ROI vs. ROAS Understand The Value of Your Marketing Efforts
By Lara Fisher of Blast Analytics & Marketing
Thanks Brian. I’m looking forward to speaking today. I hope you all are enjoying the Virtual Analytics Summit and enjoying all the sessions today. Today, I’m going to be talking about improving your marketing ROI with digital analytics.
There are a lot of things I want to talk to you about today. Primarily, I want you to learn about how you can explore digital analytics to compare marketing campaigns, measure your performance, and improve your return on investment.
I also want to take a look at some core marketing concepts as well as some difference between ROI and return on ad spend. How to calculate both. But fundamentally, I want you to understand the value of your digital marketing efforts and how you can look at monetizing your efforts and make sure the things are properly allocated. And we’re going to talk about attribution a little bit at the end so you have an understanding of why that is important.
Our agenda for today is really six key things.
- The importance of marketing channels.
- Focusing on improving your ROI.
- Making sure you don’t neglect the basics of the how to set up your marketing channels and tracking.
- Pulling through your marketing channels into the funnel.
- ROI versus ROAS and which is the better metric.
- And then talking about the state of attribution modeling in 2018.
So to kick this off we’re going to talk about the importance of marketing channels.
So what are marketing channels? I think a lot of us know the answer to this question. The marketing channels tell you how to get your visitors to the site. For those of us who are using Google Analytics or Adobe, this is set up somewhat differently. In Google Analytics, it’s something you can use straight out of the box. It’s set up for you, but there is a fair amount of tweaking you can do if need be in Google Analytics. Also, in Adobe, there are processing rules you can set up to customize what you want to track.
Now, also note, that Adobe Analytics and marketing, there is a way to kind of set things up out of the box, but in both cases, set up is crucial in ensuring that your marketing strategy aligns with your goals and what your KPIs are.
How do the marketing channels impact my ROI? To put it very simply, they are the heart of your ROI. Your marketing campaign really drives your success. And this is heuristic of how successful you are at driving people to your site. Now, the marketing channels impact the amount of money you spend and the amount of money you earn. Dollars in and dollars out. If you think about that, there is really nothing more fundamentally important to your business than how your spending your money and what you’re getting back on those dollars. So let’s dive into that a little bit more.
There are a lot of ways to improve your marketing channel reporting. Now, if you want to get a good read on getting those dollars in and dollars out and making sure that the dollars coming back in are exceeding what’s going out. You can’t move the needle without the right strategy. So the important thing to remember here in setting up your marketing channels is garbage in, garbage out. The purpose of this session is to help you take out the trash.
So, let’s focus on improving your ROI.
Most of your KPI’s don’t tell you the whole story. In a lot of situations we tend to focus on things that are sort of fundamental but very basic. In media based strategies, you look at impressions, you look at clicks, you look at the click through rate. You use those as your measurer of, “How well did my banner do? How good was my SEO?” Then on the flip side of that you’ve got the people landing on your site and you’re looking at your page views, your visits or your sessions, your users or your unique visitors.
The question in all of that is, “So what?” What does that really tell you about what your marketing channel is doing? How well you’re doing. What kind of money you’re making off of your marketing efforts. The trick is, how do you tie those two together? You always need to ask yourself if your KPI’s are getting you any closer to achieving your financial goals. Two million likes or one million unique visitors are nice, but did your customers do what you wanted them to do on your sit?
We need to focus on three key steps. The how, the why, and the ever important so what.
- First, we’re going to look at the channel people used to get to your site.
- What actions they took.
- And the result affecting achieving your goals.
The how is how people get to your site. I'm sure all of you are familiar with the different channels. Paid search or SEM, organic search/SEO, affiliates, email, social media both paid and earned, direct, referral, display. And the list is getting ever longer every day.
Then the why. What are the actions they took? Well, this is why I care. Either they spent money on us, maybe they signed up for a newsletter, perhaps they watched my video or downloaded a PDF, or filled out an application.
That’s great, but what did that do? Well, this is really why I care. What are the results affecting achieving your goals? So let’s take a look at those steps once again. They spent money with us and it increased my revenue. They signed up for my newsletter and added to our subscriber base. They watched my video and found out more about my new RX prescription drug. They downloaded a PDF and learned more about the new show on our channel. Or perhaps they filled out an application and got one step closer to closing that loan with us.
So that’s really why we care.
But let’s talk about taking those high level concepts of the marketing channels and talking about the how we get to measuring those.
So, there are four key ways to get to a splendid strategy.
- The first thing is, make sure your campaign, source, and medium are consistent and well thought out. I can’t emphasize that enough.
- Step two, is document, document, document. You always want to make sure you’ve got it down and everyone can refer to it.
- Three, this is simple but something often overlooked. Get a second set of eyes.
- And four, keep in mind this is a living, breathing thing. You want to monitor and adjust faithfully.
Let’s talk about this first step about making sure your campaign, source, and medium are consistent. You need to make sure that there are standardized tag names to make analyses easier and more effective down the road. That’s a fundamental thing here. We need to talk about why it’s important to have a standardized tag name. I’m going to go into that a little bit more in a second.
Secondly, you want to share your strategy both internally and with your agencies so there is consistency. Consistency is key here. You want to make sure your agency is on board, all of your internal stakeholders are on board. Because if you don’t align properly, you’re going to have problems down the road. Now, for non digital campaigns, tracking is possible by including utm parameters or directing visits to a specific landing page. But here are a few things you want to remember in setting up a clean campaign strategy.
Number one, don’t use spaces. The example I have here, you want to use “october_newsletter” not “October Newsletter.” Secondly, don’t use punctuation and/or uppercase letters. So again, using that “october_newsletter”, you want to make sure it’s all small case, not capital “October” capital “Newsletter.” Also don’t use campaign source as a description. You want to use “new_release” not “New Release 2014-newsletter.” It’s too much, keep it simple. And lastly, number four here is, don’t create new mediums or include medium in the campaign name. So you want to use “product_launch” not “October 2014 Newsletter” because that’s redundant, and it’s not very specific to what you’re trying to do with the medium.
Next we want to talk a little bit more about the campaign, source, medium being consistent and well thought out. You’ll notice I’m referring to campaign, source, and medium and this largely relates to examples in Google, but this also can be easily applied to Adobe.
So, utm_campaign is a campaign name that should be consistent across all marketing channels whenever is is humanly possible. Again, use that lowercase and underscore to separate words.
Utm_medium, the medium is a channel through which the visitor is coming to your site. So this is what we talked about a little bit earlier. It’s email, display, social, paid social, cpc, affiliate. All of those things. Obviously, cpc is reserved for paid search.
Utm_source. Source is where the visitor was sent from. So, site.com, a database name such as customers for email or sms or email_type or corp-gov-biz or other, all of those things.
Then the last point is utm_content and that’s the only one of these four that’s optional. If there is a need for greater granularity add in the utm_content parament and it will help you look at the different options to have to get down a little level deeper than just what you get from campaign medium and source.
The next point I wanted to make was document, document, document. Now, we’re going to have a handout for you today which talks about all of these things I’m covering right now. So if you need sort of a cheat sheet to take back with you to the office, we’ve got one for you.
This is just a quick little look at social non-paid versus social paid campaigns and how you would set this up. So you’ve got the utm campaign names are the same. So it’s campaign_name in this case. There is utm_medium so you’ve got social utm_medium which is paid social on the right. For source you’ve got social site name and then if you want to use the utm content, you’ve got post_text and then there’s description. So, what you want to do is apply this to a spreadsheet that contains all of your URLS and the QSP’s. So add in the query string parameters, make sure you’ve got exactly what you need. And again, maintain this as a living document.
Number three is get that second set of eyes. Those excel spreadsheets can get onerous and unwieldy, always have at least one person sanity check your spreadsheet to make sure that things are consistent. To make sure that you didn’t accidentally capitalize something, use a hyphen instead of an underscore, got the wrong unit or the wrong date. All of these things are the things you want to do upfront so you’re not scratching your head three months down the road looking at your data and saying, “Where did this go wrong?” So do this every time you make a change, no matter how minor it is, just ask someone sitting in the cube next to you or someone else in your department to take a look at your spreadsheet for your marketing channels.
Then step number four is just to monitor faithfully. Ya know, test, test, test. Make sure before your campaign goes live that your URLs have been thoroughly tested, that the links are working, and I know that sounds really basic, but it’s important. It’s again something, if you monitor the “data uh-oh’s” 3 months down the road after it’s too late to fix, can be prevented. So do that upfront.
Step number four, we want to take a look at pulling through your marketing channels into the funnel.
So, there are three things to talk about here.
- We want a segment based on channel.
- We want to continually adjust and improve.
- And then we want to divide your results into two categories which I’ll talk about in a second.
If you segment based on channel, your funnel is going to look something like this. You have the paid search url, someone clicks on it, they land on your great site, imagine they browse or they leave. You get a bounce and that happens and they exit on that red bucket there. The next item, maybe they put something in their cart or in another case, they download a pdf or they click on your video. Whatever it is, that important action you need them to take on your site to consider it a success. In this case, if it’s a retail site, they check out and you can say, “Yes, I’ve succeeded.”
In terms of the analyzing, adjusting, and improving. You want to ask yourself a few questions. One, did I gain new customers? And if you answer that and say, “Yes, I got new customers, this was a success,” you’ve cut yourself off too soon.
The next question is a very important question is, “Did I gain the right customers?” Are these people who are going to come back? Are they people who are a worthy customer? Did they just come because they had a coupon? Were they just browsing and downloaded something accidentally? Or did you gain a lot of customers that maybe bounced and you’re calling them customers, but maybe they’re not.
The last thing is the penultimate. Will the customer’s return? What is their lifetime value? This is getting into that story that they had about L.L. Bean. At L.L. Bean, you are not really considered a consumer until your second purchase. That says a lot about the way they do business, and obviously, they’ve been in business a long time, but that first purchase is sort of the litmus test of will you make it to the next step? But step number two is that second purchase, and then you’re really the customer and then you start to really show what your lifetime value is.
Next you want to divide your results into two categories. Which are immediate wins, things that happen within 30 days and you get a result you wanted. And then a long term win which is 30 plus days out. Where you get people to do things that are something that engages with you, but maybe you’re not going to receive the results until later down the road. So those immediate wins are things like leads or purchases. It could be a video view or sale or booking, an app download. Those are all something that are tangible, that you see the results and effect immediately. Now the long term wins are where people have said, “Yes, I want to engage with you.” It’s an email signup, social media activity, customer reviews, and loyalty club sign ups are all good examples of people raising their hand and saying, “Yes I want to engage.”
Next we want to look at the differences between ROI and ROAS. Which is the better metric?
Well let’s take a look at the two of them next to each other. In ROI, you have profits minus costs times 100 divided by costs. In ROAS, it’s very simply, revenue divided by ad spend. So, the ROI is measuring how your ads contribute to your bottom line, which is a great thing. In ROAS, the gross revenue made for every dollar spend on advertising is what you’re looking at. So, which one of these is better in terms of marketing channels and tying back to your overall ad spend?
Well, personally, I like ROAS. Traditional marketers are used to ROI, but we as digital marketers need to focus more on ROAS because it is focused on incremental conversions. Incremental conversions are really something that’s very very important. Incrementality measure the lift of marketing spend and gives the conversion rate. So, if you think about it, ROAS is something that you are putting out there that tells you, “Okay, I spent x dollars to get this promotion out there to get people to the site, and I need to look at the lift over people who saw the promotion and didn’t do anything as opposed to the people who did see the promotion and acted.” So that lift is something that’s critical. You sort of have a test of control there, and it gives you a real sense of the incrementality and the lift and the conversion rate and it all kind of comes together which is why ROAS for digital marketers is a slightly stronger metric.
Lastly, we want to talk about attribution modeling in 2018.
There is a lot going on in terms of attribution and how we want to look a these marketing channels and what’s going on and how they’re measured and why they’re worth measuring.
So, let’s start with, “What is attribution?” Attribution very simply determines what media are driving purchases or other key actions. The goal of attribution is to understand where your marketing dollars spent on marketing touchpoints. How did they cause someone take an action on your site or did they take an action on your site?
Now, in attribution modeling, the marketer gets insight into which of their marketing channels cause a positive outcome by looking at the cost of media against the return on dollars spent. So that’s tying back to the ROAS. You’re getting more into the dollar spent and where you’re getting the most bang for your buck.
In our world, I think a lot of us have very much been tied to look at first touch channel or last touch channel marketing. And really this is only part of the picture, and unfortunately, it’s often inaccurate.
So if you think about this journey and you have someone who lands on affiliate site and then does a few searches for your brand and then perhaps they land on your site through a banner ad that they happened to see. Maybe, perhaps, some remarketing. Well, if you’re looking only at first touch, you really only see that affiliate touch. If you’re looking at last touch, you only see that banner ad. But what may have gotten them really revved up about your product were those searches in the middle where they were reading about your product and landed on your site and really interacted with the site a whole lot. But they didn’t really pull the trigger until they got to that banner ad. Well, the question is, how do you account for all of that? Because we’re not in a linear world anymore.
The customer journey is just not linear anymore. The fundamental problem is we have interactions across marketing channels, we have cross device interaction, you’ve got cell phones, tablets, computers, tvs, and as I mentioned earlier, the marketing channels list is growing longer by the day. We need to really beef up how we think about attribution.
So, I think the bottom line is a marketing measurement revolution is upon us and it’s time to take action.
We’re going to talk a little bit about Google Analytics versus Adobe Analytics because both have been very active this year in really trying to get their attribution modeling where it needs to be in today’s day and age. Google Analytics now has Google Attribution 360 which is, from all we’ve known, Adometry. Google Attribution 360 helps you optimize your marketing program. So, I think Google has coined the term “micro-moments” and that describes the fractions of time people turn to a device throughout the customer journey. So, you know, you can be on your cell phone and then maybe you switch over to your laptop and then later you’re on your couch with your tablet. All those little touchpoints are something that we have not really been able to consistently track until now.
If you look at the Google Analytics 360 and the Attribution 360, digital attribution modeling includes all the touchpoints on a customer journey. Goes beyond first-touch, last-touch, or some arbitrary attribution models we have out there. That’s a great thing. That’s good new for Google.
In terms of the marketing mix modeling, this also integrates offline channels such as radio and print into the mix for an aggregated view. This is also wonderful news because it’s something we haven’t really touched on a whole lot. But as much as we in the digital world want to pretend that the offline channels don’t exist, they’re still out there with a vengeance.
And then lastly, you’ve got TV attribution. Are the TV ads impacting sales? So the Google Attribution 360 combines ad airings data with search data for analysis. So a lot of great things going on with the Google Analytics Data and Attribution 360.
Probably a lot of you are asking, “Okay that’s great, but I use Adobe. What do I do for attribution now?” Well, as you know, Adobe announced Attribution IQ. There are ten different models you can use in Attribution IQ to take a look at the different touchpoints that you’re looking at in the attribution modeling.
The ten different models, you’ve got First Touch and Last Touch, which we’re all used to. And if you look at this little example we have in the right hand corner, the bars kind of tell you how this is dispersed. So First Touch you see the large bar of how first touch is attributed basically everything. Last Touch is sort of the inverse where the endpoint is attributed to everything. Then they’ve added Linear which spreads things across all different channels equally. U Shaped gives allocation to the first touch and last touch more heavily than the middle touches, but they’re still accounted for. Time decay, you can see, it allows time to elapse over how the marketing touches have been shaped. Same Touch marketing, J Shaped marketing, Inverse J, Participation, awesome new stuff from Adobe that are worth looking at. But the thing that I think is the most worth looking at is the Custom attribution modeling which allows you to allocate different proportions to marketing channels in your campaign and allows you to adjust and weight appropriately.
In attribution IQ, each model captures different customer touchpoints. It enables you to dive into the different campaigns and promotions on a very granular level. Any dimension, metric, or event can be applied to the models both in real time and retroactively which is nice. You can also assign the real impact of channels and show differences amongst individuals, campaigns, products. And Attribution IQ, as I mentioned, has ten different models that allows for customization. So there are some great things going on with Attribution IQ as well as GOogle Analytics 360.
So, in closing, I think in my 23 years of experience having worked in digital analytics and research, this is a very exciting time to be an analyst. The Attribution IQ allows you to go ahead and start really getting into your marketing channels and understanding how the funnel works, how your dollars are being spent, and what you’re getting back for those dollars. As well as the Google Analytics 360 Attribution, there is just a ton of new stuff out there for us digital analysts to play with. I think it’s an exciting time to be working in marketing channels and attribution.
So, thank you for your time today. I look forward to taking your questions in the Q&A, and I’ll turn it back over to Brian from ObservePoint.