What to Look at When You Measure Results
in this next lesson, we'll talk specifically about paid social media and how you can look at measuring results of that as well as how to optimize your paid efforts. Some best practices for paid campaigns is to check in on the campaign at a regular cadence very often when I'm launching a new campaign for a client, I check in somewhere between 24 to 48 hours after launch, just to ensure that the campaign is delivering for various reasons. There have been many times more than I can count where I'll launch a campaign and think that everything is running check in a day later and find out that it's not delivering. Maybe it's because I've set my targeting parameters in a way that doesn't comply with the guidelines. There was one time when I was trying to sell tickets for an event where we had served alcohol and I forgot to adjust the minimum age to 21 for example. So mistakes happen and because of those things, the platforms might not be delivering your campaign. So you want to make sure that...
at the very least it's on and it's active 24 hours later. And then depending on the type of campaign you're running, if you are experimenting where you have a lot of different versions of the campaign and you want to get some quick results or if you have a lot of money invested in the campaign in that very first week, you might want to check in every two or three days just to ensure that the results that you think you should be getting are actually happening and that the campaign is pacing appropriately, some other best practices is that you want to check in maybe about once a week if it's a new client or product launch for the same reasons, just to make sure that whatever your hypothesis might be, is actually being delivered, that the audiences are responding appropriately and that everything is working the way that it should very often, especially if it's a new client or a new audience. People respond to content in different ways and you might find some outliers as you get in there. So it's just important to keep an eye on things and it doesn't take a lot of time. I would say just log in, make sure things are delivering, make sure that you're, you're getting the traffic that you aim to be getting or the purchases that you thought you might be getting. And most importantly that your budget isn't being spent more quickly than you had anticipated. And then finally, if you've been running this for a little while checking in twice a month if it's an existing clients. So if it's an ongoing campaign and it's a maintenance kind of campaign checking in every two weeks to just see how the results are tracking to make sure that things are working is about enough and then I'd say every month, once a month you should definitely generating a report for your client. It's important to compare your results month over month, so that you get about a 30 day read on how things have performed overall as well as how they've performed compared to the previous time period. Some other best practices for optimization is to try to test as often as you can, meaning if you're going to launch a campaign, create two different versions of the ad or create two different versions of the campaign where you might be testing objectives, you might be testing placement or you might be testing audience. So there's a variety of things you could choose to test and even within the content itself, perhaps it's testing two different headlines, two different calls to action. All of these social media platforms makes it very, very easy to do a B testing or split testing. So I highly recommend doing that and I highly recommend running that for at least four days and up to two weeks. Some of the platforms like facebook will automatically help you to create these split tests. If you indicate that that's what you want to do and at that time period when there is a winner. So when there's a campaign that's performing better than another facebook will automatically shut off the one that isn't performing well, another important part of campaign testing is to be mindful of both the number of versions of the campaign that you're creating as well as how much budget you have to be testing. So if you're on a limited budget, it's probably smarter to create fewer versions so that you can really invest enough money into ensuring that the campaign delivers and that you're getting enough of data back so that you can tell which one is performing best. If you are fortunate enough to have a huge budget, then it is smart to test various versions but also be mindful of tracking that and really be able to get enough results from that as well. So the best practice for a big budget would be to put money in and try to get results quickly so that you don't have multiple versions running for a long period of time. As mentioned, somewhere between 4 to 14 days is the recommended time period. I would say that if you have a budget greater than let's say $500 a day for each version, perhaps test that for only four days because with the budget at that size within four days, you should be able to get some initial reads that tells you which versions would be performing better than others when it comes to paid social media and analyzing data from those campaigns. Some things that you might want to look at our trends and performance of those campaigns, paying attention to the segments that you are targeting, paying attention to the audiences that you had aimed to be reaching as well as looking at cost efficiencies. Since you are investing money whether it's a large budget or small budget into the campaign being very deliberate about cost per action cost per click as well as cost per lead are all things that you'd want to track and monitor for these campaigns. Other things you might want to look at as well are any performance outliers. So if you've noticed some trends where things are pretty normal and falling within a certain range, but there's maybe one ad or one campaign that just hasn't performed as strongly as some others, digging more deeply into that to determine why that might be whether it's the delivery mechanism, the distribution channel, perhaps even the scheduling of the campaign or maybe the content itself. But really looking at those numbers to tell a story and to figure out what's what's standing out positively or negatively. Here's an example of how data can sometimes also be misleading. So what you're seeing in front of you now is the result of a lead generation campaign that has run for the last seven days. We were testing audiences to figure out what age group would be most interested in this product. And what I found was that women between the ages of 18 to 24 were most likely to sign up to hear more about this product. So out of the three categories of age buckets that we were looking at, we captured the greatest number of leads from this group of people. However, as I dug even deeper into the data and thinking about cost efficiencies. What I actually learned is that the older demographic age 35 to 44 had a higher conversion rate. So they had a higher rate of sign ups and the cost per lead was lower than the younger demographic. So that's where simply just looking at pure results might actually lead to miss reporting in that our KPI for this campaign was lead generation and so therefore our interest was to ensure that we could capture leads at the lowest price and by looking at the cost per lead that actually gives us the accurate result of which age group we were most efficient in. So my plan now is to continue running this campaign for another seven days. Just to verify that hypothesis. When it comes to this, I do like to be more conservative before tweaking anything. So I'll run it for another week if it still looks like 35 to 44 year old women are responding with a higher conversion rate and a lower cost per lead. What I'll do is I'll shut off the campaign from the lower demographic and focus all of our efforts into the older demographic. A calculation that's very important to work on in partnership with your client is return on investment. You'll get a lot of data from the social media dashboards and sometimes you might need to get revenue generated or number of transactions that were made from your client, especially if it happens offline if you are working with the e commerce client or anybody where the transactions are generated on the website itself. Then, as discussed earlier you'll you'll be able to install the pixel whichever social media pixel you're using onto the website to gather some of this information. It's also a best practice to be able to tag and track this in google analytics as well so that you have not just number of transactions but value of each purchase made. But if you're not getting it digitally then sitting with your client to really understand total revenue generated from the campaign is case. So if it's an offline transaction or if it's a lead captured that might lead to a contract worth $10,000. These are all things that you should be gathering with your client. Once you have that information, the way you calculate R. O. I. On your campaign is to look at the revenue generated or the value of the conversions generated from that marketing or advertising campaign itself subtracted by the marketing or advertising costs only. So here you're looking at a equation for R. Oi where its investment gain which would be purchases made revenue generated. Again, average value of the customers acquired subtracted by the investment cost. In this case it would be specifically paid social media advertising costs divided by that cost times 100 so therefore if you, in the beginning of the campaign, speak to your client and they say we want to double our investment, we want to generate a 200% return on this campaign. You'll be able to also track those goals and ensure that our eye is one of your KPI s to make sure that you're delivering on these numbers. What I'd love for you to do in your quiz for this section is to calculate out the equation that you see and solve the problem that you see on the screen. So I'd love for you to think about the cost of acquiring 393,750 visitors through facebook at a cost of $39,375. With the average order value of each customer being a dollar meaning everyone that's coming through the website is spending on average a dollar on a purchase and we've acquired 393, of them the equations there on your screen and you'll be calculating return on ad spend purely for this campaign. So please leave that in your quiz. And I wish you luck on the equation