After a couple of months of creating campaigns early in my POF career, while attempting to exclude iOS and Android, I accidentally created a campaign targeting ONLY iOS & Android. I realized soon after that what I previously thought was an idiot mistake was actually dumb luck. Follow along to find out why.

In a recent post I explained how to create mobile-compliant landing pages in an effort to increase volume and profit in POF. After that post, however, I had a lot of people ask me if it was really worthwhile to target and optimize mobile traffic on POF. So to follow up on that post, I compiled data obtained across numerous campaigns over the last month and a half to go into greater detail the kind of traffic you might be missing out on if you’ve been ignoring mobile targeting with your campaigns in POF. But, before I get too far in this post, keep in mind that the traffic I discuss in this post is all Web or Mobile WAP traffic only. None of the data here represents any App traffic. I want to mention this because while POF does have a mobile app, there haven’t been any efforts to monetize their app with advertising.

First things first, the data.

In an effort to quickly and easily present some basic numbers, I created two graphs which are displayed below. To truly explain how profitable mobile traffic on POF can be, in each graph I compare both mobile and web traffic side-by-side in terms of Volume, Clicks, Conversions, Conversion Rate (CVR), Earnings per Click (EPC), Cost per Click (CPC), and profitability or Return on Interest (ROI).
Below is the first graph, which addresses volume and CVR.

You’ll notice immediately that the campaigns in this case study had almost 5x as much web traffic as mobile traffic yielding more clicks and conversions for web traffic than mobile traffic. Next, after doing a little bit of math, you may notice that the CVR of web traffic is slightly greater than mobile traffic. With just the above data, it may seem that mobile targeting in POF isn’t worth that much of your time. But take a closer look and you can see that the CTR of mobile traffic (.206%) is higher than that of web traffic (.175%), making up for the lower CVR. Now, one may ask why the CTR would be higher but the CVR would be lower. One can only speculate on why this happens but one explanation is that when people are browsing the web on their phones, they are more likely to accidentally click on something (your creative) that they never intended to click on in the first place. Also, the lower CVR may just be a result of either an offer landing page that’s not mobile-compliant or is too complicated for a person using a phone. Either way, mobile-targeted campaigns will likely have a higher CTR than web campaigns.

But what about the cost, earning, and ROI of each targeted traffic? For that, I would like to direct you to the graph below:

As you probably see right away after looking at the above graph, mobile traffic for these campaigns on average were much cheaper (a little more than half) than web traffic. Compare that to the average CPC of each targeted traffic and you see that mobile traffic actually had a greater ROI than web traffic. This may be odd to some people at first since the web traffic had a higher EPC than mobile traffic, but since the CPC for mobile traffic was so much cheaper, this means a greater overall profit for the mobile traffic.

So What’s the Verdict?

Well, at first glance, it may seem not worth your while to target mobile traffic. Perhaps you would rather spend your time going after the greater dollar and only spend your time on web traffic and completely exclude mobile targeting in your POF campaigns. Personally, I like the idea of finding multiple small bits of treasure where I thought they never before existed. Sure, there’s less volume, but by simply splitting campaigns between mobile and web targets and adjusting bids, you may find monetizing the extra volume to be quite effortless. After all (after doing some math with the above data) while targeting mobile devices we find a little over $13,000 in revenue that otherwise never would have been obtained had mobile targeting been excluded. While that may not buy you a new house, it’s not just chump change.

So don’t be afraid to split your traffic and optimize your mobile traffic in POF. You may find some more money to squeeze out of your already profitable campaigns!

Now that you’ve gotten full grasp of the basics of CPM bidding on POF, let’s jump ahead and assume that you’ve found your profitable creatives. Now you’re ready to hold everything else constant and find the CPM that produces the most profit for that campaign.

Your first inkling might be to duplicate your existing campaign and run them at 2 CPM bids simultaneously. This isn’t the best option as you are negatively affecting the quality of traffic of your lower CPM campaign since it will always be beat out by your higher CPM campaign. More on why you shouldn’t duplicate campaigns in POF.

Ideally, we would want to swap the CPM bids we want to test every 15 minutes throughout the day, every day. We can’t just test one price for half the day and another for half the day because performance varies according to time of day. In addition, different days of the week also affect performance.

Since most people don’t have the ability to swap CPM bids every 15 minutes, I’ve found the best compromise to be testing each CPM for 7 days. That way, you cover every day of the week. At the end of the 7 days, test another bid that is $0.10-$0.30 different than the one you tested before.

Doing this every 7 days will allow you to drill down to a CPM that will approximate the best bid that generate you the most profits. When comparing CPM bid performance, the only thing you should focus on is average profit per day. At the end of the day, that’s the ONLY thing that matters.

I understand that this could seem like it takes too long. Keep in mind you only have to do this once per week. During the time that your doing this, you should be making a profit most weeks since you are testing with high performing creatives already. Lastly, the extra profit that you will be able to squeeze out eventually once you find the optimal CPM bid is worth it from my experience.

So, give it a shot, and see how this system works for you!

POF’s CPM-based bidding system can be confusing for people who are more familiar with CPC bidding systems. If you’re used to Facebook’s CPC bidding system, I promise you that once you get used to POF’s system, even though it’s not perfect, it will further justify your hatred toward the Facebook ad platform.

POF Bidding System Basics

POF does not implement an incremental bidding system. Meaning, you pay what you bid. For instance, if the highest bid out of all bidders for a particular targeting is $0.60, and you decide to bid $0.61, then you would get what POF deems as the highest quality traffic (more on that later).

Not knowing what everyone else bid, however, you may decide to bid $1.00 for the same targeting. In that case, you would gain no incremental benefit in comparison to a $0.61 bid, while paying exactly $0.39 more for every 1,000 impressions.

The lack of an incremental bidding system, along with a lack of a CPC bidding option, are the only shortcomings of POF’s bidding system. Save these 2 possible improvements, POF’s system is consistent and fair. They won’t penalize you for having poor CTR’s on your ads like Facebook.

What You Are Getting By Bidding Higher

Bidding affects two things: volume and quality. Volume is pretty self-explanatory. The higher your bid, the more volume you will get to your campaign. This is assuming that you have no frequency cap nor daily budget set on your campaign. Check out our Bid vs. Volume Case Study for some empirical data on this.

The quality portion is purely governed by session depth. The higher your bid, the lower the session depth you will get with your campaign. POF gives your higher bidding campaigns priority over other lower bidding campaigns and shows your ads to their users first. Check out our Bid vs Quality Case Study for some empirical data on this.

Let’s breakdown how this works. Imagine there are only 2 bidders, one at $0.51 (Bidder A) and one at $0.50 (Bidder B), and only 1 ad spot that people can self-serve. A user signs onto POF (session depth 1), and since Bidder A has the higher bid, A’s ad is shown to the user. The user reloads the page, and Bidder A’s ad is shown again. This continues until either 1) Bidder A’s campaign reaches its frequency cap for that user, 2) Bidder A runs out of daily budget, or 3) the user stops using the site.

So bidding higher generally gets you lower session depths. Does this really translate to higher quality traffic? In general, yes. This has been tested by many people over the years, and there is a clear trend that ads shown earlier to users in a session performs better. It could be a variety of reasons, but one possible reason is users who have just logged in have not immersed in the usage of the site and become blind to the ads. Is there, however, a difference between session depth 2 and 3? Probably nothing significant.

Thus, we can conclude that changes in CPM bids ONLY affect your session depth. Higher bids get you lower session depth and more volume intrinsically. Lower bids get you less volume since there are fewer users who get to higher and higher session depths. That is, everyone who logs in gets session depth 1, but some won’t get to session depth 2. Even less will get to session depth 3, and so forth. Therefore, if your campaign is the highest bid, it will be shown to the most people possible since you get their 1st session depth.

Based on what we’ve discovered here, you can create a 2nd campaign with the exact same targeting, and it will not affect the performance of the 1st campaign as long as your CPM bid is lower. This allows you to try to monetize higher session depth traffic without worrying about it competing with your 1st campaign.

CPM Bidding Out the Gate

So, what to bid when you first start a campaign? In general, I advocate setting bids at relatively high CPM’s when you first launch a campaign. This is when you want to hold all variables constant while you test your creatives. You want a controlled environment to weed out your poor performing ads so that you know when a creative performs poorly, it’s not because you were jiggering a bunch of targeting options.

Bidding relatively high allows me to make sure that 1) I am getting the best positioning so I’m not selling my creatives short, and 2) I am getting volume as fast as possible to test for creative CTR’s. Once I have done that and have found high performing creatives, I lower the bid and find a good balance between volume and ROI.

If you’ve run a lot of campaigns in the past in the same targeting, you probably have a pretty good idea of what makes a “relatively high” bid for that demographic. For people with less experience, however, let’s come up with a gauge as to what to bid.

To do that, we’re going to solve for your CPM ceiling based on your offer payout, your expected CTR, and your expected CVR.

The equation is:
Payout * Expected CVR * Expected CTR * 1,000

If you have no idea what you expect your CVR’s and CTR’s to be, then use these general rules of thumb. If you’re promoting an offer with a $4.00 payout, you might reasonably expect a 15% CVR and a 0.1% CTR. In general, if you aren’t able to achieve those numbers, it will be difficult for you to make any meaningful money. As a marketer, you should get to the stage where you’re achieving at least that, so those expectations are conservative in my opinion.

The CPM gauge in this case would be:
$4.00 * 15% * 0.1% * 1,000 = $0.60

With that general gauge, you know you could bid in the upper $0.50′s to the $0.60′s.  In most cases, the more niche your targeting, the higher your CTR should be, and therefore your bid.

Next, start running the campaign, and track the volume the campaign gets over a few hours as you adjust the bid downward. If the volume doesn’t drop as you lower your bid, you’re good. Remember though, while your volume might not drop, the quality of your traffic might (i.e. you are getting higher session depths with lower bids), which is something you can’t measure at this point.

I would recommend that you never drop below $0.40 as a general rule when you’re starting a campaign to test for creatives until you’ve gotten a batch of profitable creatives and ready to optimize for the CPM sweet spot.

If, however, your campaign is a broad-targeting campaign, then you will be able to bid lower, perhaps much lower. Each campaign is different, and you really have to just test and get a good feel for it over time. Broad-targeting campaigns could be campaigns without login count targeting, campaigns without ethnicity targeting, campaigns targeting a wide age range, etc.

In a subsequent post, we will reveal how we collect and use data to optimize for the most profitable CPM bid levels for all our campaigns. So, stay tuned!