Naked Marketing: Big Data Marketing Checklists for Marketo and Adobe Analytics
In the first five posts in the Naked Marketing series—a big data marketing odyssey, we looked at the big picture, the team, the foundations we built, the business case and the big data marketing tech stack.
If you read any of these (especially the last two), you saw how important it is that any big data marketing program is supported by data that’s consistent and high quality.
A big part of ensuring consistent data is to create data governance processes and policies and then make sure everyone follows them religiously. The other big part is correctly setting up your CRM, marketing automation, and analytics to capture and analyze all that relevant data (fields!) and report (filters!) it correctly.
I don’t know about you, but when I think about making complex processes simpler, I think checklists (a favorite book is Atul Gawande’s The Checklist Manifesto, all about how simple checklists can make dramatic impacts in complex systems).
Since we use Marketo and Adobe Analytics (formerly SiteCatalyst), our two most important checklists are for those two tools. If you use these applications, you may want to use these checklists too, adapting them for your own processes.
But even if you don’t use either of these tools, the checklists will still give you a good sense of the kinds of set-up and governance policies you need for accurate big data marketing.
Fortunately, Anish Jariwala, our Digital Marketing Strategy and Analytics Manager, has a wealth of experience with both Marketo and Adobe Analytics. So he drew up these two checklists to keep us all on track—and keep our data healthy. We’ve annotated each list to help explain the thinking behind them.
I hope you find them helpful. And do let us know in the comments below if you’ve got some things you’d want to add.
The Marketo Data Governance Checklist
First up is our Marketo checklist. We use Marketo to track anonymous web visitors and execute and monitor campaigns and ongoing content marketing nurture flows.
If you’re deploying Marketo, run through this list and see how you’re doing. Skipping any of these could mean you’re storing up problems for later.
Notes for the Marketo Checklist
- Do all paid programs have period costs?
We run many paid marketing programs such as PPC, offline events, social advertising and retargeting. It’s important to capture the total cost of each marketing activity (a rough approximation is fine) so you can do ROI calculations: the revenue contribution (by particular program or channel) divided by costs.
We’ve seen people in meetings say, “Look: that content got 200 downloads, let’s do more like it!” without even knowing how much was spent to get those downloads. That ‘winner’ could be a loser in terms of real ROI.
- Do all relevant programs have the correct Channel and Success statuses defined?
Every program in Marketo has to be assigned to the right channel and have the right success statuses. For example, for an email program, success might be clicking a link in an email. For tradeshows, it could be attending a specific session or visiting our booth. If channels and success statuses are not sorted out before you go live, than you can’t capture the whole value of your programs in Marketo.Do this up front because it’s very hard to fix these issues once you’ve gone live.
- Have you set up exhaustive programs tags?
Tags are the easiest way to flag all of your programs in analytics, across various dimensions such as regions, themes, marketers, etc. Once you have tags, you can slice and dice the analytics for programs by these dimensions.
If you don’t have the tags, you can’t get the insight.
- Have you set up your website-related programs?
Let’s face it—traffic acquisition and the quality form fills happen on your website, including Contact Us, Pricing Pages and Product demos. Between 60 and 80 percent of New Name acquisitions happen on our website.
Prospects who are aware of your company come to the website directly—it’s “free” traffic and it’s often the highest quality and highly engaged. You need to have an exhaustive setup of Marketo programs on the parts of your site that drive lead generation. Many times, companies set up outbound programs in Marketo such as email, but ignore the inbound programs or channels, such as their own website.
- Are contact roles attached to every opportunity?
It’s very important that Opportunity Contact roles (that is email addresses) are explicitly attached to every opportunity object in Saleforce.
If sales folks don’t tie an email address to each opportunity, it’s very hard to see which Marketo programs contributed to a given opportunity. To mitigate this problem, we use account-based attribution in Marketo—it’s not precise but it helps. (We have some other solutions in the works to mitigate this challenge, so stay tuned.)
- Check the correct set up of Opportunity object, Opportunity amount, and Expected Revenue.
Salesforce gives access to only a few Opportunity fields to a Marketo user. It’s very important to understand these fields and their implications for marketers.
For example: you need to know the currency used in the “Opportunity Amount” field in Salesforce. If you’re a global company, regions could be using their local currency to populate this field. We use a converted currency field to map to Marketo’s Opportunity Amount so we can do an aggregated analysis in one currency.
- Is “Acquistion Program” set up for all leads?
Every lead in Marketo has to have an acquisition program. Acquisition credit is given to the Marketo program that helped in converting an anonymous lead to a known lead (typically using form submits). For a given lead, there will be only one Acquisition Program.
- Is a well-negotiated revenue model up and running? Will you be using multiple models?
This is critical if you’re going to use the Revenue Cycle Analytics (RCA) module in Marketo. At Informatica, we worked closely with Field Marketers, Sales Development Reps, and Sales to define our revenue model, capturing the journey of the anonymous lead through becoming known and all the way to the opportunity. (We’ll share more on this in a later blog.)
- Is there a Group Field Organizer for the Revenue Model set up?
This Group Field Organizer helps in aggregating various values of a given field to use in Revenue Cycle Analytics. For example, Lead Score may have a value ranging from zero to infinity. You may want to create buckets of 0-50; 51-99, and 100+ scores so you can analyze leads this way.
- Are all relevant custom fields imported into Revenue Cycle Analytics? Salesforce allows most of the leads and critical opportunity fields to sync into Marketo RCA for analysis in the form of dimensions and measures. Make sure you bring along the custom fields that will help you with your analysis.
Those are the notes for our Marketo Checklist.
Now for the analytics side of things.
The Adobe Analytics Data Governance Checklist
Setting up your analytics properly is absolutely essential for big data marketing. This checklist helps us keep on track:
Notes for the Adobe Analytics Checklist
- Create one Global Reporting Suite for all your relevant web properties.
This is a such a simple concept, but it’s overlooked in many Adobe Analytics implementations that we’ve seen.
All your pertinent web property traffic should consolidate into one reporting module in Adobe Analytics. If you don’t do this you will double count your visitors and you won’t be able to frame their end-to-end web journey.
For example: If your main site (informatica.com) traffic goes into one suite and then (www.blogs.informatica.com) goes into another, then a visitor who navigates to the blog portal from the main site will be counted as an individual visitor in both suites. Instead, the visitor should be counted as a single visitor and you should be able to see in Analytics that the person went to the blog site from the main site. This is critically important.
- Think through exhaustive business questions that you want to answer for all web properties.
Based on this, assign sProps, eVars, and Success events optimally.
In any analytical framework, you have to think about two things: dimensions and metrics. You can break down metrics by different dimensions to derive analytical insights. In Adobe lingo, eVars are dimensions, success events are metrics. sProps are similar to eVars but they’re not persistent, changing value as the visitor goes from page to page—they’re mainly used in pathing analysis. (BTW, I sincerely hope Adobe streamlines eVars and sProps and just come up with one kind of dimension.)
Setting up your sProps eVars and Success events should never happen in isolation from the business. It’s absolutely critical that your analytics reflect your business goals and the metrics that track them.
Get it right and your Adobe Analytics will make it easy to do ‘mass customization’ specific to your business your converion tracking (whether it’s revenue, orders or form fills). But keep in mind: as your website changes, you need to be mindful about the impact on your Adobe Analytics implementation. Adobe Analytics is a living, breathing, listening system that watches your visitors navigate around your website. If it’s not set up to align with your business, you’ll only get half the value.
- Use friendly page names.
One of the cool thing about Adobe Analytics is that you can provide a friendly name to each and every page that is intuitive and easy to read (www:us:en:products:products-portfolio) instead of reading URLs (www.informatica.com/products). URL names can get very complex very quickly.
So set up a solid page naming convention and stick to it.
- Create Product and Solution eVars and Success events.
This is important for any multi-product company.
As I mentioned in note #2, the most powerful feature of Adobe Analytics is mass customization. We use it to give us a product-centric view, similar to that of any e-commerce business.
As you go to any retailer website and navigate to, say, a vacuum cleaner, Adobe Analytics can capture the product name (brand/make/model), the category, and how many vacuum-cleaner-related pages that you saw. Since we’re a multi-product company with a diverse product portfolio, we set it up the same way for our website. Knowing what products and solutions our prospects and customers are interested in is gold dust.
- Tag all the Marketing Automation landing pages in Adobe Analytics.
It’s important to make your CMS and Marketo landing pages seamless from a reporting perspective. That’s especially important for Form Submits, the most critical Adobe Analytics event.
This can be a big undertaking since Maketo and Adobe Analytics are very different tools. When you visit informatica.com, it’s hosted on our CMS, Adobe Experience Manager. Our landing pages (with web forms) are hosted from Marketo, so when you navigate to our landing pages we need to tag them in Adobe Analytics. That allows us to track visitor behavior seamlessly across CMS and Marketo pages.
When we set this up, we had more than a thousand landing pages globally for paid media, gated content assets on the web site, etc., built in Marketo that need to be tagged in Adobe Analytics. Not a trivial task but well worth it.
- Deploy Marketo SOAP/REST and Demandbase APIs on the website.
Another critical element to map end-to-end customer journeys between Adobe Analytics and Marketo is to have a common ID shared by these two systems.
This common ID allows us to stitch Analytics and Marketo data together to solve some of our most important use cases in our big data marketing environment.
For us, this common ID is called “Marketo ID” and it’s unique for every visitor. We expose this ID in Adobe Analytics using the SOAP API so we can track our known visitors in Marketo. (SOAP is a software protocol that allows programs that run on disparate systems to communicate.)
We do the same for integrating our Demandbase data, which gives us the reverse IP lookup that lets us enhance our profiles with company and sector information.
- Set up Marketing Channels for web traffic attribution.
Marketing Channels is a relatively new, out-of-the-box feature in Adobe Analytics and it’s east to set up. It allow us to track Paid, SEO, Direct and Referrer traffic separately so we can measure engagement and conversions across these channels. Invaluable for budget optimization.
- Assign intelligent external campaign parameters to track paid traffic.
All external paid traffic URLs should be tagged with consistent parameters so that you can track the paid traffic that’s coming to your website.
For us, the paid traffic is Display, Remarketing, paid Social, and Search. The naming convention for parameters needs to be be smart enough so that we can slice and dice this traffic by regions, products, marketing themes and channel. This can involve a lot of manual hard work—for us, it meant tagging those 1000+ URLs that are in production.
- Remove internal (employees/partners) and robot traffic from the Reporting Suite.
It’s important to remove internal (from the Informatica IP address) and any partner-related traffic (agencies, etc) from your reporting. Otherwise, this traffic will significantly distort your analytics. Also, watch out for any robot traffic hitting your site on a monthly basis an remove that from the suite.
- Avoid unnecessary bells and whistles in Adobe Analytics—anything there’s no customer for.
This one is dear to my heart. I’ve often seen people put unnecessary functionality into their analytics that no one cares about. This isn’t just a waste of set-up time, it also becomes a burden as the website generate all sorts of maintenance issues.
In our case, we were coming from the opposite end: the old SiteCatalyst instance was plain vanilla and we weren’t coming close to leveraging all that it had to offer. So the temptation is to go ‘all-singing, all-dancing’ from the start.
It takes discipline to focus on business needs and to prioritize your features and functions into a sensible roadmap. Do it up-front and you spare yourself needless pain.
Make your own checklists
These two checklists capture a lot of the lessons we learned as we built our big data marketing engine. I hope they can help accelerate your own initiatives and keep them on track.
Just as importantly, I hope you’ll create your own checklists as your journey progresses. This ‘tribal knowledge’ is a valuable business asset—but only if it’s captured and shared.
The next post will be all about data hygiene and data management—a hugely important part of any big data marketing program. I hope you’ll join us for that.