GEO expectations around "time to success" are directly impacted by how far the topics you want to rank well for are from what you are known for TODAY. Been noodling on something... in an SEO world if your brand had enough authority, if your brand went into a new area, that high authority would let you rank. i.e. If Hubspot went into small biz payments, they could probably get top 10 visibility within months on Google. But in an AI world, it could take you years to become a known entity for that topic. This phenomenon is new to people who have "bought SEO services" in the past, which could lead to really tough times trying to manage expectations in a GEO world.

Jalen Hurts wasn’t perfect yesterday in Tennessee, but when the moment demanded it, he was DANG GOOD. That final drive by the @Eagles was a thing of beauty, and their Super Bowl MVP quarterback delivered when it mattered most. I couldn’t wait to break down every throw, play by play. That’s “What Kept Me Up Last Night.” And there was no way I was doing anything else before getting to this one today. @ThePivotSports Show @Thepivot #ThePivotSports# #ThePivot# #NFL# #JalenHurts# #FlyEaglesFly#

Given that Meta Muse is a thing now and is the top app all of a sudden, I had some thoughts. Feel free to shoot holes add on top, etc. But trust me clients will be asking us, so lets banter around here and help each other. First: This would be the kind of news that I would recomend clients bookmark the day of launch and see if anything started changing in their call data, reservation data, requests for changing subscriptions and bills. I think anyone who has bills people want to lower: Energy, Insurance, Mobile phones, Reservations, Bookings, etc Looks like it is good at scanning your subscriptions and cancelling too If a client has a product that is likely to be sold aftermarket in facebook marketplace, I wonder about the usefulness. It looks like Muse has some leanings to retail sites for shopping vs review sites in the two videos I saw, which would fit the meta playbook, give the retailers free traffic for a while then at some point...ads.  Whereas publishers are a LOT less likely to buy ads. If you look at some of its default states (again from the 2 videos I watched) to get people started, you see options like "shop for a new credit card (banking)" and "plan a home improvement project (manufacturers + retailers) If I was in ecommerce I'd really want to see if there is some way for me to start tracking Muse attempts to buy products.  Is there any fingerprint (Amazon already figured out how to block them, wonder if cloudflare will be next) but how might I be able to add a honeypot or something, like "If you are the muse bot" click here on a hidden field :) What are you thinkin'?

"What decisions did you make for me without asking?" ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ This prompt is a BANGER if you are coding at all and new to it like I am. Given to me 3 weeks ago over lunch with a buddy.

The thing that is so new with GEO is how in an SEO world you did a brand search, as long as you had enough branded assets, you owned most of page 1. This definitely requires different thinking about brand search.

Remember all these changes in AI models launching, new tech launching is what you get when you gave some level of competition. Shoving ads and overcomplicating the layout (Google search), not solving spam emails (gmail), and spam comments (linkedin) is what monopolies get you. I'll take head spinning speed and change any day.

Thought: Will fanouts with brand names that are embedded in the models thinking, make it harder for upstart brands to win? I think that is one of the downsides. But then again, upstart brands doing the right things fast and on a long enough time horizon will be able to compete. Start in the niches is what I am thinking, you would struggle to compete on dog food, but "dog food for german shepards" (making that up) might be more winnable if you went all in. Then you could expand from there. I am also thinking of a new KPI for moonshot keywords, it isn't great so shoot holes, help me make it better. Could you look at a topic cluster vectorized, and see where your client name is relative to the keywords/themes they want to be known for. Could you report on "getting closer" over time, even if you braodly aren't visible yet? Could you run a prompt 100x to see if you are mentioned at least once, then twice, etc etc? Clients have to show something on a regular basis, and saying "you are not visible" this one time I ran this one prompt could miss out on the fact that running that prompt 5 times maybe you are there once. Thoughts? cc @khemaridh

Read this somewhere, love it. Five years from now, we’re going to have a new competitor that is faster, more innovative, and more successful at a lower cost, and they’re going to put us out of business. And the only way to prevent that is to become that company.

If you have high AI visibility and the trade pubs, influential people,and brands that influence your industry don't mention you, ever, you in danger girl. New KPIs for GEO. Take the trade pubs (search engine journal, search engine land, emarketer) and influential brands (ahrefs,semrush) showing up in fan outs, and use analytics + @alertmouse to see do those brands ever mention you, your trademarks or your brands. Then take the personal brands (Lily Ray, Mike king, rand fishkin, etc) showing up in fan outs. Use @officialpbuster to harvest their x posts, LinkedIn etc. do they ever mention your brand or your site? You can also use alertmouse to catch these too. Consider the tactics that got you all that AI visibility won't survive scrutiny if the models continue to use brands in fan outs.

one of the biggest unlocks for me running a remote company was introducing the weekly success list (™️ pending) here’s the context: > running a startup is super chaotic, there are so many highs and lows in a single week > by the time friday afternoon rolled around I couldn’t tell if this past week was good or not so I began compiling a list of things on sunday, that if completed during the upcoming week, I would objectively consider it a success during all hands on monday, I have the owner of each initiative speak to it for a few minutes in front of the entire company so everyone understands it and is aligned then during all hands on friday, we review the weekly success list again with a binary: “yes it got done” or “no it didn’t” I’m aware this isn’t rocket science, but it’s been so remarkably effective for our team (and for my own mental sanity) ultimately, the weekly success lists are a forcing function for the following: > prioritization: a clear list of what’s most important at any given time > alignment: everyone at the company understands this prioritization > accountability: each item has a clear lead who is accountable and must speak to it > feedback: at the end of each week we have an objective scorecard of what was done > progress: each week you see the continual progress we’re making as a team what do you do to keep your team aligned?

Make decisions faster, be less overwhelmed. I think people who are constantly overwhelmed are in that position sometimes because they leave small decisions undecided / unacted on. For me: I assess risk fast, is this a decision I can easily pull back from? If yes = close the loop ASAP, make the decision, document your hypothesis, pull the baseline data, set a reminder to check in and see if you were right or wrong. Run your next test. If no = spend more time or delegate to people better at making the call. Too many people "file away" what they could answer today. Read this, go close some loops.

Been working on my new approach to building presentations. Inputs: Readwise: highlights and notes Slack: All my crazy ideas and videos internally Linkedin: All my posts, comments + performance Code: Compare my new reads and thoughts to existing slides and notes Find what I am thinking about, saying, highlighting that could challenge my existing slides or make them better. Output is what you see in the video - I get to decide what makes sense to add and not, in a loop that trains the system why I think the way I think about what makes it into my decks.

When an ad is shown in ChatGPT, they seem to be "nudging" solutions less frequently than answers without ads. Nudges look like: asking for more detail, building a solution, etc The data showed that users got between 19% & 35% less "nudges" when an ad was present.

Maybe "how many pages of content are you going to build" is an outdated question. This research for a few clients shows that maybe we should all be doing an evaluation of what content needs updates vs what needs net new. This research makes me think we might have to change how we decide what workflows we use to determine net new vs updating older content. If you see that the old pages at one point were getting citations maybe update & consolidate is the right route. This does make me wonder if you build out content and you want to update it regularly, kind of like some of our studies that we rehash every six months, we have to figure out how to balance launching net new pages versus updating the old anybody else out there talking about this or got ideas? Also I'm not a believer in AI bots when they trigger web search following links in your posts, but I'd love to see research on this. It feels to me like they run a web search and that is it. Which also makes me think if you understand the fan out, could you include / refine old posts to make them more in tune with the fan out patterns. So if you find a fan out pattern do you update old posts to be more discoverable?