Topic: Campaign managers: If I were to make a post tracking service... (Read 379 times)

Probably and not to mention that they have a bigger case of alt accounts with such a low rank requirement to join the campaign plus most of the people who joins their bounties are low quality or shitposters. If I were a campaign manager probably a beta test will help.

I think you would have had more luck trying to pitch this sort of service in 2018 when there was more forum activity: https://bitcointalk.org/adrotate.php?adstats

Or even earlier when similar tools (already mentioned) were being developed/used.

Okay so thing is that you won't just have to pay for zapier you'd also need to shell out on domain, hosting services etc. So if you ask me, it's not worth spending so much on a campaign management bot/service. However, there's room for potential-> I would advice you to use this enthusiasm for building something that bitcointalk actually reaallly needs(like how Vod has bpip and loyce is ai). There's room for a lot of things, it's upto you really on what happens next.

It doesn't even have to be completely related to btt either think how you could use your skills in the crypto market. The potential is endless honestly...

True indeed. I have been offered for a bounty campaign management which were offering me $200 a week for all the social media campaign along with signature. I haven’t managed a bounty campaign yet (only 2 signature campaign managed), but I'm pretty sure even $500 per week for such a campaign is pretty low if you are doing your job perfectly. Result, I had to deny the offer. And they picked a random guy to manage the campaign lol.

That would be good idea but you have to keep some standard for accepting campaign. Otherwise, bounty campaign manager will be using the service but never think of donating  If someone got a handsome amount from campaign management (in fact if the campaign is bigger), they would be interested in your service. But the truth is, there's no such big campaign I have seen in my existence here, only yobit had a huge amount of participants.

I was initially thinking of hacking up something with *JS frameworks but my web dev skills are horrible so I have opted to use the low-code service Bubble.io to graphically design the front and backend. Apify scrapes the bitcointalk threads so I don't have to worry about that myself, then I use Zapier to make an integration - copying data - between Bubble and each CM's spreadsheet without requiring them to do hocus-pocus or black magic stuff on their end.

Zapier is quite pricey so without payments, I'll be operating at a net loss unless maybe I replace the payment system with a donation or tipping button.


Would you consider if instead of a subscription fee there was a donation button instead?

There are a lot of factors why this is happening:
First of all, there are people from all over the world on this forum and for those from poor nations, some bucks mean a lot. What business wants? Poor workers. This situation is similar to when immigrants enter developed countries and feel good to work on a very low salary. It makes situation worse for the local ones and overally, quality of life is going down while the capitalists are saving more and feel better.

But, at the same time, you get what you pay for. Companies prefer to hire cheap persons, who, in 90% of cases are bad in their job and run signature through them. Then, companies want to offer low rates to campaign members and morel likely get the crowd of spammers.

The problem is that companies think like this: I'll hire a cheap campaign manager who was some green trust and looks well in the eye of the community, hire spammers and my signature will be displayed on bitcointalk (Being a manager isn't to accept participants blindly and fill the sheets, 7 years old kid can do that today).

But the problem is that they don't understand one thing: You get what you pay for. They don't have a look of perfect examples of signature campaigns like Chipmixer, Bitsler, Crypto-Games, FortuneJack and others. All of them were high-paying with quality managers and sig members. They paid a lot compared to the rest of the campaigns but they succeed, they succeed a lot compared to them!

The issue these days is not the lack of people wanting to use these tools, but rather not wanting to pay for them. Companies these days are cheap as fuck. They all go on the search for a manager and pick the cheapest person rather then think about what they're getting for the money.

Now, for a manager to get a job or 2 we have to go through a rate war and end up busting ass to do the work for pennies. Then. if we wanted to use the tools you or LoyceV is offering, pay that fee. At the end of the week, we make 2$.

I know you guys don't wanna work for free, nor do I. The marketplace is just flooded with worthless wanna be managers these days who don't do the work. Kinda screwed us all over IMO.

Well said.

The idea is really really cool; however-> Consider these:

- Campaign managers don't get paid a lot lot, so chances are many managers won't bother to opt for a tool like this. History often repeats itself, bitcointalk has never had a growing and everlasting demand for campaign management bots, so keep in mind that.

- How accurate would your service be? And I am curious about your tech-stack which you would potentially use?

- How would your tool differ from just a normal excel sheet which somewhat automate post counting(Total post count from last week - Post count of this week for instance).

- If you're to make a service, the service must reduce the work of a manager by a greater extent imo. But if the manager has to still manually verify and authenticate things, he/she would definitely think twice before purchasing a service.

I am not sure how helpful it will be but, have a look at this for scrapers and crawlers: https://github.com/BruceDone/awesome-crawler

There's no way an automated system can properly assess post quality. However, for a high-quality campaign with high-quality users, I believe my automated system can get within 95% of what a manual count would do.

In this case I will suggest NotATether to create the tools what you wanted already and try to display the function of your tools and what are the new/extra tasks your tools can do. If you can show something more awesome than previous and the accuracy rate is almost like manual then hopefully bounty managers can turn to use your service for test atleast. If after that they are interested you can get some buyers.

The first signature campaign bot was created by the marcotheminer for Bit-X campaign (like NeuroticFish mentioned above). It happened back in 2015.
I suggest you to check that thread and comments around it, here OP discuss about the UI, Functions, etc: https://bitcointalksearch.org/topic/signature-campaign-bot-976910
There wasn't much interest to be fair.
There was one signature campaign run by coinomat and they were using marcotheminer's bot too: https://bitcointalksearch.org/topic/coinomatcom-signature-campaign-full-1007178
There was coinoindex also (coinomat project): https://bitcointalksearch.org/topic/coinoindexcom-signature-campaign-full-1126158

There were some campaigns that had their own bots. I remember two of them:
1. Coinroll - https://bitcointalksearch.org/topic/coinroll-signature-campaign-1491815
2. Bitmixer - https://bitcointalksearch.org/topic/closed-bitmixerio-signature-campaign-1657397


It would may worth to create a tracking service but as you see, there isn't a skyrocket demand on signature campaigns. If there were hundreds of signature campaigns, then it would succeed without a doubt and there would be even a necessarity of serious platform for not only sig campaigns but for different kind of services. I did that at some point (html/css page with great ui and functionality but to move everything on wordpress or to create custom CMS, would cost a lot, a lot of money.

So the general pereption here seems to be that signature campaign managers seem to not be interested in such a tool.


This particular tool employs their own residential proxies to circumvent all that, and a scraping job would only have to fetch a few hundred entries per week.

Again, I was given a free trial of this stuff, that's how I know it's working.

Bounty managers may utilize it for sure because they have thousands of members participating on their campaign so its a lot hell of work for them everyweek but if you are talking about the signature campaigns which has very limited number of participants then they have less chance to take your service because they are supposed to do very little work and have to take a look at each members post manually or that is what they supposed to do.

The social media sites you cite employ countermeasures to prevent scraping on any large scale, so anyone scraping those sites will need to defeat those countermeasures, which is not trivial. The forum does not employ any countermeasures to prevent posts being scraped even on a massive scale.

I remember the time of the biggest bounty hype in 2017, there are many platforms that worked automating data processing for bounty campaigns. Bountyhive, bountysuite, bountyplatform... Also as I know, most bounty managers did not want to use them, especially for signature campaigns.
post counting automation comes out of the power of Bitcointalk managers as forum users and I am not sure is it possible to check useful and non-spam posts using bots.

As a campaign manager, I am not interested to use such tools, especially for signature campaigns. For Facebook or Twitter, I think there are already tools. I don't know exactly what is that, but there is something like bot if I am not wrong. Often I do not manage big social campaigns. I am not encouraged to use it because checks by humans and AI wouldn't be the same. And counting posts isn't too hard with checking quality. Many things won't detect by AI, especially there is no clear explanation of what is a quality post. For me, I don't want to depend on AI for such as job. I want to make sure my participation working as I want based on my requirements.

I just found and tried a (pricey) service for scraping Twitter, Facebook, Instagram and Reddit: https://apify.com/ That means I could theoretically process tweets and comments and check if they meet bounty campaign guidelines.

It won't work with Telegram though. I'd need to code a bot that monitors user activity and "phone home" the post statistics given that campaign rules when a Telegram channel is involved go something like: "you must be very active on our channel and ask constructive questions, you must NOT discuss the project" and stuff like that.

So far, no buyers

Neither does the forum But I don't think that's what stops campaign managers from using my service. Maybe what's convenient for me isn't as convenient for others.

And it's fun to make

His service doesn't have an uptime guarantee.

I'm able to make at the very least a 99% uptime service (which only allows for 3d 15h 39m 29s yearly downtime according to https://uptime.is/). It's a bit low by business SLA levels but it'll do for a service that's only really used once a week.

We have seen problems with various tools and services going down and never coming back up. So having multiple services available is a net positive benefit. It's also good for competition in this field.

Loyce already offers a similar service.
And it's, by far, not the first one. Iirc in 2015, marcotheminer was using such a tool for the Bit-X campaign.

I don't know though if any good campaign manager would use only such an automatic tool in 2021. I expect the posts get checked (at least "scanned") for length, being well (enough) written, on-topic, not too far in a spammy useless thread,... I don't know how often are all these requirements checked, but I don't expect that an automated tool could do that.

Of course, you can, amongst the numbers and filters, also offer the actual posts and threads. But at this point only the targeted campaign managers can answer...

Edit: looked up for the link too: LoyceV's convenient (paid) service for signature campaign managers

I think LoyceV already offers something similar

...would you buy it?

I'm thinking about making a paid webapp with login/registration for campaign managers which automatically tracks user applications, detects alt accounts using a database, and also the posts that users make in a particular week along with filtering posts with basic criteria such as post length and not counting posts in specific boards.

Is there an interest for such an app? I think it would be helpful especially for saving time during the post-counting time immediately before the payouts.

If I somehow manage to track twitter posts and telegram user activity, would you become interested in the app? This question is mainly of interest to bounty managers who have telegram/twitter requirements

---

It would be something like this:

- You click a button that makes a new "project", where you define the number of weeks the campaign would run (corresponding to the #weeks you have escrow money for, which can be increased later)
- Then you paste the application format inside the app along with the topic title to define the application format and the thread, along with the exact times a week "starts" and "ends" for processing purposes
- Then you'd make an "open slots list", with the number of desired participants to add along with the post number in the thread to begin counting applications from (this is so that previous applications from last round are not processed again)
- It then lists all the applicants in this period which you then manually go through, and you can close the "open slots list" at any time by selecting applications in the list from the webapp (it ONLY adds them to the spreadsheet and doesn't post on bitcointalk)

- For each week the post scraper counts each participants post history between the start and ending days of each week and filters the posts accordingly, and calculates a final total.

- You then make the weekly payments yourself, etc. Removal is done by yourself and the app won't do that for you.

EDIT 2: now that I think of it this idea can be extended to "games and rounds" entries minus the weekly post counting part.