and Lauda as manager, he was needed.

[timerland]

I don't think bots will ever replace managers in this industry though. Bots are unable to check post quality.

I've seen several campaigns using automation - Yobit,Bitmixer, Coinroll as far as i remember but they do still add up managers on it.

Yep. Bitmixer is a good example when they tried to be fully automated, it was a massive mess and any regular shitposter was able to run rampant and spam posts and get paid. When Lauda came in and helped out with management, all of the shitposters were booted, even with their bot being able to detect the characters and what sections the posts were made in.

If it were largely automated you could still dip into every few posts manually to make sure they weren't posting complete and utter bilge.

I don't think that'll be possible, even if you check every 1 in 5 posts, because they'd be low content posts getting through. It'll be very hard to manage a campaign, even with automation, because you'd need to read every single posts, which I'd reckon would be over 20,000 posts a week, all read and counted in a single day. It's crazy.

[gentlemand]

I don't think that'll be possible, even if you check every 1 in 5 posts, because they'd be low content posts getting through. It'll be very hard to manage a campaign, even with automation, because you'd need to read every single posts, which I'd reckon would be over 20,000 posts a week, all read and counted in a single day. It's crazy.

It depends on how carefully you select your members in the first place. If you have a bunch of old hands that you're familiar with it's very unlikely they're all going to have a funny turn and start posting risible junk. It's a different matter if it's something like a Yobit campaign of course.

The more effort you put into initial selection I presume the easier actually running it is. Rather them than me all the same.



 

[timerland]

I don't think that'll be possible, even if you check every 1 in 5 posts, because they'd be low content posts getting through. It'll be very hard to manage a campaign, even with automation, because you'd need to read every single posts, which I'd reckon would be over 20,000 posts a week, all read and counted in a single day. It's crazy.

It depends on how carefully you select your members in the first place. If you have a bunch of old hands that you're familiar with it's very unlikely they're all going to have a funny turn and start posting risible junk. It's a different matter if it's something like a Yobit campaign of course.

The more effort you put into initial selection I presume the easier actually running it is. Rather them than me all the same.
 

Ah, I can agree with that. What could be a possible new way for signature management would be getting all the interested users to sign up via the web application (similar to bitmixer's one), and then for the manager to selectively accept and decline users via the application. Alerted users would be automatically PMed and they would be able to utilize the portal throughout the week to check how many of their posts are qualified from a basic standpoint (above certain characters, in a certain section).

I could see this working, and then the manager only checking posts ever so often. For example, a lot of high paying campaigns probably do not need a manager due to their post quality and how used they are to doing everything correctly.

[hatshepsut93]

...

That's kinda what I'm talking about, most sig campaigns have requirements like not paying for certain boards, minimum char counts, limited languages and so on, it can easily be parsed automatically and presented either as a simple totalPost/qualifiedPosts report, or go into detail like board structure of posts. Then the manager simply needs to quickly glance at post history and if necessary substract any posts that are of low quality.

As for automatically scoring quality, it might be possible to develop tool that probabilistically do it with machine learning. In fact, I was even working on that recently, but I dropped the project because a big dataset is required and I'm too lazy to prepare it.

[Stedsm]

--snip--

I think if someone creates such software and publishes it there, since effort of campaign managers will decrease, their salary will decrease too and possibly there will be less demand on them.

That's wrong because in the end, it's all campaign manager's job to effectively put things into consideration and watch out for quality and negate any posts that doesn't suit the campaign's standards. So, the manager still deserves what they are being paid. If everything could be considered correct for automated counts and everything, then why didn't Yoshit campaign become a successful story here?

But on another hand why to publish it in public? I think orders may come from campaign managers or if any of them knows programming, they'll do it for themselves and lessen their time spent on campaign management activity.
Again, I think it's not a big deal without great algorithm, bitmixer had money and etc to do that but they chose simple bot and Lauda as manager, he was needed.

Something I'd agree with, because this is something that should be taken fare of as it's totally somebody's talent they are showcasing through making such program that'd do almost everything a manager does (except the quality watch part).

...

That's kinda what I'm talking about, most sig campaigns have requirements like not paying for certain boards, minimum char counts, limited languages and so on, it can easily be parsed automatically and presented either as a simple totalPost/qualifiedPosts report, or go into detail like board structure of posts. Then the manager simply needs to quickly glance at post history and if necessary substract any posts that are of low quality.

As for automatically scoring quality, it might be possible to develop tool that probabilistically do it with machine learning. In fact, I was even working on that recently, but I dropped the project because a big dataset is required and I'm too lazy to prepare it.

I believe I've seen some spreadsheets with integrated automation which isn't fully automated but the manager used sums and subtractions in specific places to determine exact post counts and payable amounts in BTC which is quite interesting.

[Patatas]

I think a lot of tasks in signature campaign management can be automated, it's pretty easy to write a program that counts posts in given timeframe, checks in which boards they were made, checks that length was sufficient and so on. Of course it can't measure quality, so managers would still need to look at posts themselves, but it eliminates the need for counting, which is long and error-prone.

All of that can be really done manually without having software to do it. The biggest challenge is checking the quality of the post which is nearly impossible so there is no point in putting efforts. There have been tools like that in the past and no reputed manager has used it. Counting posts and length is pretty fine.

I know there are many services for bounty campaigns, and there was some service for automated sig campaign management in the past, but still I would want to know if there's a need for such tools today, because I could develop one if someone asked for it.

If you can dig into the right threads, you might find such tools used in the past.