Thread: Day 10: Setting Up a Campaign on PropellerAds - EXPLANATION

Up until now, this tutorial has been easy to follow - we've set up your first campaign together without much hiccup (hopefully!)

But if mastering paid traffic was THAT easy, then EVERYBODY would be making the big bucks.

The more hurdles you jump over, the more people you leave behind, and the closer you'll get to the gold.

After you're finished with this tutorial, and are launching campaigns on your own, you WILL need to test extensively to find out what works and what doesn't - landing pages, offers, affiliate networks, traffic networks, optimization strategies, bidding strategies...etc. etc.

There's not really a way to avoid all that testing.

What I'm aiming to do with this tutorial, is arm you with enough basic knowledge so that you can figure out things on your own when this tutorial is done. All these walls of text are written to serve that end purpose.

Today's lesson will be a LONG read - so take your phone off the hook (or whatever is the equivalent for smart phones), go some place quiet, make yourself comfortable, and say goodbye to the next few hours.


***********************************

EXPLANATION

While you were setting up your first campaign on PropellerAds, I didn't explain what the various settings were for. Now, I want to go through those options in detail.

Most other pop networks have similar campaign settings to choose from (although each of them will still be different in its own way). So learning all the options on PropellerAds will help you tremendously when it comes time to scaling your campaigns to the other networks.

Some of what I tell you below may not make much sense, because you don't yet have experience with looking at campaign stats and optimizing a campaign based on them. Just do the best you can to understand as much of it as you can for now - you'll no doubt be referring back to all this information later on.

Reading this lesson over a few times will help - because some of the stuff I explain later in the lesson can help you to better-understand the stuff explained earlier in the lesson.

I do apologize for the length of this lesson, but please know that I'm not just going into details on PropellerAds options alone. I'll talk about campaign targeting, traffic quality, bidding strategies, campaign optimization strategies - if you pay attention, you'll pick up a lot of useful bits from this lesson. I don't mean to overwhelm you with information, but many concepts just can't be explained in isolation - so, I'll just have to ask you to suck it up and make it through this lesson as best you could.

Without further ado - below are some of the campaign options on PropellerAds, and an explanation on each.



Ad Format:



The 3 ad formats available are OnClick (Popunder), Push Notifications, and Interstitial.

Popunder ads are what we're learning to run in this tutorial.

Push Notifications ads are also outside the scope of this tutorial, but you are encouraged to test them yourself! You'll find some valuable guidance in the "Push Traffic" subforum:

https://stmforum.com/forum/forumdisp...8-Push-Traffic

Especially spend some time going over @twinaxe's threads and guides on how to run push traffic! In my opinion, push traffic is the best next step to expand into from pop. In fact, once you lock down a great lander+offer combination in your pop campaigns, try taking that combo to push and test push ads there (headline+description+image).

Interstitial ads are similar to pop ads - I've never tested them so will not discuss them in the tutorial. For a description of what they are please see this PropellerAds article.

Because this is a pop traffic tutorial, I will only cover in this post all the options that are associated with the "OnClick (Popunder)" format.



Pricing Model:



The 3 options are CPA Goal 2.0, CPM, and SmartCPM.


CPA in CPA Goal 2.0 stands for "Cost Per Action". In this case, it represents the cost you're willing to pay for every conversion you get.



In order to run a CPA campaign on PropellerAds, you will need to set up your tracker to post conversion data to PropellerAds, so they'd be able to automatically optimize the campaign for you - by automatically targeting traffic that will convert at your specified goal price. (We've already put the PropellerAds postback url into the tracker in a previous lesson - so you're covered.)

What does that mean exactly? It means that PropellerAds will collect conversions to see which traffic segments (zones, OS versions, browsers etc.) are converting better and which ones are converting worse. Their system will automatically bid lower for segments that aren't converting well to save money and get less traffic, and bid higher for segments that convert well to get more traffic and conversions.

In this way, PropellerAds will strive to optimize your campaign towards that CPA goal price you specify, to allow you to pay as close to that CPA as possible for each conversion.

Here's a comprehensive explanation by PropellerAds on how their CPA model works:

https://help.propellerads.com/en/art...ations-onclick

If you, like what I once thought when I was a newbie, that "hey this sounds like a good deal - I could just set the cost to be lower than my payout and make money", let me tell you now that it's not as simple as that.

Every traffic source aims to maximize profits, and PropellerAds is no different. If your campaign's conversion rate is too low, PropellerAds won't make much money from your camp - as a result they'll send that traffic to another bidder - one that has a better conversion rate and/or has set a higher CPA than you did - and only send you the leftovers.

To find out more about the CPA bidding model, which is available on PropellerAds and several other networks, please see this comprehensive guide by mrbraun, and do your own experimenting:

https://stmforum.com/forum/showthrea...CPA-cost-model

Do NOT choose the CPA goal 2.0 UNLESS you've already tested some offers and landing pages and have found an offer (or offer+lander combination if you're using landers) that converts well. The SmartCPM price model is recommended for testing offers and landing pages - as we'll talk about later.

And when you're ready to test CPA Goal 2.0, be sure to read @twinaxe's post on how to choose a conversion type: https://stmforum.com/forum/showthrea...Type-Splittest


CPM stands for "Cost per Mille", or the cost per 1000 impressions on your pop ad. If you're direct-linking, the ad would be the offer page. And if you're using a landing page, then the ad would be the landing page.



This is the price you pay to show your ad to visitors 1000 times. This pricing model is also referred to as "FixedCPM", which means that the bid you specify will be applied to all zones/placements - you'd be paying the same rate for traffic from every zone/placement.

By the way: A Zone is a website where our ads are displayed for that website's visitors to see. It's also referred to as a publisher site, or an ad placement or simply a placement. PropellerAds pays these placement site owners for showing their visitors our ads, and in turn sells this traffic to us with a markup. I'll be using the words "zones" and "placements" interchangeably in the rest of this tutorial.

The most important thing to understand about zones/placements, is that they each converts at a different rate due to various reasons. When you set up a campaign for an offer, you'll get traffic from lots of different placements. Some of them will give you enough conversions to result in a profit, some will not. How much you're bidding matters a lot as well (more on bidding later) - just because a placement isn't profitable at your current bid, doesn't mean it won't be at a lower bid. But there are placements that just won't make you money (or at least for the particular landing page + offer you're running) no matter how low you bid. One of the major tasks in campaign optimization involves blacklisting placements that don't end up profitable - we'll discuss campaign optimization in detail in a later lesson.

The good thing about FixedCPM is that you can control exactly how much you want to pay for placements. The bad thing is you'd be overpaying for the placements that don't have a lot of competition (usually because they don't convert well), that other competitors aren't bidding as high for; you'd also be underpaying for placements that have high competition, that your competitors are bidding much higher for (usually because they convert very well) - in fact, your current bid level may not even be high enough to get you ANY traffic from some of the best-converting placements.

I would recommend that you stick with SmartCPM instead of FixedCPM - for reasons explained below.


SmartCPM is the recommended price model to choose when you're testing offers/landers.



When you pick SmartCPM as the pricing model, PropellerAd's algo will automatically bid differently for different zones/placements on your behalf. The bid you specify will be the maximum you'll bid for any one placement.

This price model will avoid the overpaying issue discussed above for the FixedCPM model, so that you get more traffic for the same budget - which is always good!

Moreover: PropellerAds allows us to manually set a customized bid on each placement. Essentially, this feature completely eliminates the need to use the FixedCPM pricing model IMO.

To manually specify the bid for a zone/placement: Go to PropellerAds' "Dashboard" section where you'll find a list of your campaigns, and either click on the "Campaign Name" or the "Show Stats" chevron at the end to show zone/placement stats, find the zone, click on the little pencil icon to edit, and type in the bid.



Here's a sample bid optimization strategy for PropellerAds, when running on the SmartCPM model:

Bid reasonably in the beginning (more on this in another section below - basically don't bid so low that you'd be getting crap traffic nobody wants). You need to let some traffic run to find out which zones/placements are good or bad at generating conversions.

a)Cut the bad zones that are wasting money without conversions. b)For profitable zones doing good ROI, manually increase the bid for them to get more traffic - and hopefully more profits. c)For zones that are not far from reaching profits, manually decrease the bid to reduce cost - and hopefully push them into green.

Because the campaign bid you specify for SmartCPM is the maximum bid, it may not be high enough for you to win traffic from some of the most-competitive (and possibly best-converting) zones. So what we can do is launch a separate campaign, blacklist all the major zones we were getting traffic for in the original campaign, and set a high bid - to trigger traffic from the top-quality zones. Then we can use step 2 to manually optimize bids per zone.

The ability to specify different bids for different placements makes the bid-optimization process a lot simpler - too bad there aren't that many networks that offer it. Zeropark is another one that does.

Again, I would suggest to stick with SmartCPM as the Pricing Model while you're still testing landing pages and offers. Then once you have a winning lander+offer combination, try CPA Goal 2.0.



Target URL:



This will be the campaign url from the tracker. Just paste the url directly into here.

There's no need to meddle with the tracking tokens displayed below the "Target URL" field. If you look closely at the campaign url, you'd see that those same tokens have already been added - this is because they were programmed into the tracker's traffic source template for PropellerAds.



Frequency Capping:



Frequency Capping is the maximum number of times you want any particular visitor to see your ad over a specified timeframe.

For example, if you put "1 times per 24 hours", that means you only want any one visitor to see your ad once over a 24-hour period.

The impact that Frequency Capping has on your campaigns in general, is as follows:

The higher the frequency cap, the more traffic you'll get, but the lower the overall campaign ROI will be.

AND the corollary is true also...

The lower the frequency cap, the less traffic you'll get, but the higher the overall campaign ROI will be.

When you set your frequency cap to 1/24 (once per 24 hours), say you're buying 1000 impressions - or 1000 chances to display your ad in front of visitors - your ad will be shown once to each of 1000 visitors.

When you increase the frequency cap, say to 4/24, and say you're buying the same 1000 impressions - then your ad will be shown to less than 1000 different visitors. The ad will be shown to some of the same visitors a second, or third, or fourth time etc., over the course of 24 hours.

Generally speaking (although not always true), when you show your ad once to each of 1000 visitors, you will get more conversions vs. when you show your ad 1000 times to a smaller group of visitors by showing each person your ad multiple times.

Since you'll be paying roughly the same amount of money for each display/impression of your ad, you'll get the most bang for your buck (i.e. the highest ROI) by setting the frequency cap to the lowest setting possible - which on many traffic sources is 1/24.

However, ROI is not the only thing we're looking for. If we increase the frequency cap, we'll get more impressions or traffic volume, which means that in spite of the lower ROI, we may STILL end up making more profits.

This is not always the case, but you wouldn't know unless you test. It's synonymous to how wholesalers will sell products at cheaper prices so retailers would buy in bulk, to result in higher profits overall.

This is an important concept by the way: Always be aiming to maximize profits, and not ROI. We'd rather be making 30% ROI by spending $1000/day to make $1300/day = $300/day in profits, than be making 1000% ROI by spending $1/day to make $11/day = $10/day in profits. I talked about that in this post, and matuloo later wrote a more-detailed post here.

But I digress. The approach I would suggest for Frequency Cap, would be to start at 1/24 (once per 24 hours; which was what we had frequency set to in the "ACTION" part of this lesson), optimize until the camp is green first, and THEN test higher frequencies. There would be little point in decreasing your ROI if you're still in the red. I talked about this in the "Play with Frequency" section in this post.

One approach to maximize your profits would be to clone the original 1/24-frequency campaign, set it to a higher frequency, and bid lower. The idea is to pay less for lower-quality traffic.

Lastly: If you toggle "Frequency/Capping" to off, it means you allow every visitor to see your ad for as many times as PropellerAds' algorithm wants to display it to them. This can be good if you have an offer that converts extremely well, or has high profit margins, or require a lot of convincing/preselling in order to convert (so that you need to show the ad to each visitor multiple times) - or some/all of the above. Again, I would suggest to wait until you have a green campaign before testing this setting.



Traffic Options:



Propeller Ads vs. Brokers Traffic: The "Propeller Ads" option lets you buy traffic from websites/placements/zones that work directly with PropellerAds. "Brokers Traffic" lets you buy traffic from websites/placements/zones that are working with OTHER traffic networks - and PropellerAds have made arrangements to buy (or "broker") this traffic from these other networks.

Generally speaking, brokered traffic will not convert as well as traffic from PropellerAds' own publishers, but it's also cheaper so there's no telling whether you'll make profits, until you actually test it.

In the beginning when you're split-testing offers and landers, you want to be confident in the traffic quality (i.e. that the traffic is good enough to give you conversions) - otherwise you'd have an extra unknown, i.e. when you don't get good results you won't know whether to blame traffic quality or the offers/landers. From this perspective, leaving brokered traffic out of your campaign in the beginning would be wise.

Once your campaign is making money, you can include brokered traffic and compare results before and after, then decide whether to revert back to unchecking this option, or to leave it checked and continue cutting bad zones/placements.

You could even leave your original campaign alone, duplicate it and checkmark this option for the new camp, but set your bid lower to counterbalance the lower traffic quality (i.e. pay less for lower quality traffic). This is more advanced stuff, so if it sounds confusing don't worry about it. You can always come back to review this information when you have more experience.

"Include AntiAd Block":



This option allows you to select "Anti-Adblock" zones, which are zones where PropellerAds is using their technology to bypass adblockers, so that ads can still be displayed to visitors.

Because there's no telling how much of this traffic will actually be able to see your lander, I would suggest to leave this option unselected in the beginning, and only test it if the campaign is green.

And when you're ready to select this option, be sure to do one thing first: Check to verify that your landing page url and landing page domain are NOT blocked by adblockers. In the case where you're direct-linking to offers without using landing pages, you'd need to check your offer page url and domain. (Basically you'd need to check the url and domain of the page that is first displayed to the visitor.) Otherwise, the visitors you buy from anti-adblock zones will not see your landing/offer page, and you'll be wasting your money.

So how do you verify whether a page is blocked by adblockers? PopAds has provided a detailed explanation:



To find out a bit more on how anti-adblock works, here's an article on Propeller (ignore the stuff about integration - the article was written for website owners / publishers):

https://propellerads.com/blog/no-mor...se-of-adblock/

So in summary, I'd recommend to leave this option unchecked in the beginning. Once you have a profitable campaign, you can test anti-adblock traffic. There are different ways to do this, and one way is to clone the profitable campaign and include anti-adblock traffic for this new campaign. Then see whether the campaign will be profitable. If not, decrease the bid gradually to see if you can turn it green..



***********************************


Continuing in 2nd post below...




Amy

Countries and Bid:




Automatic Bidding and CPM: If you toggle "Automatic bidding" to enabled, PropellerAds' system will automatically set the bid for you - to an "optimal rate".

Different traffic segments have different average prices. For example, these days mobile traffic tends to cost more than desktop traffic, and 3G/LTE traffic tends to cost more than wifi traffic. So depending on the targeting options you specify - country, desktop/mobile, 3G/wifi, etc. - PropellerAds will figure out what the best pricing is, and bid that amount on your behalf.

And this optimal bid would be shown in the "CPM" field when you toggle the automatic bidding option ON.

I'm usually skeptical when a traffic network recommends bid amounts, because I've found that in most cases the prices are inflated - likely to maximize revenue for the traffic network. But these days, PropellerAds' optimal bid seems to be quite reasonable - I would suggest to follow this suggested bid.

IMPORTANT: I'm NOT referring to the bid amount shown in the screenshot below, which you'll see when you click inside the "CPM" field:



For some reason, that amount is way higher than the optimal bid that is shown in the "CPM" field when automatic bidding is enabled.

Moreover: I would suggest to bid a bit higher or lower than the optimal bid value, in order to avoid competition (because other people advertising on PropellerAds may set their bid to the exact recommended amount). I would suggest to bid 10-20% higher or lower than the optimal bid for this reason.

This was why, when we were setting up our first campaign in the previous lesson, I asked you to toggle automatic bidding to enabled first, to check what the optimal bid is, then toggle automatic bidding to disabled and enter the modified bid (+20%) in the "CPM" field.

One REALLY important thing to note here is that the optimal bid is dependent on your campaign targeting. So before you make a note of what the optimal bid is, make sure you first specify all your campaign targeting options such as country and desktop/mobile etc.

Next - I want to talk about why it's important not to bid too low.

Remember I mentioned above that it's important for us to be confident in the traffic quality in the beginning when split-testing offers and landers? Well the same applies to our bid - let me explain.

We've already learned that the conversion rate for each zone will be different. This is mainly due to 2 reasons:

Because each zone has a different audience, or a different group of people visiting it, the traffic from each zone will react differently to our ad and type of offer.

This is the much bigger reason: Some zones will send fake visitors generated by software, in order to make more money. This is commonly known as Bot Traffic, which does not convert. And different zones or placement sites will send different amounts of bot traffic.

Now, because every advertiser (i.e. your competitors) on the traffic source can choose which placements/zones they want to target, and which ones they want to cut, and also how much they want to bid, they will bid higher for the better-converting zones, and cut the ones that don't convert.

The result? Market needs will drive up bid prices for zones that convert better.

What this means for us is this: If we bid low, we may not even see traffic from the best placements. As we increase our bid, we'll often get better and better converting traffic. (We'll also likely be overpaying for traffic from the not-as-good placements as well - thus the need to cut them.)

This is why I often propose the following bidding approach for people that are just starting out:

Bid at least around average or slightly higher while still split-testing offers/landers in the beginning, to be confident about traffic quality, i.e. to be confident that we're not stuck with all the garbage traffic nobody else is bidding on. Once we find an offer+lander combination that converts reasonably well, we can test higher-than-average and lower-than-average bids.

So the question would be, how do you know what the average bid is? The optimal bid shown by PropellerAds, when we have "Automatic bidding" enabled, would be a pretty good amount to use for this purpose.

There's another check you can do to ensure you're not bidding too low. Notice that PropellerAds provides a traffic distribution graph:



I would suggest to avoid bidding below the first "hill" as I've indicated with the arrow in the screenshot above - [B]basically don't be one of the lowest bidders and get stuck with bad traffic (traffic from bot-infested placements) that nobody wants.

While we're looking at the graph: Note that the number of impressions on the graph is only the total traffic you can bid/compete for - the amount you will actually win/get tends to be much less. Since you're new, you want to receive a good amount of traffic in order to collect stats in a reasonable amount of time, so you can learn quickly without waiting many days to test each offer. So I would suggest to bid high enough to compete for at least 50k-100k impressions/day.

So all this is for picking an initial bid for a new campaign. We'll talk about bid-testing strategies in a later lesson.


Countries: Simply start typing the name of the country you're wanting to target and select it.

Traffic from each country will have different average costs, so if you've wanting to save money, avoid targeting multiple countries in the same campaign - as you can only specify a single bid in a campaign so may overbid for some geos while underbidding for others.

There are exceptions to this "rule": e.g. Some veterans, when testing multi-geo offers, would target all of them in one campaign to get a quick idea on which geos are low-hanging fruit. But this will cost more money than it needs to - so only do this when you're making so much money that your time is worth more than the extra money you'll need to be spending using this "blast first, aim later" approach.


Targeting by Cities or States: If, based on your campaign tracker stats, you're finding that certain cities or states are converting better than others, you could set up a campaign to target them separately. But the usual approach would be to target the country as a whole in the beginning.



Smart Optimization Tools:



Auto-Optimization is very similar to the CPA Goal 2.0 pricing model. I would suggest to disable this option for now. Again, you can test this later. I want you to learn to manually optimize your campaigns first, before attempting to apply any sort of automation. If you jump directly to using automation, you may never truly understand what the optimization process actually involves.

I may add a lesson later on to talk about CPA Goal 2.0 and Auto-Optimization. And even if I don't, please feel free to experiment with these options - just, AFTER you've had a chance to run and optimize campaigns manually first, without using optimization.

For more information on auto-optimization, please see: https://help.propellerads.com/en/art...eads-campaigns



Daily/Total Campaign Budget:



Pretty self-explanatory. Daily campaign budget is the maximum the camp will spend in a day, starting from and ending with midnight EST. Total campaign budget is the total spend since the start of your campaign.

Traffic should stop if it reaches either one of these budgets. The minimum you can set these values to is $10, and this is what I've asked you to set your budgets to in the first campaign, just to guard against potentially wasting money on an offer that doesn't convert.

(The reason why I put the word "should" in italics, is because PropellerAds would sometimes exceed the budget you've set. There's not a lot you can do about this.)



Targeting:





Platform - Self-explanatory - Desktop or Mobile depending on what traffic your offers will accept. If your offers accept both, you could test either mobile or desktop first to see if the offer shows promise, then test the other. Or, start 2 campaigns at the same time, one for desktop and one for mobile.


OS - Click on the text field and you'll see the options, and these options will be different depending on what you selected for "Platform". If you leave it empty you'd be targeting all OS's by default. You'd want to target all the OSs that are accepted by one or more of the offers you're rotating in your tracker's campaign settings for that campaign. As I'll explain in a later lesson, if the different offers accept different OSs, you can set up tracker rules to make sure that traffic from the right OSs will go to the right offers.

If you're running mobile, the main OSs you'll probably be focusing on the most, are Android, iOS, and maybe also Windows Phones. If you're running desktop, the main OS would be windows and mac. Once you have collected some conversions for your camp, you could drill into OS stats in the tracker, and either "Include" the best-performing and profitable OS, or "Exclude" the worst-performing and unprofitable OS, from your targeting here.


OS Versions - I mostly leave this blank in the beginning unless the offers I'm running will only accept certain OS versions. But later, if you drill into tracker stats and see that some OS Versions are wasting money but not converting, you can specify them here and toggle to "Exclude".


Device - If you've chosen to target only "iOS" in the "OS" field, you'd be offered two choices here: iPad and/or iPhone.


Browser - Again, you can decide to cut certain browsers based on stats. I would highly recommend excluding "Android Webview" at the beginning of every campaign. That traffic often converts badly, probably because it processes Javascript and CSS differently than "normal" browsers do and therefore will break some lander elements. However, once you've found a promising offer+lander, feel free to retest Android Webview - it's not always crap.




Browser Language - Again, this is another stat you can drill down into in the tracker, and you can "Exclude" languages that seem not to convert well.

Or - you could go one step further, by identifying the main languages that are spoken by your audience, then serve to them offers and landers that are in their language. For example, say you've been promoting an offer to Belgium, where the offer page and landers are all in Dutch. Then you drill down into browser language and notice that a lot of visitors have their browser language set to French. In this case, you could translate your landers into French, and set up tracker rules to direct visitors to either set of landers based on browser language. You could even try to look for offers with offer pages in the French language, that accept Belgium traffic, and split-test those against the original offer.


Connection type - Options are "All", "3G/LTE", or "Wi-Fi/Broadband".

Choosing "All" will target Wi-Fi and Carrier/3G traffic at the same time, which is often not the wisest thing to do, in part because carrier traffic is significantly more expensive than wifi traffic. So if you target them both in the same campaign using the same bid, you'd be either paying too much for wifi traffic, or too little for carrier traffic.


Mobile ISP - If you selected to target "3G/LTE" in "Connection type" above, and specified one or more countries in the "Countries" field, this section would be available for you to specify the carrier(s) you want to target.

Worth noting here is that some mobile carriers have multiple names, because mobile companies are merging and unmerging all the time. The result is that some traffic sources will still refer to some carriers by their older name. So if you can't find a particular mobile carrier on a traffic network, do a search in google to find out what other names they've had before, and see if they're referred to at the traffic network by those names.


Zone limitation - This is where you can specify a list of profitable placements/zones you want to receive traffic from, and toggle to "Add" - which is referred to as "whitelisting"; or specify a list of unprofitable placements/zones you want to cut, and toggle to "Exclude" - also referred to as "blacklisting".

I've had better experience with the blacklisting approach rather than whitelisting when it comes to pop, and many other people have reported the same, but whitelisting could work out in some cases. Feel free to experiment with this, but often when you only whitelist profitable placements, the traffic would somehow dry up. This is why I always recommend going the blacklisting route to start.

Another way to blacklist zones would be to go to PropellerAds' "Dashboard" section where you'll find a list of your campaigns, and either click on a Campaign Name or the "Show Stats" chevron at the end to show zone/placement stats, find the zone you want to blacklist, and click on "Exclude".




Audiences - This does not apply to your first campaign, but here is where you can retarget audiences you've collected in previous campaigns. That is, if you have chosen to do so. Please see the section below on how to collect audiences.



Collect an Audience for Retargeting:



In this section, you can create a retargeting list of visitors and subsequently target them. There is a TON of information online on what retargeting is, when to use it, and the benefits. You probably won't be needing to use this feature for low-payout offers, so can safely ignore it for now. You can find details here:

https://propellerads.com/blog/propel...eting-feature/

Personally I've never bothered to build audiences for retargeting on pop traffic. But this is because I haven't run high-payout offers such as gambling or biz op where a customer can be worth hundreds of dollars in commissions or more, in which case it could be worth it to retarget those people with similar offers. I would encourage you to do your own testing if you're interested.



Campaign Schedule:



Here, you can specify which hours and days of week you want for your campaign to run.

If you DO want to select specific hours/days, use "Select Timezone" to choose the timezone first.

Often, after you've been running a campaign for a while, when you drill down into "Day" and "Hour" stats in the tracker, you'll see that certain hours of day and days of the week are converting better than others.

Cutting the worst-performing time segments is one way to increase your campaign ROI - this is referred to as Dayparting. However, it would also limit the amount of traffic you'll get. Because we're running carrier traffic, for which traffic volume can be an issue, I would suggest that you not daypart in the beginning unless you already know (e.g. based on stats) that certain hours tend not to convert well, and unless you have lots of traffic at your disposal.

However, if you're targeting something that has a LOT of traffic - such as a big mobile carrier in a massive geo, or wifi traffic in one of the bigger geos - you can potentially start your campaign by targeting the evening hours (local time of your target country), which are sometimes the best-converting hours, when people are off work and relaxing and have time to look at our offers. This can be one way to conserve your budget, but of course any testing and optimization you do would take longer as well.

As for "Set display period", you can leave it unchecked if you want the campaign to run immediately after approval and keep running until you pause it, or checkmark this option and specify a start and end day for receiving traffic:





Ad Delivery Method:



The options are "Standard" and "Distributed".

"Standard" means you'll get the full amount of traffic you're able to get at your bid level, and when your spend hits one of the campaign budgets, traffic will stop.

"Distributed" means that if the budget you specify is lower than what you WOULD be spending if you had chosen "Standard", traffic would be throttled so that you wouldn't spend more than the budget you specified, while getting traffic spread out consistently throughout the day. In other words, you'd be getting traffic "slower" in order to stay within the budget you specified.

If you want to limit the amount of daily spend, BUT want the data to represent average performance instead of being skewed because traffic keeps starting at midnight EST and stopping around the same hour every day when your spend reaches daily budget, then choosing "Distributed" would be a good idea.

If again, you want to limit the amount of daily spend, but you want the traffic to run out ASAP and don't really care about it not being a good representation of average performance throughout the entire day, then you can go with "Standard" delivery. For example, when you're wanting to quickly test an offer to get a rough idea on whether it would convert, then this may be a good option.

For newbies, it would be better to go with "Distributed". This would rule out the possibility of you spending your whole budget with the "Standard" option during hours where conversion rate is worst, e.g. in the middle of the night.



***********************************

WHEW! Wasn't that the longest thing I've ever had to write!

If you're still awake at this point, MAD PROPS and KUDOS! And mega thanks for bearing with my walls of text!

Please don't feel bad if you only understood a fraction of this lesson. You can always come back to reread later on.

After all that reading, I'm sure you're eager to take some more action! The next couple of lessons will be on how to optimize your first campaigns - which hopefully will lead to your first profits!



Amy