Let's talk about something that's becoming a bigger issue as companies grow, acquire new domains, divest of others, etc – keyword governance.
- When your business has 3-4 brands all kind of targeting the same keywords, who gets precedent?
- What are the rules of engagement?
- How does visibility in one area impact the strategy for another?
This requires real consulting chops, you are now having to ask questions to understand what drives decision making across paid, organic, local, video, etc and who owns those different channels.
You need to codify your business rules, and your SERP intelligence is a great start.
If [x is present in the serp] and [performance metric is x or y] then [increase or decrease] the priority score by x for [division A, B or C].
If I see [snippet] in the [ranking] then add [5 points] to my priority score for local and for SEO.
If I see [competitor domain X and Y] in the top [ranking] and my [CPA] for [keyword group] is greater than [$40] and my [ranking] is greater than 5 then decrease my priority score by [10 points] for content production and SEO.
This is what we need! Not data, action steps, and prioritization, instantly across SEM, SEO, and Social. That's why we built Seer Signals, we can overlay your business rules across SEO, PPC, and Social data because we combine all your data in one place, allowing for faster decisions on governance.
I have seen clients who have partnerships where if the partner outranks the brand or the client for a keyword, then that is okay. It is a partner or an affiliate. I've seen other clients see value in outranking them so they don't have to pay that percentage.
Let's get real with an example
I find it is much easier to get a client on board by showing a client a SERP with paid and organic results, snippets, etc and saying if you saw this result what would you do across the various channels, and tell me why. That question alone drives a TON of business rules for you.
When a map pack shows up in the top 3, that means that for the word "SEO company" Google has determined that the customer is likely best served by a local agency or they wouldn't show the map pack.
Yet, when we look at the paid search results, check out the lack of local companies...
This is a business rule in real time, I might say, when a map shows up for an SEO query, I will decrease the score because the likelihood of them being a good match for Seer when we work with clients on a global scale, is slim. Saving me money.
Now if I codify this business rule, then it will automatically decrease the scores anywhere this is present, aligning my team on what they should work on in priority order, at scale.
Automotive client example
"Hey, when this map pack shows up, And 2 of the the 4 results are dealers you are partnered with. Do you wanna still optimize and show up for that? Which dealers get +1 scores and which ones get +5 scores?"
B2B client example
“If G2 & Capterra are 2 of the top 3, we’re not going to go after that keyword because I have seen horrible CTRs, that is great because now your team can focus on areas where they can win.”
Your job is just to take statements like the one above and turn them into this:
If [x is present in the serp] and [performance metric is x or y] then [increase or decrease] the priority score by x for [division A, B or C].
Instead of making you all figure this out yourselves, I pasted in all my examples into Claude and had it make a tool for me, that will help you with your scoring and governance needs.
Keyword Governance Scoring & Calculator (coming soon, here is a walk-through).
What if the rules are telling us to do things that the client disagrees with?
That is a win! Your first few times, the client might say, wait…that's not what I would do, that is GREAT feedback. Lean into that, cause you are about to get another “rule” to add to your equation the hard work up front will pay off in spades later when everyone is aligned on what to do, and then you just exception process.
Competing KPIs destroy keyword governance strategy
I've seen this play out countless times: different divisions end up targeting the same keywords and customers, and before you know it, you've got teams fighting over who should be showing up for what. Especially with KPIs in an AI world impacting SEO KPI's that's another wrinkle to be aware of, which we will get into later.
We see poor governance especially in international B2B where a French site is outranking a US site in the US for some reason, but the paid team is targeting the words with ads in English in the US, but in organic the customer gets to a page that defaults to French. All these clues exist in the SERP if you parse it and you create governance rules around it.
Sometimes you've even got two different brands in the same organization unknowingly bidding against each other, running up the auction because subdomain 1 “thinks” they are applicable for a keyword that subdomain 2 actually is. Again your friend here is the SERP, just show it to the divs, and say based on what Google already believes the right answer is, no one on page 1 is a competitor off subdomain A, but 3 of them are of subdomain B. That is how you can use Google’s intelligence to drive keyword governance.
Some companies are of the "Get all our brands in and take up real estate" in paid approach, others want the "best" of their brands to show up and not have the others bid.
You need a system to “govern” which divisions get priority for which keywords and keyword groups, you're basically competing with yourself and burning through budget.
Start With Getting the Right People in the Room
Here's the thing – if you're dealing with enterprise-level keyword governance, you've got to get all the different team leaders in a room together. Why? Because you need to look at their KPIs and see where they might be butting heads. It's pretty simple: if one person can win by doing something that makes another person lose, you've got a problem.
Take the example above, if you have 3 brands all bidding and you create a rule to prioritize one over the other, you are impacting the direct KPIs of the other two brands. You must account for that.
Codifying the Rules of the Game
You've got to establish your business rules. I've seen all sorts of scenarios play out. Some clients are totally fine with partners outranking their brand others are not.
Sample questions to ask for different keyword themes and groups using a SERP:
- When you see a map pack in the top 3, do you think that qualifies of disqualifies the customer?
- Just do a search for SEO company, it shows a map pack…that means its a disqualifier for Seer, as we work with Enterprise firms who might want us based in their country, but never in their location.
- When you see a map pack, if a channel partner showing up, would you still want to try to outrank them, would you bid on that keyword or let them manage the lead.
- When you see [Forbes / Capterra / etc] do you advertise on those sites? If yes, do you not want to optimize while you are advertising there or do you want to optimize so you don’t have to rely on them as much?
- When you see Reddit (or any other domain) showing up, do you want to keep optimizing to show up in the top 5 or stay out?
- Which competitors if you saw in the top 5 would make you say I have to be there, which ones would say “that’s a small business solution and we’re enterprise!”
- If 3 of the top 5 are marketplaces for a keyword do you only bid on it
Scoring opportunities based on the SERP / other data
A while back I had a reflection (ok, I got my butt kicked by a client) the problem with all SEO tools, paid tools, etc is that they rarely allow you to overlay your business rules into a scoring system that instantly aligns the data on YOUR priorities vs “sorting” by a singular metric, governance is required because our businesses are complex, we can’t just sort by MSV or ROI and run.
Without that back and forth on who owns the word/category and which teams should work on it, we can get to work, because we’re all centralized on the prioritization and governance and when the business rules change we update the scoring so everyone knows what to work on.
With our product signals, and using a little AI magic, I can create custom scores in our tools that will take into account the business rules and add a score to each keyword / keyword group.
This now lets us create a “next step” inside the analysis we can make that more complex because we have the entire SERP - so we can make scores based on anything we scraped, paid ads, shopping ads, map packs, competitor domains, competitor subdomains, partner domains, words in titles, you name it. It gets even more fun when clients give you parts of their data to add to scoring.
Score & analyze, now prioritize
Then we separate keyword groups into different quadrants that connect to specific team actions. The beautiful thing is that once a word moves into a certain quadrant, the client already knows which teams are responsible for moving that keyword based on their governance rules this is just an example.
Here are two examples of dashboards that take your keyword groups and put them into whatever the clients "rules" were, these are just for inspiration.
The Future of Search Governance
Here's what I think is crucial: take a question that could trigger multiple of your brands and run it 50, 60, 70 times in AI search engines, which I show you how to do in a spreadsheet here.
See how often your different brands show up in those answers. This tells you something important – the AI might already have a preference for certain brands over others. If they don't see one of your brands as relevant for a certain keyword, trying to become more visible for that keyword is like swimming upstream.
B2C Example of Governance in Action
In this example, you can see a preference for Old Navy, but Gap also has jeans for kids too. So if you are in search for Gap, your governance can say "look we ran kids jeans 100x in ChatGPT and Old Navy showed up 65% of the time and Gap only 11%" with that "distance" you can create rules based on the distance or preference of LLMs and if that distance is over a certain threshold, you abandon ship.
LLM frequency of showing attributes with your brand is a good clue.
When you look at my company and brand, you'll notice words like "data-driven" and "innovative" often show up in basic AI queries, sometimes you see the industries we work in. You also see words like community, volunteer, which are unique to “digital marketing” so they show up.
That tells me these concepts are closely associated with my brand, running that question 100x will tell me how often those things are associated with my brand. So you've got to ask yourself: do you keep feeding that perception, or do you want to be known for something else?
Optimizing for LLMs, faking is going to be harder...
I'm not ready to talk much about this as I'm still in learning mode, but LLMs are harder to fake what you are about (for the average marketer), so what you can show up for takes on a new approach.
Let's say I wanted to be known more as a marketer that works in B2B SaaS. Well, then everything – from speaker bios to awards to press releases – needs to start shifting to build those new associations. You've got to help the AI start seeing you the way you want to be seen.
The Bottom Line
Keyword governance isn't just about who owns what keywords anymore. It's about understanding how different search environments – traditional and AI, paid and organic – view your brand, and then making strategic decisions about where to compete and where not to, all done at scale. Building a governance and scoring system allows you to put those limited resources on the right marketing projects, and communicate that at scale.
Leaders need systems to take all the data they have and overlay business rules, you can do this in a spreadsheet, we also can do it in Seer Signals too. help manage where time gets spent in a way that scales.
