10 Free and Effective Ways to Promote Your Webcomic

[LibertyCabbage]

Sometime next week, I'll have a write-up about an experiment I did in Project Wonderful.

[VeryCuddlyCornpone]

Neat!

I tried having an adbox on my own site, but the views were too low so I got delisted. The weird thing was I was getting a lot of pageviews according to SJ stats, but I guess people weren't scrolling enough for the ad to be visible? For some reason I thought it counted as an ad view as long as the code loaded, I didn't know it had to actually be on the screen. I didn't care much for the way the ad looked when I had it "above the fold," so I dunno I was making a bleeding pittance anyway, and even though I wasn't looking to make $$ off of it, seeing how slowly the fractions of cents were adding up was frustrating to me for some reason.

I'm only running one (paid) ad right now, which isn't doing spectacularly, but I don't care enough to cancel it. I think in my initial zeal I overdid it and used my ads too much and now I'll need to scrape together some new ones if I want to have effective things people are interested in clicking instead of things that people have seen so often it might as well be part of the ad site's wallpaper.

I think my strategy will be to run a few more free ads during the end of this chapter because what's there to lose, and then when next chapter starts I should have some new ads ready (maybe even have ones that actually pertain to the storyline of the chapter?!? scandal) so that if I do decide to run some paid ads they'll actually be interesting enough to get some clickins.

[Humbug]

It's hard to say what makes people stay, but I seem to be doing alright. After paying running a number of ads for 1 cent on PW for over a month, I'm noticing a slight increase for ToP on update days. According to PW stats, about 80+ clicks so far.

[robotthepirate]

Hmm, this is all good advice.

I already do FB and I'm not sure I have the time to get into Twitter. I think DA works better on more traditional comics, if you can post cool sketches of your characters in cool poses it'll get you interest, all I can do is put up pictures of my characters in their fairly unimpressive normality. Plus I'm always distracted by delusions of graduer that make me try "real" art. I'll have to see about tumblr when I have more time.

One thing I did recently was to start a Blogger account. All I'm planning on doing is using it as a newsfeed for my sites to keep people up to date with progress but I've had 33 blog views since about 6pm last night (it's now just before 1pm GMT, so 19hrs). I don't know how many of them are from me because although I've told it not to count my views as page views it might still be counting the views I make of the iframe on my homepage. I'll have to see how it pans out but as long as I include links to the website in the blog posts I should get some traffic.

[Humbug]

In regards to Tumblr and Deviantart, I'm getting more hits from deviant art and not even a handful from the former. Maybe someone has better luck on Tumblr than me, but that's what I'm experiencing at the moment. DA is a bit more promising but constantly submitting to groups gets tedius as hell and I'm not quite sure of other methods of getting more views. (Beyond buying ads) Maybe joining contests will help, but frankly I've no time for those.

[robotthepirate]

In regards to Tumblr and Deviantart, I'm getting more hits from deviant art and not even a handful from the former. Maybe someone has better luck on Tumblr than me, but that's what I'm experiencing at the moment. DA is a bit more promising but constantly submitting to groups gets tedius as hell and I'm not quite sure of other methods of getting more views. (Beyond buying ads) Maybe joining contests will help, but frankly I've no time for those.

Groups is something I've never explored fully on DA, shall have to look into it more.

[LibertyCabbage]

]I think my strategy will be to run a few more free ads during the end of this chapter because what's there to lose, and then when next chapter starts I should have some new ads ready (maybe even have ones that actually pertain to the storyline of the chapter?!? scandal) so that if I do decide to run some paid ads they'll actually be interesting enough to get some clickins.

As someone who's been bidding on free ads (and, I guess, endorsing doing it), I'm pretty much over it now after running a campaign. I think I'd much rather throw a few bucks at Project Wonderful and let its server do penny bids than deal with doing tedious free bids three times a week, 95 percent of which will either get outbid, never be approved, or not get clicked on.

Groups is something I've never explored fully on DA, shall have to look into it more.

I've been submitting all of my pages to #Webcomics-Community, #Web-Comic-Alliance, and #SmackJeeves. I doubt I've been getting a notable amount of traffic from it, but it's so quick and easy to do that I don't really care.

------------------------------------------------------------------------

I was curious to see how a Project Wonderful campaign works, so I put in $5 and scheduled a campaign to make penny bids over a course of five days to promote my blog. However, I couldn't help but wonder if there was a more effective strategy. So, instead, I decided to do an experiment by breaking up the campaign into five separate campaigns, each lasting a maximum of 24 hours and with a maximum budget of $1. Here are the results.

(All of the campaigns were scheduled to start and end at 2 p.m. EST.)

Day 1 - $0.01 max bid - Tuesday to Wednesday
Min. hits/day: 50; square banner: no; bid style: cautious
Clicks: 34; views: 84,031; expense: $0.27; cost per click: $0.01

Day 2 - $0.02 max bid - Wednesday to Thursday
Min. hits/day: 50; square banner: yes; bid style: cautious
Clicks: 16; views: 73,379; expense: $0.89; cost per click: $0.06

Day 3 - $0.03 max bid - Thursday to Friday
Min. hits/day: 100; square banner: no; bid style: conservative
Clicks: 24; views: 86,972; expense: $0.94; cost per click: $0.04

Day 4 - $0.05 max bid - Friday to Saturday (lasted 8 hours)
Min. hits/day: 500; square banner: yes; bid style: aggressive
Clicks: 23; views: 113,425; expense: $1.00; cost per click: $0.04

Day 5 - $0.10 max bid - Saturday to Sunday (lasted 4 hours)
Min. hits/day: 1,000; square banner: no; bid style: cautious
Clicks: 26; views: 75,234; expense: $1.01; cost per click: $0.04


Conclusion: The first day, which used penny bids, had great results, while the rest weren't nearly as positive. I was surprised to find that several factors I thought were significant -- day of the week, minimum hits per day, and bid style -- had no noticable impact. The last three days, for example, all had around 25 clicks and a $0.04 cost-per-click ratio despite their criteria varying heavily. Also, I was initially pleased that the campaign was bidding on $0 ads, which I've previously done manually, but these free ads only resulted in two clicks.



Ad type
Skyscraper - clicks: 45; views: 22,651; expense: $1.88; cost per click: $0.04
Button - clicks: 42; views: 319,426; expense: $1.62; cost per click: $0.04
Half banner - clicks: 21; views: 54,204; expense: $0.71; cost per click: $0.03
Square - clicks: 10; views: 26,162; expense: $0.65; cost per click: $0.07
Rectangle - clicks: 5; views: 10,598; expense: $0.19; cost per click: $0.04

Conclusion: There's no big winner here, although the square ads are a clear loser. They seem to be somewhat overpriced for their size, while the larger skyscraper and rectangle ads did a better job of getting readers' attention. The small ads performed reasonably well because of how inexpensive they are, as sites often have four or more ad spaces for them.



Traffic region
United States - clicks: 38; views: 91,905; expense: $1.16; cost per click: $0.03
Europe - clicks: 34; views: 171,251; expense: $1.33; cost per click: $0.04
Elsewhere - clicks: 26; views: 95,176; expense: $1.35; cost per click: $0.05
Canada - clicks: 25; views: 74,709; expense: $1.22; cost per click: $0.05

Conclusion: United States ads got the most clicks while costing the least amount of money; however, the inexpensive European ads weren't far behind, as I was able to advertise on high-traffic sites where U.S. ads were outside of my price range.



Sites with more than two clicks
Girl Genius - clicks: 11; views: 163,600; expense: $0.05; cost per click: $0.004
Flaky Pastry - clicks: 4; views: 5,750; expense: $0.04; cost per click: $0.01
chainsawsuit - clicks: 3; views: 4,400; expense: $0.04; cost per click: $0.01
out at home - clicks: 3; views: 33; expense: $0.002; cost per click: $0.001
School Bites - clicks: 3; views: 1,100; expense: $0.02; cost per click: $0.005
Mock Girl - clicks: 3; views: 6,898; expense: $0.02; cost per click: $0.008

Conclusion: Girl Genius accounted for about 9 percent of my clicks while only costing about 1 percent of my budget, making it a great value. I think the reason it works so well is that it has a lot of space for button ads -- 14 slots, in fact -- which means that bidders don't need to be particularly competitive. In any case, with the exception of Out At Home, which appears to be a fluke, all of these are high-traffic sites with four or more ad spaces.



Overall: The experiment was somewhat discouraging, as the cost-per-click ratio was higher than the $0.01 or $0.02 I was hoping for. In addition, while adding up each campaign's expense gives a total of $4.11, my account was actually charged $5.05, leaving me with a negative balance of $0.05. I expect that I'll be able to advertise more effectively in the future, though, now that I've learned more about how the process works. If I were to start a new campaign today, I think I would go with a $0.01 maximum, no square ads, no traffic threshold, no expiration date, and only bidding on United States and Europe ads. However, another option to consider would be to do manual bidding, using half-banner ads for United States traffic.

[LibertyCabbage]

]I think my strategy will be to run a few more free ads during the end of this chapter because what's there to lose, and then when next chapter starts I should have some new ads ready (maybe even have ones that actually pertain to the storyline of the chapter?!? scandal) so that if I do decide to run some paid ads they'll actually be interesting enough to get some clickins.

As someone who's been bidding on free ads (and, I guess, endorsing doing it), I'm pretty much over it now after running a campaign. I think I'd much rather throw a few bucks at Project Wonderful and let its server do penny bids than deal with doing tedious free bids three times a week, 95 percent of which will either get outbid, never be approved, or not get clicked on.

Groups is something I've never explored fully on DA, shall have to look into it more.

I've been submitting all of my pages to #Webcomics-Community, #Web-Comic-Alliance, and #SmackJeeves. I doubt I've been getting a notable amount of traffic from it, but it's so quick and easy to do that I don't really care.

------------------------------------------------------------------------

I was curious to see how a Project Wonderful campaign works, so I put in $5 and scheduled a campaign to make penny bids over a course of five days to promote my blog. However, I couldn't help but wonder if there was a more effective strategy. So, instead, I decided to do an experiment by breaking up the campaign into five separate campaigns, each lasting a maximum of 24 hours and with a maximum budget of $1. Here are the results.

(All of the campaigns were scheduled to start and end at 2 p.m. EST.)

Day 1 - $0.01 max bid - Tuesday to Wednesday
Min. hits/day: 50; square banner: no; bid style: cautious
Clicks: 34; views: 84,031; expense: $0.27; cost per click: $0.01

Day 2 - $0.02 max bid - Wednesday to Thursday
Min. hits/day: 50; square banner: yes; bid style: cautious
Clicks: 16; views: 73,379; expense: $0.89; cost per click: $0.06

Day 3 - $0.03 max bid - Thursday to Friday
Min. hits/day: 100; square banner: no; bid style: conservative
Clicks: 24; views: 86,972; expense: $0.94; cost per click: $0.04

Day 4 - $0.05 max bid - Friday to Saturday (lasted 8 hours)
Min. hits/day: 500; square banner: yes; bid style: aggressive
Clicks: 23; views: 113,425; expense: $1.00; cost per click: $0.04

Day 5 - $0.10 max bid - Saturday to Sunday (lasted 4 hours)
Min. hits/day: 1,000; square banner: no; bid style: cautious
Clicks: 26; views: 75,234; expense: $1.01; cost per click: $0.04


Conclusion: The first day, which used penny bids, had great results, while the rest weren't nearly as positive. I was surprised to find that several factors I thought were significant -- day of the week, minimum hits per day, and bid style -- had no noticable impact. The last three days, for example, all had around 25 clicks and a $0.04 cost-per-click ratio despite their criteria varying heavily. Also, I was initially pleased that the campaign was bidding on $0 ads, which I've previously done manually, but these free ads only resulted in two clicks.



Ad type
Skyscraper - clicks: 45; views: 22,651; expense: $1.88; cost per click: $0.04
Button - clicks: 42; views: 319,426; expense: $1.62; cost per click: $0.04
Half banner - clicks: 21; views: 54,204; expense: $0.71; cost per click: $0.03
Square - clicks: 10; views: 26,162; expense: $0.65; cost per click: $0.07
Rectangle - clicks: 5; views: 10,598; expense: $0.19; cost per click: $0.04

Conclusion: There's no big winner here, although the square ads are a clear loser. They seem to be somewhat overpriced for their size, while the larger skyscraper and rectangle ads did a better job of getting readers' attention. The small ads performed reasonably well because of how inexpensive they are, as sites often have four or more ad spaces for them.



Traffic region
United States - clicks: 38; views: 91,905; expense: $1.16; cost per click: $0.03
Europe - clicks: 34; views: 171,251; expense: $1.33; cost per click: $0.04
Elsewhere - clicks: 26; views: 95,176; expense: $1.35; cost per click: $0.05
Canada - clicks: 25; views: 74,709; expense: $1.22; cost per click: $0.05

Conclusion: United States ads got the most clicks while costing the least amount of money; however, the inexpensive European ads weren't far behind, as I was able to advertise on high-traffic sites where U.S. ads were outside of my price range.



Sites with more than two clicks
Girl Genius - clicks: 11; views: 163,600; expense: $0.05; cost per click: $0.004
Flaky Pastry - clicks: 4; views: 5,750; expense: $0.04; cost per click: $0.01
chainsawsuit - clicks: 3; views: 4,400; expense: $0.04; cost per click: $0.01
out at home - clicks: 3; views: 33; expense: $0.002; cost per click: $0.001
School Bites - clicks: 3; views: 1,100; expense: $0.02; cost per click: $0.005
Mock Girl - clicks: 3; views: 6,898; expense: $0.02; cost per click: $0.008

Conclusion: Girl Genius accounted for about 9 percent of my clicks while only costing about 1 percent of my budget, making it a great value. I think the reason it works so well is that it has a lot of space for button ads -- 14 slots, in fact -- which means that bidders don't need to be particularly competitive. In any case, with the exception of Out At Home, which appears to be a fluke, all of these are high-traffic sites with four or more ad spaces.



Overall: The experiment was somewhat discouraging, as the cost-per-click ratio was higher than the $0.01 or $0.02 I was hoping for. In addition, while adding up each campaign's expense gives a total of $4.11, my account was actually charged $5.05, leaving me with a negative balance of $0.05. I expect that I'll be able to advertise more effectively in the future, though, now that I've learned more about how the process works. If I were to start a new campaign today, I think I would go with a $0.01 maximum, no square ads, no traffic threshold, no expiration date, and only bidding on United States and Europe ads. However, another option to consider would be to do manual bidding, using half-banner ads for United States traffic.

[VeryCuddlyCornpone]

I got an absurd amount of hits from Out at Home as well. Like if I recall correctly it bizarrely resulted in hundreds of clicks. It was cool but kinda weird; it seemed like PW would tell me x amount of people clicked my ads but the views I'd expect to see in my statistics didn't seem to match up with that or even be anywhere close. I guess a lot of it might have been clicks where people go "Oh shit I clicked an ad by mistake" *immediate back button before site loads* ? Who knows.

[LibertyCabbage]

I got an absurd amount of hits from Out at Home as well. Like if I recall correctly it bizarrely resulted in hundreds of clicks. It was cool but kinda weird; it seemed like PW would tell me x amount of people clicked my ads but the views I'd expect to see in my statistics didn't seem to match up with that or even be anywhere close. I guess a lot of it might have been clicks where people go "Oh shit I clicked an ad by mistake" *immediate back button before site loads* ? Who knows.

I've seen some funky stats as well, like getting clicks from a site even though it shows zero page views. (Whuh?) Although, keep in mind that PW keeps track of two kinds of clicks -- total clicks, and unique clicks -- so it might be more useful to look at the unique clicks statistic in strange cases like that.

[robotthepirate]

Without doing the experiment multiple times it could be hard to be certain of your results. On the other hand it's your money, so spend it as you like.

I image certain factors, like the type of ad, are fairly conclusive as they've been test multiple times over the different days but I'd say you would need to play around with the minimum bid,the day of the week and the "min hits/day" in order to confirm these weren't chance results.

I'm probably still going to try (in June) the $0 bid thing for a while just because my justifiable budget for my comic is exactly £0:0. At least for now it is.

[LibertyCabbage]

Without doing the experiment multiple times it could be hard to be certain of your results. On the other hand it's your money, so spend it as you like.

Oh, absolutely, and I'm not going to pretend like I was doing anything scientific. I'm just curious about the subject because I wasn't able to find any in-depth information anywhere on effective PW strategies. It was a fun intellectual challenge for me to try to make sense of all the data, and I probably got a lot more entertainment out of my $5.05 than I would have spending it on anything else.

I think my next move will be to drop a quick $5 on manual U.S. ads, and then let a $5 penny-bid campaign do its thing. I'll post the results here of each strategy.

I image certain factors, like the type of ad, are fairly conclusive as they've been test multiple times over the different days but I'd say you would need to play around with the minimum bid,the day of the week and the "min hits/day" in order to confirm these weren't chance results.

Another issue that's important, which I didn't really have a chance to test out, is how appealing the ad design is. For example, an unappealing/uninteresting skyscraper ad may perform poorly, but that doesn't necessarily mean that large ads are a bad investment, and it could also mean that the particular audience of that webcomic isn't as attracted to the ad (e.g., a romance-oriented ad on an action webcomic). I did use different rectangle ads, but I couldn't get much of an impression from it since publishers don't use them much.

I'm probably still going to try (in June) the $0 bid thing for a while just because my justifiable budget for my comic is exactly £0:0. At least for now it is.

I guess free bids get a decent amount of traffic for the time invested if you're really strapped for cash. I'd estimate that you can probably get the same amount of traffic from 200 free bids (the maximum) as you could get from spending $0.25 or so, though, in which case the difference is really more psychological than pragmatic. Doing something like manually bidding $0.01 on an ad for 30 days should be within anybody's budget.

(Also, apologies for double-posting earlier.)

[djracodex]

The break-down is super helpful LC, thanks! I signed up for PW, but I haven't had time to throw together any banners yet. Maybe one day, lol

[robotthepirate]

I'd estimate that you can probably get the same amount of traffic from 200 free bids (the maximum) as you could get from spending $0.25 or so, though, in which case the difference is really more psychological than pragmatic.

Now when you put it like that...

Well that's July's problem.

[LibertyCabbage]

The break-down is super helpful LC, thanks! I signed up for PW, but I haven't had time to throw together any banners yet. Maybe one day, lol

I figure it's better than nothin'. I should have a better picture of things once I've gotten a chance to make more bids.

Now when you put it like that..

Actually, one thing you could do would be to host an ad on your site, and then put any revenue you get into advertising. That way, it's "free" in the sense that all it costs you is digital real estate.

But yeah, I guess most of all, I'm just kinda tired of free bidding after doing it 50 times or whatever. I'm at the point where I'd rather just throw another dollar at the server and do something more enjoyable than spam-clicking stuff.

[Humbug]

Decided to take advantage of Dominic Deegan being its last week, and doing ok I suppose. I haven't had an ad up for full 24 hours because I keep getting outbid and I prefer to stay at a certain amount, but so far about 2% out of the pageviews actually click my ad, and Analytics matches up to the unique clicks stated on PW, so it seems I'm doing something right. I would likely get more clicks if it were a leaderboard or skyscraper, but for a square it's not too bad?

[LibertyCabbage]

That seems like a smart idea, and I see that Cope's gotten in on some of the action as well. You also have the advantage of a genre match, so there should be a good chance of those new readers sticking around.

Also:

from spending $0.25 or so

I forgot that Project Wonderful has a minimum deposit of $5, so you can't actually just, like, randomly put in a quarter like that. Instead, it'd be more, like, putting in $5 and then budgeting yourself $0.25 (or whatever) a week until it runs out.

Anyways, here's what happened with my next go-around. I think my next "experiment" will be a lengthier campaign, spending $5 on U.S. and Europe penny bids with only button ads. I'm curious to see how low of a CPC I can get with that. My previous campaign spent $0.27 in the first 24 hours on penny bids, so I expect this new campaign to last at least 20 days.

$5 budget - U.S. traffic - 2 p.m. EST Wednesday to 12 p.m. EST Thursday

The Winners
Gregor - button - 1 click - $0.01 - $0.01 CPC
Quantum Vibe - button - 14 clicks - $0.22 - $0.02 CPC
Selkie - half-banner - 5 clicks - $0.10 - $0.02 CPC
Girl Genius - button - 16 clicks - $0.37 - $0.02 CPC
HUGE Cartoons - rectangle - 1 click - $0.03 - $0.03 CPC

The Losers
Gunshow - button - 1 click - $0.33 - $0.33 CPC
Flaky Pastry - button - 1 click - $0.23 - $0.23 CPC
Widdershins - half-banner - 1 click - $0.20 - $0.20 CPC
Busty Girl - rectangle - 1 click - $0.11 - $0.11 CPC
Shotgun Shuffle - skyscraper - 4 clicks - $0.38 - $0.10 CPC

Previous Winners
Girl Genius - button - 16 clicks - $0.37 - $0.02 CPC (vs. $0.004)
Flaky Pastry - button - 1 click - $0.23 - $0.23 CPC (vs. $0.04)
chainsawsuit- button - 0 clicks - $0.00 - inactive (vs. $0.01)
School Bites - half-banner - 1 click - $0.06 - $0.06 CPC (vs. $0.005)
Mock Girl - half-banner - 0 clicks - $0.02 - n/a CPC (vs. $0.008)

Ad Type
Button - 34 clicks - $1.22 - $0.036 CPC
Half-banner - 41 clicks - $1.91 - $0.047 CPC
Rectangle - 14 clicks - $0.89 - $0.064 CPC
Skyscraper - 14 clicks - $0.95 - $0.068 CPC

Conclusion: Like in my previous experiment, the smaller ads had the best value. And if you take out the two biggest losers, Gunshow and Flaky Pastry, it brings the cost-per-click ratio for button ads all the way down to $0.021. It's important to keep in mind, though, that 88 percent of the button clicks came from just the Girl Genius and Quantum Vibe ads. The former provided a ridiculous 213,400 page views, so getting a decent amount of clicks from that's a pretty safe bet, and the latter had a 0.11% click-through-rate, which is even better than I got from most of the rectangle and skyscraper ads. The best explanation I have for why Quantum Vibe's rate's so high is that the webcomic's anti-government, which could mean that its readers were particularly interested in an ad featuring an intimidating police officer.

I'd say that bidding on United States ads is risky, as my $5 got me about 20 percent less clicks than the same amount got in the previous experiment. The five ads in the Losers category ate up 25 percent of my budget while only netting eight clicks, and a lot of the ads were in the $0.04- to $0.07-CPC range, which isn't any better than what I was getting by letting Project Wonderful run campaigns for me. I expect that I was also overly concerned with being the high bidder, and I may have bid too aggressively on some of the ads. Moving forward, when it comes to manual bidding, I think it might work better to focus on a few particularly rewarding sites rather than spreading my resources out over a variety of them.

[VeryCuddlyCornpone]

Moving forward, when it comes to manual bidding, I think it might work better to focus on a few particularly rewarding sites rather than spreading my resources out over a variety of them.

That's what I plan to do lately. Next time I get into it I'm going to try looking at other historical comics to see if I'll have more luck posting on "similar" places than just spraying my figurative seed into the wind accross the pasture.

[LibertyCabbage]

That's what I plan to do lately. Next time I get into it I'm going to try looking at other historical comics to see if I'll have more luck posting on "similar" places than just spraying my figurative seed into the wind accross the pasture.

Yeah. Although, I had a new idea after I made my post, which would be to use a site's CPC to figure out a target price. Take the Derelict skyscraper ad I bid on, for example:

$0.70 max bid; $0.20 expense; 5 clicks; $0.04 CPC

If I want a lower CPC, I can get it by multiplying the max bid and the CPC by the same percentage.

100% -- $0.04 CPC = $0.70 max bid
75% -- $0.03 CPC = $0.50 max bid (rounded down from $0.525)
50% -- $0.02 CPC = $0.30 max bid (rounded down from $0.35)
25% -- $0.01 CPC = $0.10 max bid (rounded down from $0.175)

At that point, it's just a matter of picking a target CPC and making the corresponding bid. If it's the high bid, then you hopefully get a good deal, and if you're outbid, then you don't get charged anything. It's the real-life equivalent of waiting till something goes on sale before you buy it, but it's even better because the transaction's done instantly and automatically.

[robotthepirate]

Now when you put it like that..

Actually, one thing you could do would be to host an ad on your site, and then put any revenue you get into advertising. That way, it's "free" in the sense that all it costs you is digital real estate.

CG don't like that though as it's their space to sell.

The best explanation I have for why Quantum Vibe's rate's so high is that the webcomic's anti-government, which could mean that its readers were particularly interested in an ad featuring an intimidating police officer.

Would it be worth designing ads specific to certain sites (or at least types of sites)? So if you have a comic audience you'd like to target (because you think people who like that comic will like yours or something) you could make an ad to appeal more to them.

Also I imagine it'll be worth working out a bit of a crop rotation system of which sites to target. Like Cuddly found, if your ad appears on the same site too often all the people who are likely to have clicked it will have done so already. That said it'd be good to return to high yield sites after a while to attract the attention of new readers and also to reattract people who a) read a bit of your work then forgot about it or b) were too busy to click it last time.

This all seems to suggest quite a lot of research into who you're advertising with, I wonder how effective it would be to advertise on these sites by other means instead of PW. So if Cuddly's going to hunt down other costume drama comics (ok, technically not a logical description) would she get more views/advertising by doing fan art or connecting with the author in some way on deviantArt? Find and befriend rather than find and bid. Obviously that takes a lot more work to do so it's a question of free time vs cost.