Is This a Good Bidding Strategy?

Just wondering what others think of this bidding strategy. One PoF what I’ve been doing is uploaded a lot of adds (100+), pausing all but 5 then running those to see what has the highest CTR.

Most of my ads have .15-.25+ CTR, which is cool. I usually bid around .49 cents for this–my question is, is it a good idea to reduce my bid after I get a solid CTR? Or is the idea to get a good CTR, bid high and also get it converting so you can maintain the volume (from the high CPM bid) and make a profit (with hopefully a good CR)?

Just looking for feedback on how people bid on PoF and whether or not my method is viable/makes sense…thanks!

User Comment:
with POF it’s more a factor of going after volume and impression quality. If your ads are good and you want more traffic you can bid higher to get better position and have your budget filled first. (one thing to always remember is that concept of budget priority)

on the flip side you could reduce your bid and try to squeeze more ROI out of a camp and see when you start to lose impressions.

A lot of times i’ll leave a good camp alone and copy it to split test a higher/lower bid or another variable so i can compare them side by side.

User Comment:
on POF you pay exactly what you bid, so unfortunately the only way to optimize bids is by split testing them.

User Comment:
I’m fairly new to all this and have been tackling dating on PoF for a little while now and am finally seeing some success. But there’s still some things i’m struggling with. Was reading this post and as I have also been looking for a good way to test bids amounts. I decided to ad my own questions to this thread

So when you copy a campaign to split test different bid amounts, how do you know if the increase in CTR and or CR is due to the change in bids or because a given user may see your ad more times? Would split testing the same campaign not increase the frequency cap at the same time?

For instance I read this post earlier and decided to try split testing bid amounts on one of my campaigns as I hadn’t thought of doing it that way before. I didn’t see much of an increase in impressions, but saw a bit of an increase in CTR and possibly CR though I probably don’t have sufficient numbers to say. Here’s the stats from the few hours I tried this:

CampA Hour1: imp 2068/clicks 3/conv. 0/Bid $.60 Hour2: imp 2281/clicks 3/conv. 0/Bid $.60 Hour3: imp 2027/clicks 3/conv. 0/Bid $.60
CampB Hour1: imp 2078/clicks 5/conv. 0/Bid $1.00 Hour3: imp 2261/clicks 3/conv. 1/Bid $1.20 Hour3: imp 2100/clicks 4/conv. 1/Bid $1.00
(copy of camp A)

So, with not much data it would look like trying a higher bid may be beneficial but this got me thinking about how split testing these two campaigns would probably be increasing the freq. cap for my ads and that possibly what i’m seeing may be due to the increase in freq. cap, especially with the amount of impressions staying pretty much the same? The only way I can see of testing it would be to the split test freq. cap as well, but do i test all 3 at once or just one at a time using my original campaign as the baseline? How do you guys tackle this?


User Comment:
i see what your saying here about the more camps you throw up the more times your ads get seen. A lot of it boils down to taking as much data as possible and drawing a few conclusions.

From you data above your success could be due to either:
-Higher bid = better quality traffic and more conversions
-Higher freq = more conversions
-time of day = more conversions for certain hours

I think the numbers of impressions are pretty tight to look at on an hourly basis. Maybe take a step back and look at a full days traffic or better yet several days traffic and draw a few conclusions there.

Unfortunately (and this is particular to POF) camps can burn out quick so it’s tough to test things knowing that your data is constantly influenced by a number of factors.

I’m learning the longer I’m in this game to let things run and don’t micro manage campaigns. It’s great for ROI but will keep you away from long term and large dollars. Test 1 variable at a time and do what you can to let it run. Don’t worry so much about perfecting a campaign and all it’s variables before letting it take on traffic and make you some money. Get things going, scale, and then optimize whenever possible.

You need to think of your data as possible trends over time more so than hard facts unless you have a very large amount of data to analyze. Hope this helps

User Comment:
^ some great advice

User Comment:
Thanks that does help alot. I have been running the above campaign for close to 3 weeks now and it’s close to break even. The problem i’ve been having is that one day it’ll be at 50% – 100% ROI and the next it’s at -50% or less. I’ve been collecting data each day and each hour to try dayparting/weekparting but it’s varying so widely from week to week and day to day that it’s hard to do. I did make the mistake of assuming certain things too early when i first started and pausing the campaign for the day if I didn’t see it performing too well and then started just letting it run, if i let it run throughout the day by the end of the day i’m usually around the break even mark give or take a few dollars. I did that for a few days and then figured I had enough data to try dayparting, so tried that for a few and found it’s still varying widely. LOL I’ve also been constantly testing new images and new ad copy throughout these past few weeks. I’m now at a point where I have about 6 or 7 creatives that are consistently getting CTRs of .15% – .37% and CRs of 25% – 42%. Total spent on Camp so far $198.54, made $174, net -$24.54. Also have 2 other campaigns (same offer, same targetting, diff age group) with very similar overall stats. All in all, with these campaigns i’m about $150 in hole (tested diff login counts, age groups, etc at the beginning and the 3 camps mentioned above won out) But here’s what i’ve learned:

1. Spend more money sooner after initial tests to see if it might work are done.
2. Only test as many creatives as budget will allow at one time.
3. Once you find winning creatives keep those running while adding a couple more at a time. Don’t pause your good creatives to test a whole new batch. 😛
4. Let the campaign run right from the start, don’t pause it if you don’t see much happening after a few hours and then wait a few hours and run it again to see if this is a better time, etc. Just let it go, so 2 weeks later your not trying to puzzle together a decent dayparting plan and then realize you have no or very little data for certain hours of particular days of the week. 😛 Or realizing a week in (after you throw your hands up in the air and just let it run) that certain hours of the day which you "assumed" wouldn’t be good at all for your demo are actually the best converting hours LOL
5. Don’t try to read too much into way too little information. (yikes! i’m bad for this)
6. If a creative is *close* to being profitable just pause it, no miracle is going to happen to make it suddenly start working. Creatives that do work will jump out and start screaming "Keep Me! I’m a Winner!" pretty much right from the get go.
7. hmmm….did i mention once a campaign looks like a winner spend more and just let it run right from the beginning?

Anyway, we’ll see if these campaigns hold out much longer or die out, going to keep at it until I start to see a steady drop in CRs. Here’s to hoping i can actually end up making these profitable. If not $150 seems like a good price for the lessons learnt above

Anything else you guys can add to this after reading above info please do so.

Cheers 🙂

The Article Published IN 08-24-2011 04:34 AM

Share To More ()