Ana Kravitz is a web analytics master. She worked with Condé Nast, Barnes & Noble, and as a Google Analytics and Google Tag Manager consultant for hundreds of clients. Right now she’s focusing her efforts on her own project, Mixed Analytics, developing Google Sheets add-ons that are used by over 100K people.
Ana also worked as a Senior Online Marketer. Now more than ever, understanding data is a key skill for marketers. We wanted to hear Ana’s thoughts on the topic – can marketing decisions get way smarter when marketers improve their analytics skills?
Ana, do you feel that there are two fronts out there, in a company: the Marketing front and the Analytics one?
Can you draw a line between them, or do they intertwine?
While there are of course aspects of both fields that don’t intersect, I don’t think marketing can operate independently from analytics. Analytics is key to marketing, so much so that a lot of companies even locate analytics within their marketing organization. Anyone running marketing campaigns needs to be comfortable checking and understanding performance metrics to see what’s working.
Also, we could say that all elements of a business get measured, but there is something special about marketing. For example, I think it’s much more common for companies to add new features to their website without measuring the impact than it is for companies to send out an email campaign without checking the results.
This isn’t really due to a lack of interest in the data in one department versus the other, but it seems related to the analytics-driven culture of marketing, plus the fact that it generally requires a much higher level of technical effort to implement site analytics. Digital marketers are in an enviable position where most marketing platforms provide some type of analytics out of the box.
What about this intertwine between marketing and analytics in a tiny company? How can you look at data, when there's not enough of it?
I‘m not sure lack of data is a problem, even at a tiny company. I have a tiny company myself, but I still have a huge amount of data from my website (tracked in Google Analytics), sales data (tracked in Stripe), and various marketing services (Facebook, Google Search Console, etc).
Still, I do lack data about marketing channels that I haven’t delved into yet, so at some point, I will likely run some small tests with paid channels. It’s pretty easy and inexpensive to run tests. And then there will be more data 😃 .
We often hear about T-shaped marketers, growth hackers, marketing specialists, and marketing generalists.
At the end of the day, which should be the most valuable skills of a marketer? Are web analytics important to marketers' toolkits?
I’m biased, as I’m coming at this from a web analyst’s perspective, but YES I do think web analytics knowledge is a must for marketers. Having a solid understanding of analytics is critical when checking results and developing campaigns, and it’s not practical or efficient to route all analysis through a separate analytics team.
Many companies struggle with how to structure analytics in such a way that analysts are able to support the teams that need them. But you can’t get closer than doing it yourself, and you don’t need to be an expert analyst to check your campaign results in Google Analytics.
Not to throw more on the plate of the marketer, but I also think online marketers should have some basic understanding of how to tag marketing campaigns. They don’t need to do it themselves, but this way they can recognize if something looks off (like an email campaign missing “UTM” strings), and communicate tagging requirements effectively.
You asked about the intertwining of marketing and analytics, and I think we could ask something similar about the intertwining of tech. A dash of tech understanding goes a long way.
Not to throw more on the plate of the marketer, but I also think online marketers should have some basic understanding of how to tag marketing campaigns.
(…) this way they can recognize if something looks off (…) and communicate tagging requirements effectively.
What do you think the ultimate goal of an online marketer is?
How do you think it will transform as new technologies emerge or as marketers learn to work with more tools, like Tag Manager, for example?
In one of your articles on the Mixed Analytics blog, you’re saying: “But what if… you had to implement e-commerce tracking without any outside developer assistance, entirely through Google Tag Manager?”
Tag management tools are awesome for getting around development bottlenecks, as you can insert and customize tags yourself. It’s pretty powerful stuff.
However, I think tagging should generally be left to specialists (in the same way an implementation specialist probably shouldn’t be running the email campaigns). With that said, it’s absolutely important for marketers to have a basic understanding of tagging so they can effectively communicate their requirements.
Also, in a smaller company, a lack of resources and the wearing of many hats may produce a situation where the marketer is in fact the one implementing tags themselves (this is the situation I was referring to in my article). This might not be ideal in the long run, but for people who find themselves rising to and enjoying the challenge, it can also be the gateway towards a more technical career. I’ve seen that happen more than once!
Sometimes we draw the wrong conclusions looking at people’s behavior on our website, when in fact we did not attract the right audience.
How can a marketer discern that it’s not the behavior on the website that’s flawed, but the targeting is?
In cases like these, I think it’s all about segmentation. We can differentiate factors by creating user segments such that there’s some benchmark to compare to.
If you see high bounce rates and low conversion rates from users who visited from Campaign A vs. Campaign B, that’s a pretty clear signal that Campaign A is attracting the wrong audience. If metrics are negative across ALL campaigns, that’s a signal that something is wrong with the site.
What are the Analytics reports a marketer shouldn’t live without?
Google Analytics’ Traffic Sources report is where I always start. Its greatest flaw may be its one-size-fits-all attribution model, but this simplicity and even-handedness is what makes it so popular.
One interesting experiment is to enter Facebook Ad Manager, Google Ad Manager, your email platform, etc, and add up all the transactions you find within their reports. You will almost certainly arrive at a number far greater than your actual total transactions since each platform attributes conversions in a way that gives itself the greatest amount of credit.
On the other hand, Google Analytics treats each platform in exactly the same way, so you can compare across channels. And, once you fully understand the traffic sources report, and start to recognize the limits of that default attribution model, that’s the perfect time to move onto the custom Attribution reports.
As we all know, proper tracking helps decision-making, budget spending, etc. How do you think Google’s Consent Mode will affect the way marketers look at data and make decisions?
From what I understand, Google Consent Mode is a way to more easily adjust the behavior of tags based on the consent status of users. I’m not sure yet how this will impact data since it doesn’t seem like a major change to what we see around the web now (cookie opt-outs that block tracking under GDPR).
But in general, I think the benefit to marketers of letting people easily block invasive tracking is that when they know they aren’t being tracked and followed around the web, they are less likely to simply install an ad blocker and block everything.
Marketers benefit greatly from having aggregated data on the performance of their campaigns, and they don’t need to know the details of each individual. So I’m for anything that helps us get useful, aggregated data while respecting people’s privacy.
Thank you for sharing your knowledge and your time with us, Ana!
Our last question - what’s the north-star data in your daily life, the data you feel your life depends on 😊 ?
Thank you for the awesome questions! For daily data, I have to go with all the data in Google Maps.
Between directions, estimated trip durations, and train / bus schedules, this is probably the single most useful app I have on my phone. I haven’t gotten properly lost in years now.