“Errors using inadequate data are much less than those using no data at all.”
Data is abundant in enterprise SEO, internally and externally. Customer data, machine data, finance data, marketing data, etc. The challenge with BIG Data is that there’s so much data, that it’s a challenge to put it together and tell meaningful stories with it. It’s difficult to validate data integrity. Enterprise marketing reporting suites need to be set up and configured to the desired attribution modeling. For the SEO, it’s a constant battle of knowing when the data’s good, and when the numbers lie.
Questions Worth Asking
Where does the company store the most raw form of web data? Does SEO have access to extract or view any of this primary source data?
The company’s secondary (SEO’s primary) data source is commonly a digital analytics /reporting suite, and/or a business intelligence software such as IBM Coremetrics, Adobe SiteCatalyst, Tableau, etc. Which of these does the company use, and what level of access should the SEO team have?
Who creates weekly, monthly, quarterly, and/or ad-hoc reports and dashboards for the SEO team? How should you teach each other the fundamentals of (Organic and Paid) Search, and on the flip side, Data?
How are SEO clicks (sessions), orders, conversion rates, and sales attributed and calculated compared to all other digital marketing channels?
What is the digital marketing attribution model? For example, first touch, last touch, multi-touch. What are the pros/cons of your company’s approach?
Most enterprise reporting calculates Organic Search performance by taking total traffic referrals from Google, Bing, etc., and subtracting by all landing page URLs containing Paid Search tracking parameters.
SEO data points are like jigsaw puzzle pieces. Each data point such as web logs, search volumes, rankings, click-through rates, traffic, orders, sales, and customer segment data can be found from multiple data sources. How should the SEO identify the most important pieces, and compile them together in standard reports and dashboards, so that they can complete the puzzle and share complete stories with the business?
What KPIs are most important to the SEO team? To the Digital Marketing team? To Finance and Accounting? To Leadership? What metrics should be used to calculate the KPIs?
It’s common for SEOs to go through stretches where they spend more time generating reports, estimating project ROI, and building case studies than actually doing SEO. And for good reasons: SEOs report to handfuls of stakeholders, use dozens of tools and data sources, work on lots of unique projects, deal with constant changes in search engine and business landscapes, and for a “free” channel, the business invests lots of time and money in projects. Communication is almost as important as the work itself.
How should you standardize as many presentations, reports as possible?
How should you collect data from 3rd party sources such as Google Webmaster Tools, enterprise keyword tracking tools, etc. so that you can aggregate, manipulate, and analyze historical data in-house, and combine it with your internal reporting?
How can you crack Google’s 90-day rolling period limitation? Their 1000-row of search analytics data limitation? Answer: The API pull.
Example: Using API calls, you can pull keyword or landing page data from Webmaster Tools using a tool such as Neil Bosma’s SEO Tools for Excel, so that you’re not just restricted to a rolling 90-day period of data.
If using API calls and aggregating Webmaster Tools data, the BI team can guide you on where and how to warehouse this data alongside your primary internal data. How should you standardize this process?
Advanced example: once you begin pulling Webmaster Tools data, you can align it with the business fiscal weeks, so that you can take the top 5000 landing pages, their average CTR, ranking, impressions and clicks, and combine this with your reporting suite such as Coremetrics. You can now translate average rankings, search volumes(impressions) to meaningful sales data.
It’s very normal to get lured into and distracted by too many features inside of enterprise tools, such as Conductor. If you’re using it to focus on on-page optimization efforts, how should you incorporate the tool in an efficient workflow?
How should you justify the expensive cost of an enterprise tool, such as Conductor? Will you use it daily, or at least weekly? Who else will use it outside of SEO?
What scraper / crawler tools should you use to diagnose site issues?
Standard web logs reporting is difficult because files are too heavy, and the infrastructure teams aren’t always sure of what data to pull that’s SEO-specific. Are you familiar with ELK technology? Should you consider using it so that you can obtain server logs data of Googlebot activity on your site?
How should you marry on-site search data with external (whether it be from Google, or click stream data) search data to help customers find what they’re looking for?
What alternative ways can you segment SEO performance by? Examples:
- Website Merchandising: category pages vs product pages.
- Business Merchandising: department vs department.
- Customer Segments: new vs existing (active, inactive, etc.)
- Keyword type: brand vs non-brand
Low-hanging SEO opportunities are plentiful in enterprise SEO. There’s always a chance to optimize “striking distance” keywords (keywords in positions 4-20 on un-optimized pages), to redirect broken links, to claim un-linked mentions on the web, etc. How should you work on these?
Sophisticated Excel and Access usage can allow you to look at your data in many different ways, like a Rubik’s cube. Who on the SEO team should possess advanced Excel and Access capabilities?
Data’s important, and the more data available, the smarter your decisions. Data is a polarizing topic, you can see it in the NBA media: some swear by advanced statistics, others swear by anecdotal / gut-feeling analysis, the smarter ones realize there’s truth to stats and stories both. Think of all data in your company’s data libraries, as well as your external SEO data, as jigsaw pieces, and put the meaningful ones together to create high-resolution picture stories, and make smarter business decisions. As Google takes away more and more SEO data, you’ll have to adapt by using your own historical and internal data.