What I Love: Adobe and Google Analytics*
While in Atlanta last week for ACCELERATE, I got into the age-old discussion of “Adobe Analytics vs. Google Analytics.” I’m up to my elbows in both of them, and they’re both gunning for each other, so this list is a lot shorter than it would have been a couple of years ago. To wit:
- Cross-session segments are in both!
- Both enable multi-suite/property/view tracking!
- Sequential segments are in both!
- Custom dimensions/variables and custom metrics/events are plentiful (not that users won’t always pine for more)!
- Classifications/dimension widening is in both!
- They both have reasonably robust eCommerce tracking, including a native concept of “product” and “cart” (yeah…that’s a new one for one of these guys)!
Are these features identical in both platforms? Of course not! Does one platform handle any of these capabilities in an empirically better way? Perhaps…but the comment trolls and platform homers will take over the comments if we let ourselves stray down such a path. Let’s just say that they both provide these capabilities and accept that personal preference and “the-tool-I-learned-first” quickly enter the picture and make such a debate subjective. Let’s call it close enough to binary parity and move on.
Other historical knocks against one platform or the other are rapidly going away, too:
- Sampling is becoming less and less frequent an occurrence in Google Analytics
- Correlations and subrelations are increasingly available between any two props/evars in Adobe
- Google Analytics has gently eased its terms of service as Universal Analytics continues to get broadened so that user-level tracking is allowed (as long as reasonable privacy lines aren’t crossed)
- Adobe has drastically simplified their product/pricing model so that users just get many of the most powerful features that used to require additional expense
Right?! Convergence ruuuuullllles!
Nonetheless, I made a bold claim last week that I could write a brief post that hits my highlights — geared towards analysts, and avoiding topics that are more religio-philosophical than fact-based — as to the capabilities of each tool that, to me, are meaningful differentiators between the platforms. And, this is my attempt to back that claim up, listing these capabilities as positives of what each tool has that stands out, rather than as them’s-fightin’-words-type criticisms of either platform.
(As it turns out, I was wrong when I claimed I could write a brief post. I circulated the initial draft internally among the Demystified partners and Team Demystified…and the end result is a much more complete — not as…er…”brief” — and clear work.)
Things I Love That Google Analytics Does
Here’s my list of some features I use all the time in Google Analytics that might be lacking in another platform:
- Multiple Segments — applying up to 5 segments to a view in the tool’s web interface (I almost never use 5…but THREE is a magic number!). “In the web interface” is the operative phrase here — more on that in the next section.
- Retaining Segments — when multiple segments are applied in the tool’s web interface, Google Analytics is really good about retaining those applied segments no matter how you click around among reports; you’re stuck with them until you click to remove them!
- Segment and Report Template Sharing — sharing segment and custom report templates (“templates” is the operative word) with a simple emailed URL (read it again: sharing templates for segments and reports; not the segments and reports themselves).
- There Is a “Free” Version — yes, “free” is in quotes, because I’m just talking about the licensing fees. But, for a company on a tight budget that is looking to get off a dying platform or that, somehow, doesn’t have web analytics on the site yet, the availability of a free platform removes one barrier to getting internal backing for the effort. Plus, from a talent pool perspective, the existence of a free option means lots of analysts and marketers can get their hands dirty with the platform on small sites and be able to hit the ground running faster with analytics for larger, more complex sites.
- Automatic Adwords (and DoubleClick) integration — Google owns these products, of course, and the very existence of Google Analytics is driven by that fact, but that doesn’t change the fact that it reduces the need to implement campaign tracking variables for a, generally, significant traffic source.
- Google Spreadsheet Integration for Free — there’s a handy Google spreadsheet extension that lets you pull data straight into a Google spreadsheet.
- Captures the URL and hostname of the Page by Default — URLs still matter, so having the URL (with gratuitous parameters stripped appropriately in the configuration) readily available is super handy. And the hostname is useful to be able to easily get to for multi-domain/subdomain sites and to figure out if another site (and which site) has inadvertently hijacked your tag.
Things I Love That Adobe Analytics Does
Here’s my short list of some features I use all the time in Adobe Analytics that might be lacking in another platform:
- Ad Hoc Analysis / Discover — this has to be at the top of the list. I go for days without getting into Reports & Analytics, and that’s as it should be. Slicing and drilling, comparing different segments side by side, quickly trending a specific item that I’ve drilled down to, and so on. I’ve said it before, and I’ll say it again, if you are an analyst using Adobe Analytics and you spend more time in Reports & Analytics (SiteCatalyst) than Ad Hoc Analysis (Discover), you really, really should be kicking yourself. If you match this description, and you’re at eMetrics in Boston and are willing to sign a waiver, I’ll happily deliver the appropriate kick in the pants for your poor judgment. (Business users heads still spin when they dive into Discover, so I don’t think Reports & Analytics can go away…even as I hanker for extending its capabilities).
- Hit Containers — It took me a while to fully grasp the robustness of the Hit/Visit/Visitor segment paradigm in Adobe Analytics when it was rolled out, but I find myself using all three container types all the time! Hit (page view) containers, in particular, stand out as being a unique plus to Adobe Analytics.
- Calculated Metrics — These can be created by individuals for their own temporary (even throwaway) use, or they can be deployed to all users. Very handy!
- Segment / Dashboard / Bookmark Management — Being able to actually share these items (rather than templates of these items) comes in extraordinarily handy. And, giving the user the option to either copy or simply link to dashboards…is genius.
- Segment Stacking — the ability to apply multiple segments at once (“I want to see only the visits who are First-Time Visitors — one segment — and who did not purchase — another segment — without building a whole new segment.”) This used to be an Ad Hoc Analysis-only thing, but it’s now available in Reports & Analytics and Report Builder. Woot!
- Pathing on any traffic variable — because Adobe Analytics has a long-standing history of extensive custom variables, the ability to do pathing on those variables pops up as being super handy when you’d least expect it.
- Excel Integration for Free — Adobe Analytics comes with Report Builder, which enables a high level of control over what data gets pulled into Excel, when, and how (and enables scheduling of well-formatted results).
- “Page Names” decoupled from “URLs” — a page is a page is a page…and being able to build a meaningful taxonomy for your pages that makes that core unit neither too granular nor too broad is pretty powerful flexibility.
The Thing I Would Love but Am Unable To
I’ve got to put one item on the list that doesn’t exist in any web analytics platform: better data visualization. Choosing from a 2×2 or 2×3 grid (or even from a longer list of limited layouts) and then dropping in widgets where I can control substantially less of the presentation of the data than I could manage with Excel 95 is embarrassing. Especially since there are oodles of third-party visualization platforms that are built to be embedded in products. I don’t get it. I’m forced to pull data into third-party tools (for me, that’s generally Excel, but plenty of people use Tableau for the same reason — and good on ya’, Adobe, for supporting that with the Tableau export format!) if I want to deliver recurring reports to business users that can actually be meaningfully understood. I need better control of:
- The layout of widgets — not necessarily pixel-level granularity, but give me a grid that is at least 25 columns wide
- Labels and dividers — grouping and organizing of content
- Trending of data — chart size and style, axis label display and formatting, line/bar color
- Sorting control — for lists of numbers…and let me decide if I want to include a gratuitous “% of total”
- And much, much more…
I’ve bought a copy of Stephen Fews’s Information Dashboard Design for a web analytics product manager before, and I’ll happily do it again. Let me know if that’s your role and you’re interested.
So, Clearly, the Better Tool Is…
…really? Surely, you didn’t think I was going to bite on that, did you? I’ll even stop short of, “It depends on your needs.” I’d hazard that, north of 75% of the time, a company could flip a coin between these two platforms and then invest all the time they were planning to put into an exhaustive RFP process, instead, into finding people who really know the platform inside and out to implement, maintain, and use the tool. And, they’d come out ahead of the company that obsesses about which of these platforms is “right” for them (they’re both right, and they’re both wrong!).
As Adam Greco says in his Top Gun training classes, “It doesn’t matter what tool you’re using if you’re only using 20-30% of its capabilities. Make sure you have the people and the commitment to not only learn and apply the full power of the tool now, but to stay current on new capabilities as competing platforms chase each other.” Competition is stressful for the vendors…but it’s a boon for analysts!
* This post doesn’t attempt to cover all web analytics platforms. There is no mention…except in this footnote… of Webtrends, IBM Coremetrics, Mixpanel, AT Internet, KISSmetrics, Piwik, or any of the other potential dozens of vendors who may comment here about their particular platforms. I’m sticking with what I know and what we see with our current client base. It’s in the post title.
Image Source: Flickr / Klearchos Kapoutsis