Every so often, SEOs should revisit their keyword strategy. This can be a daunting process for many of you, depending upon how large your websites are and how many non-branded (and even branded) keywords there are driving traffic and producing conversions. I am putting together a two-part series to share some of our process and get feedback from other Conductor customers and blog readers on your keyword strategy process.
In this first post, I will cover tips for getting buy-in from management to implement a keyword strategy re-evaluation and how to organize all the keyword data that you will need.
Before You Step: Get Buy-In
Since assessing a website’s keyword strategy can be a time consuming process, it’s important from the beginning that (if applicable to your organization) you get buy-in and support from your manager and/or team by explaining the business case for this activity. Though every organization and business case is different, here are some general points I recommend highlighting:
- Generate more conversions: This activity will identify keyword opportunities that can drive conversions in organic and/or paid search channels.
- Save money: This activity can help identify opportunities where it may be possible to test reducing paid search spend on keywords with an average first or second position whose organic SERP position are also first or second.
- Content opportunities: Identify high-volume keywords that are currently driving clicks and conversions in paid search but do not have visibility in the organic search channels where content can be developed.
If you need help evangelizing SEO within your organization, then you need to get in touch with Conductor’s Stephan Bajaio, as he is an expert in the art of SEO evangelism.
Know and Organize Your Data
The first step in structuring a new keyword strategy is to identify all of your keyword data sources and define a look-back time period. Some of the data sources we look at are Organic, Paid, SERP/Visibility, and Keyword Discovery. The look-back time period is subject to how frequently you evaluate your keyword strategy, but my rule of thumb is to use data from the last three months. Of course this depends on a number of things, including how much data you have. Without using a robust data set, your keyword strategy will suffer so always err on more data than less.
The first step in structuring a new keyword strategy is to identify all of your keyword data sources and define a look-back time period. Some of the data sources we look at are Organic, Paid, SERP/Visibility, and Keyword Discovery.
Once you’ve established your look-back period, it’s time to export the actual data. Before we pull any data, though, we need a place to store this data. We’ve built out a ‘Keyword Scorecard’ in Excel which is used to house data from each of the sources, independently of one another, in separate tabs. Eventually, we will use this ‘Keyword Scorecard’ to categorize and prioritize our keyword strategy. Disclaimer: The metrics identified below are only examples. I recommend pulling in as much relevant data available into these tabs from the onset to create a holistic view.
We work with our web analyst to pull organic search keyword data from our Analytics provider into the Analytics-Organic tab. Prior to running the export, we make sure to include all of the metrics we have deemed to be crucial (as derived by our previously established KPIs.) Again, these metrics will vary depending upon what your business goals are. Next step is to pull in the paid search keyword data.
There are two primary aspects to note with paid search data. First, we work closely with the paid team to include campaigns and keywords which were active throughout the majority (at least 75%) of our look-back time period to control for accuracy. Second, we coordinate with the team to include the metrics which are crucial to their channel’s success. Lastly, we make the conscious decision to only use exact and phrase match data. Using broad search (in my opinion) would only skew our assessment of that keyword’s effectiveness from an organic search perspective, so we ignore it.
At this point in the process, we have three tabs within our Keyword Scorecard: Analytics-Organic, Paid Exact and Paid Phrase. The next step is to bring in data that shows what our current organic visibility is for the keywords we are tracking. At this stage we’ll do an export of Searchlight (into a fourth tab, labelled Searchlight) to pull in our current Google SERP, that keyword’s ranking URL and Google Local (exact) Monthly Search Volume data (so we can have an idea as to how much demand the keyword generates.)
Next we create a fifth tab, called Keyword Discovery. This tab contains keywords we have identified through our own research that we want to further assess and potentially target. We primarily discover these keywords through Searchlight’s Visibility / Traffic Explorer. However, if you aren’t using a solution like Searchlight you could use (or supplement your discovery efforts with) the Google AdWords Keyword Tool. If keyword discovery is a crucial aspect of your team’s responsibility, I recommend investing in an AdWords API Key*, as it will save you a tremendous amount of time and stress, or a keyword research product. For purposes of the ‘Keyword Scorecard,’ we track and record the same metrics for these keywords as those we captured in the Searchlight tab. *Richard Baxter of SEOGadget, put together a great post about using the AdWords API for SEO.
Bringing It All Together
At this point we have a wealth of data within this Excel workbook, now it’s time to roll that data up into one tab – the ‘Dashboard’ tab. The Dashboard contains a snapshot of how each keyword performed across sources, during the look-back period. This is the tab we use to categorize and prioritize the keywords based upon their performance and opportunity that exists for them within organic search. First, we visit each tab, copy all keywords and paste them into the Dashboard. Again, this list could be quite substantial depending upon your site’s popularity – don’t let that frighten you, as we can quickly parse this list down. It’s better to have all of the data in one place so that we can confirm no opportunities are missed. Next, we’ll de-duplicate the list to ensure that a list of only unique keywords remains. At this point, we identify the metrics we want to import into the Dashboard from each of the data source tabs and define the columns accordingly. I recommend focusing on the most important metrics (always look to your KPIs) first as others can always be added later. Finally, using several nested ‘Lookup’ formulas, we will pull in corresponding data from each column from each of the respected tabs.
The first draft of your Keyword Scorecard is now complete and ready for your organic team (or even the paid search team) to organize the data set. I’ll be publishing a follow-up post about how you can categorize and prioritize your Keyword Scorecard within the next week.
Until then, I welcome your feedback and thoughts in the comments below – Are there data sources you use (or would use) which aren’t listed here? Is there another format or methodology you use (or would use) to analyze this data? What is your process for keyword discovery?