Which Metrics Really Matter in Digital Advertising?

Filtering vanity metrics to focus on real commercial KPIs, attribution models, and incrementality in paid media.

Speakers: Sevda Yurtvermez, Onur Durdağı

28 May 2024

Video Transcript

⏱ 13 min read• 2,336 words

Read the full speech or switch languages

— Hey everyone, welcome to a new episode of BPC Talks. Today Onur and I are talking data-driven performance. We'll cover why measurement and optimization matter in digital advertising. Welcome, Onur.

— Thanks for having me. Hi everyone.

— Before we get into why data-driven performance matters in digital advertising. Let's start with the basics. What is measurement, really? Measurement is a tool that tracks when a user completes a specific goal or action. Measurement is the single biggest thing separating traditional advertising from digital performance marketing. Because we can measure, we work differently on this side. In digital performance advertising, we watch user behavior with different strategies and analyses. Then we optimize our ad accounts based on what we see. So first up, what are the measurement metrics? How should we read the data we see in advertising? And how do we build strategies for those campaigns? Let's start there. What are the measurement metrics, and what do we watch for in each one?

We can start there if you want, Onur.

— Sure. At the heart of analyzing data and measurement sits conversion. That whole idea of a conversion. And conversion ties into a bunch of metrics. What those metrics are worth. How you read them. How efficiently we spent the budget. What kind of profit margin we ended up with. That matters a lot in digital marketing. Click-through rate leads the pack. CTR, or click-through rate spelled out. It tells you how many times an ad got clicked versus how many times it showed. Clicks create value in every conversion campaign. They do the same in impression campaigns.

Because it shows how much curiosity you sparked in the audience. It measures how strongly users reacted. That makes it one of the serious numbers to watch.

— Click-through rate, like you said, isn't just an e-commerce thing. It matters for our form lead clients too. We need to measure how well an ad performed. Text ad or image ad, doesn't matter. How many clicks did it pull? Then we take action based on that. So we can't overlook click-through rate either.

— Absolutely, Sevda. Click-through rates tell us a lot. They give feedback on the whole advertising industry. Digital advertising too. They show the creative is good. They show the copy works. They show whether we're reaching the right audience. Of course, we have to set some things aside. I'm talking about a real, genuine CTR here. Not clickbait. Not a bot attack. If it happens that way, it's a serious signal. The second metric is conversion rate. Conversion rate shows how many site visitors actually converted. It tells us how effective our ads really are. It's a clear indicator of how well the whole thing performs. Whatever we measure and analyze as a conversion. Conversion rate shows how efficiently that output gets used.

Then there's cost per conversion. It's a metric showing how much we spend to get a conversion. People call it CPA. Cost per action. Cost per conversion is basically what one conversion costs us. And like you said, this comes down to measurement. Conversion measurement, specifically. It could be a newsletter signup. It could be a form fill. On e-commerce, it could be a purchase. We track how much ad spend went into that. A purchase event, a form event, whatever it is. That gives you a cost figure. Cost per conversion. The lower your CPA, the more efficient you are. That's what cost per conversion tells you. So it's one of the most valuable metrics. Especially when we need cost-focused outputs. Especially when the analysis hangs on that.

— Yes, exactly. We can actually see what a single conversion costs us. And we have to build our strategy around that number.

— Yes, absolutely. Sevda, you've probably noticed it too. Plenty of e-commerce cases. Other form conversion actions too. If we don't measure that in the cost math, we're flying blind. We need to know the cost per conversion. We have to track the conversion itself to know it. So the heart of that analysis and tracking is measuring the conversion metric. That means setting up the Pixel and the other measurement tools. Only then can we calculate a real cost per conversion.

— That's the other thing brands really watch. How much am I spending per conversion? So the goal here is to push that number down. Bring the cost per conversion lower. That way you get more conversions on the same budget. Once we can measure that data, the next question follows. How do we improve the actions we take with better strategies? How do we get more conversions while spending less? It's one of the first metrics we look at when deciding those actions.

— Absolutely, Sevda. Cost per conversion tells us a lot. Is our spend moving in the right direction? Are we building the right milestones? Are we squeezing maximum efficiency out of a campaign? It gives you a serious signal on all of that. I'd call it one of the most important metrics here.

— Right, and that leaves ROAS. ROAS is one of the top metrics for e-commerce brands. Same for e-commerce campaigns. It shows how much we spend on ads. And how much we earn back on that investment. Like, I spent 1 on ads and made 5. Or I spent 1 and made 3. So we keep optimizing our strategies along those lines. For e-commerce brands, ROAS is the headline metric here. Okay, so we've talked a bit about conversions. Why they matter, and the conversion metrics. When we start managing a brand from scratch. The first step is adding those conversion codes to the web site. They have to actually work. Otherwise we don't get the data.

How do you set up those conversions when you're starting a brand from zero? Like, through the web, through GeoForge. What variables come into play for e-commerce or lead brands? We could touch on those briefly if you want.

— Of course. Sevda, look at social platforms as a whole in digital advertising. Google Ads, Facebook Ads, TikTok. Every one of them has a tracking tag. We add a unique conversion code to the site through that tracking tag. Then we handle the setup in tag manager. We do a separate setup for each platform, and that's how conversion tracking gets built. The details go a lot deeper. I don't want to confuse the listeners or the viewers. Those tags let us track the shopping behavior of the customer, the user. Right from the site. We measure it directly, and that's what makes analysis possible.

Like I said, this applies inside Google Ads, Facebook Ads, TikTok and similar platforms. Yandex too. So those tools matter a lot. The next phase is web analytics tools. Once we've done the analysis and the tag tracking, we need to check it. Was it set up correctly? And how do users actually behave? Web analytics tools show us where things come from and how. GeoForge leads the pack here, what we call Google Analytics 4. These tools are non-negotiable. We analyze exactly which source the traffic came from. Which device it came from. Which segments drove the purchase behavior.

More precisely, we classify all of it. Then we turn that classification into a report and a presentation. That's where we show how efficient our investment really was. And how usable the whole thing became. GeoForge makes that visible. On the other side, there's event and goal event definition. Which actions count here? A thank-you page visit, maybe. Or a user behavior like subscribing to the newsletter. We measure those through GeoForge in our reports. Reports like Looker Studio. Could be ecommerce behavior. Could be a form conversion in conversion tracking. Could be the newsletter. Could be a wider conversion rhythm like click conversions. To measure those rhythms, you define those pages. And measuring with GeoForge is genuinely essential.

That's roughly how these things come together, Sevda.

— Yes, you brought up a really important point. With GeoForge, a digital performance marketer doesn't just watch panel dashboards. They also track user behavior through Analytics. And the events you want to follow have to be set up correctly from day one. Take an e-commerce brand. Add-to-carts, cart abandons, the checkout page, time spent on specific pages. How many pages someone browses. You can analyze all of that in far more detail. Then you feed it back into your ad campaigns as strategy. Of course, like we said at the start. The Analytics setup has to be done first. The conversion tracking for each ad platform also needs to sit on the web page. Once all of that runs correctly, we can actually see these reports and analyses.

Reading and interpreting it matters just as much. So: which pages are we tracking? Which actions did we want on those pages? And what are we trying to do? That's the path you follow on this part. Alright, say we finished every conversion setup. The event setups here got built. Conversion tracking too. Then comes optimization on those campaigns. How does that actually run at a given brand? How do we read this data? How do we analyze it? Let's cover that as the last stage.

Sezi, I'll hand it to you. Take a brand. An e-commerce brand, let's say. No problems with conversions here. Success means lifting sales volume for a specific product or category. Lifting conversion volume too. And there's a nice campaign running. We've got a 50% off campaign in that category. How would you handle tracking here? What about interpretation and optimization? Walk me through the scenario.

— First off, in an action like this, you're reading purchase behavior. That's the last leg of a funnel strategy. Optimization is one of the most strategic phases of this tripod. It has to be built well. It's the final phase, really. But for strategy development and improvement, campaigns matter a lot here. Say we have an e-commerce client. A campaign setup was done for them. Conversion tracking or events got installed. E-commerce behavior data started landing in our panel, bit by bit. From there you run performance analysis on keywords. If there's a search campaign, on audiences too. Sorting efficient keywords from wasteful ones matters for improving performance.

Negative keywords matter a lot here. Test the ones with high CPA and low conversion rates. Then filter them out. Sorting your keywords is a critical early phase. That's if we're talking about search campaigns. Next up is budget adjustment. That means checking whether the campaign spends its budget efficiently. It's one of the big pillars of this optimization process. For budget adjustment, first check if the campaign is even spending the budget. Once it spends, check if it spends efficiently. If it does, make some budget tweaks to push it further. The panel sometimes suggests daily budgets. You can probably improve on those.

It ties directly to search volume, and performance follows that. So optimizing the daily budget accordingly really matters. Another phase is ad copy and creative updates. We talked about metrics with low CTR. And we said how important CTR is for every campaign. For ad copy and creative updates, take the campaign with low CTR. Swap in a different, stronger ad copy or a better creative. Optimize it. Run tests around that. It's key to keeping the campaign efficient and running at its best. Another breakdown applies to e-commerce and every other campaign too. Targeting improvements. At the top of that list: age, demographics, shopping behavior. Then the audiences we call detailed targeting. In-market or affinity segments.

Inside those, we look at buying behavior. Is that audience performing well or not? We can improve from there. We can tune the demographics accordingly. Another option: look at the hours when people actually spend. Ad scheduling lets us push the most efficient hours forward. If certain districts or cities perform well, we can raise budget there. That's a customization practice we can run on campaigns. These are serious optimization steps. Another big one is changing the bid strategy. Google Ads has strategies like cost per conversion and maximize conversion value. If we're talking Google Ads specifically, those matter. Or target ROAS. These carry real weight on the bidding side.

Manual bidding, target CPA, maximize conversion value, updates like that. Making the bid strategy type efficient is huge, Sevda.

— These are the core steps. This is what a performance marketer has to do on account optimization. Beyond all that, say we're an e-commerce brand. Say we're running a discount campaign in a specific category. If conversions come in under target, the first place we look is our best sellers. We check for broken size runs and stock issues. Because we launched with a great campaign. We did every optimization we needed to do behind the scenes. But it stalled out somewhere. So we ask: do our best sellers have enough stock? Are the key sizes actually available in that category?

Say we have a best-selling model. Size M sells the most. But M is out of stock now. Only XS or XXL are left. That alone can drag sales down. So as performance specialists we watch for this. And we bring it straight to the brand. Yes, we ran a great campaign in this category. But we don't have enough stock here. It's tied to the top sizes on our best sellers. That can hurt the campaign's results. So we raise it with the brand. Then we take action on it together.

Nicely put, thank you so much. Today we covered data-driven performance marketing. We talked about the metrics and what you can do inside accounts. That was our topic for today. Want more content and different topics? Want to get the info right away? Subscribe to our YouTube channel. Follow us on our social media platforms too. See you in the next one. Take care.

— Take care.

This transcript was generated using AI speech recognition and verified against subtitle style rules.

Yiğit Konur

Content author

Yiğit Konur

both

Contact us

How we handle your data

Open as a full page