{"id":18143,"date":"2021-12-08T17:18:58","date_gmt":"2021-12-08T11:48:58","guid":{"rendered":"https:\/\/www.karooya.com\/blog\/?p=18143"},"modified":"2021-12-08T17:18:59","modified_gmt":"2021-12-08T11:48:59","slug":"ppcchat-mer-vs-roas","status":"publish","type":"post","link":"https:\/\/www.karooya.com\/blog\/ppcchat-mer-vs-roas\/","title":{"rendered":"PPCChat | MER vs. ROAS | 7th December 2021"},"content":{"rendered":"\n

Greetings Happy Readers! This week’s PPCChat session was hosted by Julie F Bacchini<\/a> & guest host William Harris<\/a>. Experts expressed their views on MER ((media efficiency ratio) and ROAS, are they currently tracking MER and\/or ROAS in their accounts, what are their biggest challenges when it comes to tracking and more.<\/p>\n\n\n\n\n\n\n\n

<\/div>\n\n\n\n

Q1: Are you at all familiar with the term MER (Media Efficiency Ratio)?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

Not this nomenclature, but the concept, yes. It feels like digital-only, or small-scale MMM studies. @JuliaVyse<\/a><\/p>\n\n\n\n

New to me & excited to learn more! @sonika_chandra<\/a><\/p>\n\n\n\n

I think I first heard it here on #ppcchat<\/a> a few months ago. Seems like a smart metric, but I haven’t incorporated it into reports yet. @dan_patterson <\/a><\/p>\n\n\n\n

I have become familiar with the term MER, as it has been all over the place in the last 4-6 months! I have always felt attribution was composed at least partly in wishes & fairy dust, so I like seeing new metrics that take more into account! @NeptuneMoon<\/a><\/p>\n\n\n\n

MER is used in contrast to ROAS. MER is a more blended version of ROAS, used to smooth out attribution to be able to give the first step toward a full media mix model, typically focused on digital ads only. @ferkungamaboobo<\/a><\/p>\n\n\n\n

Yes… as I follow a ton of ecom marketers. But less useful for me in healthcare as the “total sale” amount is variable to the insurance. So – a little more delay before I can get the right numbers. @Galliguez<\/a><\/p>\n\n\n\n

Yes, but the name is new. The philosophy is to measure all of media instead of the single-channel view. @lchasse<\/a><\/p>\n\n\n\n

Yes from the Foxwell FB group. Been meaning to dig into it more. @beyondcontent<\/a><\/p>\n\n\n\n

Yes, still working to get clients up to speed & convinced. But we have been reporting on similar metrics for a while so it’s not a huge leap.@selley2134<\/a><\/p>\n\n\n\n

Growing more familiar each and every day, being that I work with @wmharris101<\/a>. One of the things that really helped make the metric click for me was when he used cake as an example in this podcast: unofficialshopifypodcast.com\/episodes\/mer-v\u2026<\/a> @BrettBodofsky<\/a><\/p>\n\n\n\n

Yes, we are… since our agency is 80% ecom and DTC brand work. @duanebrown<\/a><\/p>\n\n\n\n

Same, due to Q4 I\u2019ve not yet had time to lock down the right frame to explain the value prop easily enough. If we can\u2019t explain value easily & simply, it will be rejected as a concept. @beyondcontent<\/a><\/p>\n\n\n\n

Don\u2019t know yet how accurate Google\u2019s modelling is industry wide in Q4 but idea that FB is making up 80% of conversion data regardless of Conversion API is why this is a minefield. @beyondcontent<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q2: Are you currently tracking MER and\/or ROAS in your accounts? Why are you using the method you\u2019re using?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

We track MER as well as ROAS\u2026 as well as aROAS and pROAS, but let\u2019s not get distracted from the main focus, haha. The key thing to realize is that the platforms aren\u2019t TRULY reporting on ROAS. @wmharris101<\/a><\/p>\n\n\n\n

ROAS is a lie. Wait\u2026 what? I go into that in more detail in this video with @kurtinc<\/a> – twitter.com\/wmharris101\/st\u2026<\/a> @wmharris101<\/a><\/p>\n\n\n\n

Report on ROAS by platform & Total ROAS. Also, report on total revenue and percentage of that from our campaigns. This is the next step to incorporate. @selley2134<\/a><\/p>\n\n\n\n

That’s a hard sort of for me. Foot traffic has an agreed on avg value and we report from there, but it’s far from exact. We then bake in sales data and broadcast data after the fact. It’s quite a few inputs. @JuliaVyse<\/a><\/p>\n\n\n\n

ROAS yes, MER no. Though we track it more on a profitability\/margin basis instead of straight revenue @dan_patterson<\/a><\/p>\n\n\n\n

We use the blend of these different methods so we can better triangulate the actual effectiveness of ad spend. ROAS still has a use, it\u2019s just not measuring what you think it\u2019s measuring, so you need to adjust how you use it. @wmharris101<\/a><\/p>\n\n\n\n

I feel like ROAS and MER are better suited to e-commerce, where sales come more quickly and at higher volume. It is trickier to track any kind of return on spend in short windows with lead gen, especially lead gen with long conversion timelines. @NeptuneMoon<\/a><\/p>\n\n\n\n

Tracking both to get the 10K and 1K views of the world….facebook is blind with Apple’s update in April. Have to be careful to not just assume Facebook (or any channel) is driving business if it’s not. What does GA say too? @duanebrown<\/a><\/p>\n\n\n\n

Though I do have a blended PPC\/Remarketing ROAS that I watch. Just need to add a few other channels to be a true MER @dan_patterson<\/a><\/p>\n\n\n\n

I do a lot more lead gen than e-commerce. I really like talking about blended CPA (from digital) because it takes into account non-media spends like SEO, retainer, website costs, etc. Harder at the agency level because of…. agency reasons, let’s politely say. @ferkungamaboobo<\/a><\/p>\n\n\n\n

This is client-specific, but for eCommerce accounts, we try to shift new clients to a MER model for overall direction while setting softer channel-specific ROAS targets so we have a translated number for setting bids and reporting. @PPCKirk<\/a><\/p>\n\n\n\n

I say “softer” ROAS targets, because I think aiming at some sort of target calculated by the client goals \/ margins \/ etc is helpful for us PPCers (esp in bid targets), but should be able to flex as things change in accounts, seasonality, shipping costs, etc, etc. @PPCKirk<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q3: Is MER something you are interested in using in your accounts? Why or why not?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

The obvious answer from me is\u2026 YES! But I\u2019m curious about what all of you think and I\u2019ll do my best to weigh in where I can. @wmharris101<\/a><\/p>\n\n\n\n

Yes! it’s a really good way of understanding the whole channel balance and not just the one magic tactic. Silver bullet thinking will not work often if at all. @JuliaVyse<\/a><\/p>\n\n\n\n

I use (the equivalent of) MER for our e-commerce account but it’s definitely challenging because logistics are different between, say, website sales and Amazon. @ferkungamaboobo<\/a><\/p>\n\n\n\n

Yes, especially as we get more clients on to Facebook. We had been doing mostly Google \/ Bing for ecomm (& generally small orders) so attribution wasn’t as bad for us as most everyone else. Was easier to use & Trust ROAS \/ Profit. @selley2134<\/a><\/p>\n\n\n\n

The hardest part is always just explaining it to clients who are use to another way of looking at success. Education takes times. @duanebrown<\/a><\/p>\n\n\n\n

Yes, I am interested in it conceptually. I think much of the way the “return” or “profitability” of paid search or social ads has been calculated or even thought of has been messed up for, well, ever. @NeptuneMoon<\/a><\/p>\n\n\n\n

100% yes, yes, and yes. We have found it (can, not always) helps avoid the temptation for over-analysis on individual channel marketing tactics and keeps the exec team more focused on a big picture look of what is actually happening across all channels, together. @PPCKirk<\/a><\/p>\n\n\n\n

We in the industry appreciate it needs to become a helpful guideline. But as long as clients believe it is the gospel truth in the platform interface, we have a big challenge on our hands. Clear education piece to invest in or clients get frustrated and walk. @beyondcontent<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q4: Is ROAS becoming more difficult to track in accounts since attribution has gotten murkier with iOS 14.5\/Apple\u2019s ATT? If you’re finding this so, how are you handling this?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

The best way to handle this is by measuring a few different metrics, and using them to create a 3-D \u201cx-ray\u201d of your data. I explain the \u201cx-ray\u201d in more depth about halfway down on this article: elumynt.com\/facebook-roas<\/a> @wmharris101<\/a><\/p>\n\n\n\n

Yes and no. we have agreed on points of truth with our clients, and that has REALLY helped in the ongoing iOS saga. we look at media efficiencies, and then measure sales, recall, and lift. I kind of love this because clients are so integrated! not just a hit send. @JuliaVyse<\/a><\/p>\n\n\n\n

Not as much as most of our clients are Google & Bing & tend to be quick purchase items. It has gotten worse but not as bad as I am seeing for others. @selley2134<\/a><\/p>\n\n\n\n

Spend more time in Google Analytics. Spend more time talking with clients @duanebrown<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q5: What are your biggest challenges when it comes to tracking and\/or reporting on things like ROAS or MER in your accounts?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

I can see it becoming a bit of a rabbit hole… do we include this expense or that expense? Seems like it could get a little crazy to compile if taken to the extreme. What guidelines do you all have for starting? @dan_patterson<\/a><\/p>\n\n\n\n

The biggest challenge to using MER is being more thoughtful about how you run tests. You can\u2019t just shotgun things like a lot of people do. You need to set up a true testing approach, make sure all channels are aligned, and then measure the test correctly. @wmharris101<\/a><\/p>\n\n\n\n

Integrating sales data. we really don’t have clients who are ready to go on in-platform reporting, so the time and effort needed to create these full-program reports is still a chore. but it sure does bring us together, rolling up our sleeves @JuliaVyse<\/a><\/p>\n\n\n\n

Actually, I take that back, the biggest challenge is educating clients on what these numbers actually measure, how they work together, and how to correctly evaluate ad spend after adopting this approach. People have grown to appreciate their ROAS \u201ccrutch\u201d @wmharris101<\/a><\/p>\n\n\n\n

There are multiple complicating factors, especially if you want to have initiatives other than PPC in the model\/mix. But, I think the biggest challenge is a fundamental one and that is backing people off of the “certainty” that they felt with attribution. @NeptuneMoon<\/a><\/p>\n\n\n\n

Education & shifting clients to look at different metrics & more holistic attribution. Seems like this is a big shift for them & even thinking about who they report to. Most clients come to us with ROAS goals (sometimes assigned by their boss) so its a big shift @selley2134<\/a><\/p>\n\n\n\n

One of the historical selling points of PPC has been “you can track what your dollars are doing and put more where the return is highest”. That is still true, but not in the absolute sense it has been perceived as before and the state it is in now. @NeptuneMoon<\/a><\/p>\n\n\n\n

As I understand it MER is challenging to calculate at a very granular level. Good for making channel level decisions, sometimes possible at the campaign level, v difficult at the ad criteria level @RichardFergie<\/a><\/p>\n\n\n\n

The game is “what’s included.” One major challenge with any blended metric is that it’s pretty easy to fudge what’s in the blend. IDK i struggle with any metric – It’s all a game to make numbers bigger, really. @ferkungamaboobo<\/a><\/p>\n\n\n\n

That not everything is 100% trackable. @duanebrown<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q6: How do you think we need to be thinking and talking about attribution and success metrics in the future? How might MER factor into this for you?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

The same way we should have been talking about it – scientifically. We became myopic and trusted what the platforms told us instead of running tests to discern what effect ex. brand search or ex. fb retargeting was actually having on the client\u2019s bottom line. @wmharris101<\/a><\/p>\n\n\n\n

And not to overuse the \u201cx-ray\u201d analogy, but relying on a single data source for attribution just doesn\u2019t cut it, and never should have. We need to do a better job showing how data triangulates an attribution story. @wmharris101<\/a><\/p>\n\n\n\n

Getting back to the roots of marketing, which has always been done by traditional marketers who utilize everything from mail, radio, catalogs, online, etc… Good marketers understand the various channels do not work in a silo. @lchasse<\/a><\/p>\n\n\n\n

I feel like this is getting back to my pre-internet marketing roots in a lot of ways! But thinking about your advertising as a totality and looking at the sales it is driving should be the foundation. Amazing it hasn’t been for a lot of biz! @NeptuneMoon<\/a><\/p>\n\n\n\n

I think automated MMMs are where we need to go as an industry — there’s too much noise + too many gaps in attribution for it to be reliable at any sort of meaningful scale. @DigitalSamIAm<\/a><\/p>\n\n\n\n

I think for non-ecom, even selecting the metric that = success and sticking with it is a challenge. I think we get real about what we’re attempting, and then build measurement around that. (easier said than done!) @JuliaVyse<\/a><\/p>\n\n\n\n

And looking at the types of customers you’re getting via paid channels. Are they your most desired type of customer? Are they your least desired type of customer? What drives them to convert or buy? Sales\/discounts? Something else? @NeptuneMoon<\/a><\/p>\n\n\n\n

Relatedly, as automation + ML become more pervasive, the control lever for brands\/agencies becomes data — so your challenge becomes (1) push conversion points to lower intent levels in the maze + (2) use aMMM to identify value + feedback to platforms. <\/a>@DigitalSamIAm<\/a><\/p>\n\n\n\n

It’s one of many metrics you can use as a KPI. The goal is to find a KPI up-front and stick to it – whatever that KPI is. Otherwise, you’ll (do the equivalent of) p-hack your way to a positive report no matter what you do. @ferkungamaboobo<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q7: Which ad platforms have you noticed being the most affected by tracking and\/or attribution issues?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

The obvious answer is Facebook, but that\u2019s only because they\u2019ve done the best job advocating on our behalf. This affected all of the other (app-based) social platforms just as much, including YouTube, and even affects paid search (to a lesser degree). @wmharris101<\/a><\/p>\n\n\n\n

But what I think people often miss, is that this affects ALL tracking, which includes Google Analytics, Shopify Analytics, etc. @wmharris101<\/a><\/p>\n\n\n\n

And BEWARE of all the snake oil out there\u2026 so many people\/apps claiming they have \u201cfixed\u201d something that all these giant ad platforms haven\u2019t fixed. There isn\u2019t a logical way for them to get around this (that I\u2019ve seen), most of what I\u2019ve seen is pure lies. @wmharris101<\/a><\/p>\n\n\n\n

Facebook. @JuliaVyse<\/a><\/p>\n\n\n\n

It had a lesser impact on a more closed ecosystem like Amazon and using their campaigns, but I feel like it has impacted every platform to some degree. Facebook really pushed it, but everyone has been impacted by it. @lchasse<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q8: If you\u2019re using MER, how are you tracking effectiveness across multiple platforms?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

That\u2019s the money question. The simple answer is testing. Scientific tests including geo-holdouts and other tactics to truly assess incremental performance and trends. It\u2019s not easy, but it\u2019s powerful if you do it right. @wmharris101<\/a><\/p>\n\n\n\n

We used blended totals – so recall lift, brand lift, sales, whatever the metric is. we start there, then dive into each channel to see how it contributed. @JuliaVyse<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

PPCChat Participants<\/strong><\/p>\n\n\n\n