MTurk Mentors

[Metallica]

I fully understand. No worries. Having a lonnnng birthday party for a friend so can't comment a lot now, but ....
I work about 9 hours a day and am lucky if I can make $16. $30 or even $20 is VERY rare for me.

"Is this due to being scared of rejections or is it something like not being able to do them fast enough to make it worth doing?" Not afraid of rejections. Do surveys almost exclusively (and I tend to do even those about 15% slower than most other Turkers, I think). Just not fast enough on batches to do well on any given day.

I use MTS and HIT scraper and PC, but few actual "scripts."

I see what you and I are doing as a "process." THANKS for your help, sir.

Given I'm waning in and out of sleep, I'll start with the easier part

Would you mind sharing the settings you use on HF/HIT Scraper/PC? Are you using a combination of Hit Finder and Scraper? Or are you not using the HF functionality from MTS? You can PM the settings to me if you like, no need to place them here if you aren't comfortable with doing so

 

[Jerami]

Hey, I'm sorry for not responding last night like I said I would. I feel asleep too soon and now I have to pay for it by staying up until 8 am

I'm going to go off some things that I loosely remember, so correct me if I'm wrong on any of it.

I want to recall that you said you work for 8 hours uninterrupted, and come away with around $15-25. Is that correct?

And you prefer surveys to batches? Is this due to being scared of rejections or is it something like not being able to do them fast enough to make it worth doing?

Lastly, which scripts are you using? Full disclaimer: I do not turk often anymore and outside of common scripts like Hit Forker, MTS, PC, and so on, I won't be of much help. I tried to get another Mentor to hit you up regarding that, but it doesn't seem like any have. I'll tag @Jerami anyway.

I'm sorry for playing 20 questions with you I'd just like to get some sort of baseline established so that I'm covering exactly what you're looking for. My soft goal is around that $20 mark as well.. some days it can be that, or $50, or $100. I don't think you're doing terrible, but there's definitely some room for improvement there given you work for 8 or so hours. I understand that we are very different people, so I'll be sure to keep that in mind when giving you advice.

(When you get the chance to respond to this, I'll make a separate post for each question answered)

Hey sorry, I meant to reply last night. I don't batch often and am not a big user of scripts and definitely nothing recent, minus MTS. If you'd like to see my method of teaching people to turk, my "down and dirty get-to-turking NOW" method, I'll invite you to a convo where I taught someone. It has a couple of scripts. I'll PM you. Check.

 

[Jerami]

I'd be interested in this too. Now that I've been at it since the beginning of the year I need to learn how to step it up a notch. I'd like to be able to get over $100/week more easily. I don't know if that will just come with time but if you guys can enlighten me/us with your expertise that would surly be appreciated. Thank you to all you mentors out there!

Did you want to look over my "Turking 101-Expedited Jerami Method"? It's not a lot of scripts and very minimal, in general. I feel that there are a lot of things that aren't necessary to make $20 a day easily, given adequate time per day. I honestly feel that you should easily be able to do about $5 an hour, minimum, with this method. It's most of what I do, excluding having multiple screens, which is my number 2 suggestion. I feel that, in 8 hours, you should be able to reach a goal of $40 a day, consistently. I know that's not a ton of money but, it's more than a lot of people.
Let me know and I'll invite you to read the convo.

 

[ChristopherASA]

Given I'm waning in and out of sleep, I'll start with the easier part

Would you mind sharing the settings you use on HF/HIT Scraper/PC? Are you using a combination of Hit Finder and Scraper? Or are you not using the HF functionality from MTS? You can PM the settings to me if you like, no need to place them here if you aren't comfortable with doing so

Right now I've got HF set at 1100. Sometimes I'll do a snatch and grab (catch a HIT as soon as it pops up), other times I halt it to channel HITs into HIT Catcher. (I seem to get better results with HF than I do PC.)

Any HITs posted on the forum I'll route to PC and try and keep those down to 8 HITs loaded. I delete lower-paying HITs in favor of higher-paying HITs as time progresses.

I also use Scraper; just find what look like good HITs and click the HIT name to see if it's still available and, if so, I accept it. If not, I'll click "Reward and Panda" to PandA it.

I really think I don't work enough on batches. So what I've started to do is run Scraper ("most available") and see if I can find batches that way.

Today I made $16.45: $1.71 p/hr.

Thanks!!

 

[ChristopherASA]

Did you want to look over my "Turking 101-Expedited Jerami Method"? It's not a lot of scripts and very minimal, in general. I feel that there are a lot of things that aren't necessary to make $20 a day easily, given adequate time per day. I honestly feel that you should easily be able to do about $5 an hour, minimum, with this method. It's most of what I do, excluding having multiple screens, which is my number 2 suggestion. I feel that, in 8 hours, you should be able to reach a goal of $40 a day, consistently. I know that's not a ton of money but, it's more than a lot of people.
Let me know and I'll invite you to read the convo.

$40 a day would mean FREEDOM for me. Not kidding.

 

[Metallica]

Right now I've got HF set at 1100. Sometimes I'll do a snatch and grab (catch a HIT as soon as it pops up), other times I halt it to channel HITs into HIT Catcher. (I seem to get better results with HF than I do PC.)

Any HITs posted on the forum I'll route to PC and try and keep those down to 8 HITs loaded. I delete lower-paying HITs in favor of higher-paying HITs as time progresses.

I also use Scraper; just find what look like good HITs and click the HIT name to see if it's still available and, if so, I accept it. If not, I'll click "Reward and Panda" to PandA it.

I really think I don't work enough on batches. So what I've started to do is run Scraper ("most available") and see if I can find batches that way.

Today I made $16.45: $1.71 p/hr.

Thanks!!

Your settings and methodology actually looks just fine. In the nicest way I can say it, I thought you might have HF running at 5000 or something I leave my HF minimum reward setting at 5 cents, by the way. You could probably do better than me with a lower setting - I just find it to be a good number since most 4 cent and below batches suck.

Personally I'd avoid looking at the forum for posted HITs unless they're like $1-1.50 or above. I find it challenging to catch much of anything these days when I'm in the zone, and that extra few seconds it takes for the HIT to be posted on the forum means you're almost certain to not catch it. To be at your most efficient turking, you'd want to avoid looking at the forum instead of whichever scraper.. just those few seconds can make a big difference to your daily earnings if something big posts I'm not saying to avoid using catcher scripts, but watching a scraper as much as possible will reduce your reliance on catchers as well as the following frustration that comes when you have to add yet another panda to it.

Batches.. unfortunately I don't have any "get quick fast" solution for this, but you don't seem THAT slow.. have you tried warming up for the day by taking a typing speed test or something? I find those actually help my efficiency at times. And I know we covered this and you said being scared wasn't an issue, but sometimes you can do a HIT so fast it feels like you're doing them wrong - usually you aren't. If you have a big cushion, I'd recommend submitting a hit or two from whichever batch you come across and feel like you can do it efficiently. If you're doing them wrong, you'll find out pretty quickly anyway. Most batch work revolves around the same types of work (categorizing things, harvesting emails from a company's webpage, etc.), so after a while you won't be second guessing anything. Of course, you should continue to not touch batches that have terrible ratings. Another thing to cover would be finding new, good batches that people haven't brought up.. if you see a batch sitting untouched for hours, it may not be because they suck - people just may not realize its worth doing. Consider this, too: @C to the J is one of the more efficient turkers out there, and 9 times out of 10 he isn't going to share a super awesome batch he comes across to the forum. Nothing wrong with that, but it means you'll need to be able to judge for yourself how good a batch is. (you'll learn this with experience)

The last thing that I could cover (and would take a while to teach someone new to scripting) is writing scripts yourself for a batch you come across. I'm in the process of teaching a small class on AHK at my job, and if I remember enough I'll try to adapt that into something for you. AHK most likely isn't a program you're familiar with, but you can use it to select the first radio button in a HIT by pressing 1 on your numpad, or so on.

The last point brings this to mind.. have you used it? https://greasyfork.org/en/scripts/24650-mturk-radio-keybinds
Its a really nice script to have and is super easy to edit to a particular batch even as a beginner. It provides the same functionality most AHK scripts would.

 

[Jerami]

Right now I've got HF set at 1100. Sometimes I'll do a snatch and grab (catch a HIT as soon as it pops up), other times I halt it to channel HITs into HIT Catcher. (I seem to get better results with HF than I do PC.)

Any HITs posted on the forum I'll route to PC and try and keep those down to 8 HITs loaded. I delete lower-paying HITs in favor of higher-paying HITs as time progresses.

I also use Scraper; just find what look like good HITs and click the HIT name to see if it's still available and, if so, I accept it. If not, I'll click "Reward and Panda" to PandA it.

I really think I don't work enough on batches. So what I've started to do is run Scraper ("most available") and see if I can find batches that way.

Today I made $16.45: $1.71 p/hr.

Thanks!!

So, I just want to be sure I'm reading this right. You're running HF, and HitScraper(what delay[time in ms]?) and PandaCrazy with 8 Pandas and HitCatcher with some more PandAs? Just trying to visualize what you've actually got running, and the order of operations for the system that you are currently operating. Because, if I'm reading it right, there seems to be a lot going on that doesn't leave much room/time for working. Also, HitFinder or Scraper should be living in it's own spot, right out in front. Even if I'm not actively working anything, such as when I'm watching a movie, HF lives on my left screen, always on top. I know it's always there(can be done with split screen instead of second monitor) and when it moves, I can see it out of the corner of my eye. Human eyes have evolved to catch small movements more easily in their peripheral vision so, see HF move is just natural. I know the most recent HIT, in 'Logged HITs' will always be at the top, so that's the only part I have to focus on. I know the accept and the pay are in the upper right, narrowing what I have to focus on further. So, if HF moves, that means a new hit posted. Then I immediately move my mouse and my eyes to that upper right corner and click if the pay is remotely good. I'd rather chuck it back than try and rely on a PandA after reading and making a decision about it's quality. Click first, ask questions later. Am I a woman? "Nope." Chuck it back. But, at least I had the option.

So, HF having a space front and center is really key for me. It makes me the most money. I catch most of my stuff there so it get's priority. Muscle memory will eventually kick in so that you can always land on that 'accept' link without too much effort. But, you really want to try and be the one catching the HITs from HF, or Scraper, or whatever. To make the leap to the next step of turking, you should get into the mentality of trying to be a HIT predator and jump on them when they first hit HF. Once you miss that first chance, a PandA is like hoping that you can find a half a sandwich that someone was finished with. There's a chance, if you're not too picky about what kind of sandwich. And, PandAs from the forum are usually like carrion. If you hang out with predators, there's a pretty good chance you'll see all kinds of dead stuff but, you're only going to get what they don't want or need anymore. You're hardly going to fill up or, make money, if you understand the metaphor.

I don't think there are any actual figures I could reference but, in my estimation, your chances of catching a HIT are at least halved when you have to use a PandA over a 'post and grab'. Honestly, it's probably much less, even as low as ~25%. So, with chances that low, if PandA management is taking a significant amount of your time, you may want to adjust that. Because, I really feel like, if your mouse pointer lived on the top right of HitFinder, and you didn't even turn on any PandA apps, you could probably catch, at least 2, $1 HITs and complete them within an hour.

I have more to say, I'm sure but, that'll do for now. Keep at it.

 

[Metallica]

Also, given most people get nowhere near maxing out on a daily basis, try to live by the "accept first, ask questions later" motto. See a .20 survey post? Try to catch it off of HF (but don't waste a slot in PC for it). Worst case scenario, it looks like garbage so you just return it.

 

[GemStone]

Did you want to look over my "Turking 101-Expedited Jerami Method"? It's not a lot of scripts and very minimal, in general. I feel that there are a lot of things that aren't necessary to make $20 a day easily, given adequate time per day. I honestly feel that you should easily be able to do about $5 an hour, minimum, with this method. It's most of what I do, excluding having multiple screens, which is my number 2 suggestion. I feel that, in 8 hours, you should be able to reach a goal of $40 a day, consistently. I know that's not a ton of money but, it's more than a lot of people.
Let me know and I'll invite you to read the convo.

@Jerami Thanks, yes, I'd like to read about your method. I've read what you've posted here and it sounds interesting.

 

[Jerami]

@Jerami Thanks, yes, I'd like to read about your method. I've read what you've posted here and it sounds interesting.

Check PM. Let me know if you have questions but, I don't think there will be. Start a new Conversation by clicking my avatar. That old one is dead. It's pretty 'solid state', if you know what I mean. Only a couple of scripts.

EDIT It is also VERY basic, just to reiterate that point. It's a solid foundation though.

 

[Metallica]

@ChristopherASA did you see an improvement in your earnings this week?

 

[ChristopherASA]

Hey, Metallica. No. I was bad. I have a slight problem with beer. I've got to get back to serious. Sorry if I've been a bother. But at least I DID force myself out of bed this morning to work (and boy, it is SLOW).

 

[ChristopherASA]

Metallica. Don't give up on me, please.

 

[ChristopherASA]

@Metallica and @Jerami I DEARLY apologize to both of you for not responding to you in a timely manner. I've got some personal problems I've been trying to deal with. Please don't give up on me.

 

[ChristopherASA]

Your settings and methodology actually looks just fine. In the nicest way I can say it, I thought you might have HF running at 5000 or something I leave my HF minimum reward setting at 5 cents, by the way. You could probably do better than me with a lower setting - I just find it to be a good number since most 4 cent and below batches suck.

Personally I'd avoid looking at the forum for posted HITs unless they're like $1-1.50 or above. I find it challenging to catch much of anything these days when I'm in the zone, and that extra few seconds it takes for the HIT to be posted on the forum means you're almost certain to not catch it. To be at your most efficient turking, you'd want to avoid looking at the forum instead of whichever scraper.. just those few seconds can make a big difference to your daily earnings if something big posts I'm not saying to avoid using catcher scripts, but watching a scraper as much as possible will reduce your reliance on catchers as well as the following frustration that comes when you have to add yet another panda to it.

Batches.. unfortunately I don't have any "get quick fast" solution for this, but you don't seem THAT slow.. have you tried warming up for the day by taking a typing speed test or something? I find those actually help my efficiency at times. And I know we covered this and you said being scared wasn't an issue, but sometimes you can do a HIT so fast it feels like you're doing them wrong - usually you aren't. If you have a big cushion, I'd recommend submitting a hit or two from whichever batch you come across and feel like you can do it efficiently. If you're doing them wrong, you'll find out pretty quickly anyway. Most batch work revolves around the same types of work (categorizing things, harvesting emails from a company's webpage, etc.), so after a while you won't be second guessing anything. Of course, you should continue to not touch batches that have terrible ratings. Another thing to cover would be finding new, good batches that people haven't brought up.. if you see a batch sitting untouched for hours, it may not be because they suck - people just may not realize its worth doing. Consider this, too: @C to the J is one of the more efficient turkers out there, and 9 times out of 10 he isn't going to share a super awesome batch he comes across to the forum. Nothing wrong with that, but it means you'll need to be able to judge for yourself how good a batch is. (you'll learn this with experience)

The last thing that I could cover (and would take a while to teach someone new to scripting) is writing scripts yourself for a batch you come across. I'm in the process of teaching a small class on AHK at my job, and if I remember enough I'll try to adapt that into something for you. AHK most likely isn't a program you're familiar with, but you can use it to select the first radio button in a HIT by pressing 1 on your numpad, or so on.

The last point brings this to mind.. have you used it? https://greasyfork.org/en/scripts/24650-mturk-radio-keybinds
Its a really nice script to have and is super easy to edit to a particular batch even as a beginner. It provides the same functionality most AHK scripts would.

I've copied and printed this out and will read it today, and respond. As I will @Jerami's. THANK YOU!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

 

[ChristopherASA]

So, I just want to be sure I'm reading this right. You're running HF, and HitScraper(what delay[time in ms]?) and PandaCrazy with 8 Pandas and HitCatcher with some more PandAs? Just trying to visualize what you've actually got running, and the order of operations for the system that you are currently operating. Because, if I'm reading it right, there seems to be a lot going on that doesn't leave much room/time for working. Also, HitFinder or Scraper should be living in it's own spot, right out in front. Even if I'm not actively working anything, such as when I'm watching a movie, HF lives on my left screen, always on top. I know it's always there(can be done with split screen instead of second monitor) and when it moves, I can see it out of the corner of my eye. Human eyes have evolved to catch small movements more easily in their peripheral vision so, see HF move is just natural. I know the most recent HIT, in 'Logged HITs' will always be at the top, so that's the only part I have to focus on. I know the accept and the pay are in the upper right, narrowing what I have to focus on further. So, if HF moves, that means a new hit posted. Then I immediately move my mouse and my eyes to that upper right corner and click if the pay is remotely good. I'd rather chuck it back than try and rely on a PandA after reading and making a decision about it's quality. Click first, ask questions later. Am I a woman? "Nope." Chuck it back. But, at least I had the option.

So, HF having a space front and center is really key for me. It makes me the most money. I catch most of my stuff there so it get's priority. Muscle memory will eventually kick in so that you can always land on that 'accept' link without too much effort. But, you really want to try and be the one catching the HITs from HF, or Scraper, or whatever. To make the leap to the next step of turking, you should get into the mentality of trying to be a HIT predator and jump on them when they first hit HF. Once you miss that first chance, a PandA is like hoping that you can find a half a sandwich that someone was finished with. There's a chance, if you're not too picky about what kind of sandwich. And, PandAs from the forum are usually like carrion. If you hang out with predators, there's a pretty good chance you'll see all kinds of dead stuff but, you're only going to get what they don't want or need anymore. You're hardly going to fill up or, make money, if you understand the metaphor.

I don't think there are any actual figures I could reference but, in my estimation, your chances of catching a HIT are at least halved when you have to use a PandA over a 'post and grab'. Honestly, it's probably much less, even as low as ~25%. So, with chances that low, if PandA management is taking a significant amount of your time, you may want to adjust that. Because, I really feel like, if your mouse pointer lived on the top right of HitFinder, and you didn't even turn on any PandA apps, you could probably catch, at least 2, $1 HITs and complete them within an hour.

I have more to say, I'm sure but, that'll do for now. Keep at it.

@Jerami I just copied and printed this out. Will read and respond. Sorry to be such a bother. "(

 

[Jerami]

@Jerami I just copied and printed this out. Will read and respond. Sorry to be such a bother. "(

You're not a bother. Please don't worry about it because it makes me feel like I implied that you were a bother. You aren't. If you were, I'd block you. I am in full control of the 'ignore' button on my end so, if someone bothers me, I have the solution. I can't let the forum stress me out or it's not good to come here. I wouldn't use smileys in my post if I were bothered. That's why I use them, to signify that I'm happy, not bothered.

Relax. Feeling stressed for no reason isn't going to help make any extra money.

EDIT I understand life gets in the way and people get busy. I will say, in all honesty, that when learning something and beer compete, beer will always win out.

 

[ChristopherASA]

"Please don't worry about it because it makes me feel like I implied that you were a bother." No no no! Not at all!!!!!!!! I just have a ... very negative self image and that comes thru from time to time.

"...when learning something and beer compete, beer will always win out." Not only am I spending money on beer, each time I get some I'm losing money by not being able to work the next day. Beer is bad shit. Thanks!!!!!!!!!!!

 

[Jerami]

"Please don't worry about it because it makes me feel like I implied that you were a bother." No no no! Not at all!!!!!!!! I just have a ... very negative self image and that comes thru from time to time.

"...when learning something and beer compete, beer will always win out." Not only am I spending money on beer, each time I get some I'm losing money by not being able to work the next day. Beer is bad shit. Thanks!!!!!!!!!!!

We all have our own situations to deal with but, I can say, if you get beer pretty quickly when you get up, try getting behind the pc first thing, like half awake. Then just start, slowly. Keep going and before you're even fully awake, you'll have made some money. Then, once you've made some money, if the high of making more doesn't keep your interest, maybe, at least, you can say, I'll make $5 more before I get up, at all. If you want to get up, you'll work faster. Then start raising that amount. Eventually, get to $20 before taking a break.

If possible, don't drink beer while at the pc. If you combine the 2 they're harder to separate. Also, if you have to step away to drink, you'll notice the absurdity of standing in the kitchen, while HitFinder is going off.

 

[ChristopherASA]

"We all have our own situations to deal with..." I know that, and feel just a bit guilty talking about mine, but . I'm in a half stupor today but forced myself up because I knew I had to work. I just need to cut my beer in half (and take one devoted day off from Turking) or stop all together.