Does your list of SEO opportunities feel daunting? Not sure what activity is going to have the most impact? Why not let the data decide…
Christina and Wil will walk you through how to integrate your SEO and PPC data in Power BI to prioritize your SEO work for the highest impact.
Identify optimizations and quick wins
SEO x PPC scatterplot analysis
Where you rank for converting search terms
Opportunities to defend – both within striking distance & unrealized potential
Big data SEO Competitive Analysis
Who is ranking for search terms you convert on and why?
Identify opportunities for advertising on ranking domains
When: Wednesday, April 3rd at 12:30pm EST (9:30am PST)
Can’t make the time? RSVP anyways and we’ll send you a recording of the webinar afterwards!
Where: Wherever you’re at! Zoomin’ into your computer
Christina Blake, SEO Account Manager
Wil Reynolds, Founder & Director of Digital Strategy
Search Term = the actual term you typed into Google
Keyword = the phrase you put into PPC to match to several search terms
i.e. if my PPC “keyword” is nike air max, a search term could be nike air max, nike air maxx, nike air max sz 13, kids nike air max.
4 search terms for 1 keyword.
The way I like to explain it is…
A search term is what people are actually typing
A paid keyword is a way to tag or categorize your search terms.
The data we used from Seer was leveraging a last click conversion model.
HOW DO YOU DEAL WITH THE “EXPIRATION” OF DATA IF TEAMS AREN’T ABLE TO EXECUTE IMMEDIATELY? DOES DATA GENERATED TODAY HOLD TRUE IN A FEW WEEKS FROM NOW? OR ARE THERE REAL-TIME DATA AGGREGATOR OPTIONS?
We can pull the data daily if we want – For most clients that’s not as valuable. But there are instances (like Halloween) where the SERP changes completely in July vs September. When you are pulling 50K – 2MM keywords, you are going to rack up real costs pulling daily, and the juice isn’t usually worth the squeeze for most clients.
There’s a few options…
Using something like our data warehouse has real time data.
But for a lot of ad hoc analyses – what we’re doing at Seer is creating templates. This way our team can hook up their client data and spend their time on analysis.
So let’s say I did an analysis for you 6 months ago and now you’re looking to implement on that. It would be relatively simple for us to hook it up to your most recent data into the same exact dashboard.
Christina has been working a lot on the templatization and I’ve been focused on the data warehouse – so I can dig in a little there.
Anyone on this call – you’re going to go through a progression.
I use to login into AdWords, download .csvs, login to my ranking tool, upload all the search terms, and then join them together. It was very manual. That’s where you have to start.
Then we had a bunch of our team members came along and said – hey, we’re learning how to put parameters around and templatize some of these things so you can just download the data and update the visuals.
Then our future state is we hit “refresh” using BigQuery and that’s it. Every team member is hooked into access with just their own clients and it’s a single button refresh to get the updated visuals.
Then they are right into having a robust conversation with their clients and that’s really what the future of where we’re hopefully going to be able to get to in the next 6 to 12 months.
HOW DO YOU NORMALIZE RANKINGS FROM SEMRUSH TO STAT, FOR INSTANCE, ON AN ONGOING BASIS WHEN RANKINGS CHANGE EVERY MONTH?
There’s two use cases where we use STAT at Seer. One is for ongoing ranking tracking, day to day we want to keep an eye.
When we do major analyses, we often upload a sandbox. Download data from multiple sources (like search terms from AHRefs, SEMRush, Moz, and Adwords) and put them into STAT, then run long enough that we can get that data back to do that analysis. So we do run it through STAT to normalize.
Big issue — if you don’t ever bid on upper funnel keywords then you don’t have the data. We try to encourage trying to spend for a few weeks on upper funnel so we can get that data and prioritize. Without that, we end up using ahrefs or semrush to get upper funnel keywords. Not ideal, but we understand that sometimes that is the reality.
CONNECTING KEYWORD AND SEARCH TERM DATA – DO YOU SPECIFICALLY DO IT JUST ON THE KEYWORD TO SEARCH TERM LEVEL? FOR EXAMPLE, IF YOU ARE BIDDING ON “GOOGLE ANALYTICS PRICING” AND PULL GSC DATA FOR THAT SAME QUERY REGARDLESS OF WHAT URL IT RANKS FOR?
Keyword from an SEO tool connected to search terms from you PPC platform. That is one of our joins.
Yes we pull regardless of what URL ranks b/c this is how we find if an organic URL is performing better than a PPC landing page and vice versa.
IS THERE ANYTHING YOUR TEAM CAN DO ON THE ORGANIC RANKING SIDE OF DATA IF YOU DON’T HAVE ACCESS TO A PAID TOOL LIKE STAT OR MOZ?
On the organic side of things, you usually need some tool to help you get what’s ranking for you and your competitors. You can scrape google and get position but usually you get blocked pretty quickly.
SEMRush, Ahrefs, Moz, STAT, Screaming Frog, Webshrinker, and Clearbit Reveal were the SEO specific tools.
GOOGLE RECENTLY ADVISED THAT AD POSITIONING WILL NOT BE A METRIC THEY REPORT ANYMORE. WHAT ARE YOUR THOUGHTS ON HOW THIS WILL AFFECT SEARCH?
Erin Simmons: Recommend checking out Grant Endrington’s post: Google Ad’s Average Position Metric is Going Away
We’re not really using position right now. But what has been fun is to look at your click through rate and some of your paid metrics based on where a competitor ranks that’s consistently above you. So when Amazon ranks organically in 1 or 2, how that affects your paid data. Then when Amazon drops out of 1 or 2, when do you see an uptick in some of your paid data. When we can monitor that – we know when Amazon drops into position x, we know it’s time for us to bid up. So the way we’ve looked at positioning so far is more using organic positioning to understand how a big competitor above you influences your paid data.
CAN YOU WALK THROUGH WHAT THE “OK, GOOGLE” MISFIRE MEANS THAT YOU USED IN YOUR VOICE SEARCH ANALYSIS?
When I say “Ok, Google” into my phone or device, I get an answer. Sometimes folks double speak and say “OK, Google” twice. In these instances, “OK, Google” is captured in the beginning of a search query. Then I can pull all the impressions out of our data warehouse across all our clients to identify where these instances happen as a search impression. Then I get data on impressions, clicks, and what not.
The hole in that approach is I do not know what the propensity for a misfire is. If it’s 1 in 10, 1 in 100, etc – we have to give a good, better, best assumption there since that data isn’t available.
They’re different joins between data. I didn’t know anything about joins until I started on this journey. 18 – 24 months ago I did not know what Power BI was, what joins where, the difference between a keyword and a search term — all this stuff is learnable if you take the time.
So, yes, I decided to create a shirt to showcase my inner nerdiness and it’s a bunch of joins.
WE’VE HELPED A CLIENT WITH YOUR METHODOLOGY AND FOUND THEY SPENT $300K ON KEYWORDS THAT GENERATED ZERO CONVERSION FROM ADWORDS. THEY’RE ASKING IF THEY SHOULD REINVEST ALL THAT INTO SEO. WHAT DO YOU THINK?
I would hate to move all that money into SEO. Using that methodology, you should’ve also found search terms that were doing really well. That’s an opportunity to move that money into high performance paid.
I do think you could then also suggest specific pages and content to create for SEO with the appropriate budget costs for those.
When you actually dig into data day in and day out – you see how many search terms only converted 1x all year. If the conversion is worth that spend – maybe that’s OK. The bigger thing is when are you targeting search terms that don’t make sense for your business.
So when I see 21 Savage Bank Account terms — yes, get rid of those. But I don’t want to get rid of terms that have zero conversions because the long tail has gotten so long, you’d probably do a net negative to the account there.
You could also check out our new Seer Solves It that talks about detailed ways to find wasted spend: How to Find Wasted PPC Spend Using Data Visualization