Skip to Content

What I Learned from Teaching Python to Marketers and SEOs

54 min read

RankSense Founder and CEO Hamlet Batista sits down with Jeff Coyle, MarketMuse Co-founder and Chief Strategy Officer to talk about Python and SEO. Their discussion is interwoven with interesting insight from Hamlet on his experience of building a personal brand with the help of content and community.

AMA With Hamlet Batista

Below is an excerpt from an AMA (ask me anything) with RankSense Founder and CEO, Hamlet Batista. This event was held on our newly launched Slack community, the Content Strategy Collective. Upcoming AMAs include Jill Nicholson, Senior Director of Customer Education at Chartbeat, Lisa Deignan, Global SEO Analyst, Lionbridge, Mike Leonhard, Founder of Composely, and more.

Join the Content Strategy Collective here.

How did you get started with Python? Why did you turn to that as a practice for SEO/marketing?

I have a developer background, so I learned Python many years ago. 2004 to be exact. We used it extensively over the years internally to automate tasks here and there. It is very easy to use to teach team members that are not technical.

Can you show us some examples and uses cases related to you and RankSense generating titles and subtitles using NLG?

I would love to. I’m really excited about this capability and can’t wait to get this out! We are hoping to release this in our product next week. 

Do you also focus on long-form text generation or do you stay in the 50 – 250 words range?

I try to stay up to date on the progress of long-form text generation as well. The main challenge with NLG currently is keeping the text factual. I plan to write an article covering that using question/answering. I research/try +5x what I share primarily because most stuff is not practical or too complex to share.

What resources would you recommend for people moving into the world of machine learning and applying those concepts to the SEO space?

I’d recommend this Coursera course and if you are a bit technical, you can try this deep learning course. I’ve also written many articles in SEJ that are more SEO specific, including:

Any specific Python libraries you use for SEO that we should familiarize ourselves with?

I don’t use any SEO specific libraries unless you consider this nice wrapper for Google Search Console. I submitted a pull request to enable its use in Colab. Now, you reminded me to address his comments to get it merged.

How do you prioritize product efforts at RankSense and what’s the decision-making process?

We work on weekly sprints, which gives us a lot of flexibility. I use the feedback from my articles and presentations to re-prioritize our roadmap. For example, the last SEJ article was a big home run, so we dropped everything else and pushing it out ASAP!

Do you manage a longer-term (6-12 month) roadmap on top of this that paints your broader vision?

My product vision expands 2-3 years, including things where the latest tech is not there yet. We have a lot of computer vision work that is too slow, expensive or impractical to add to the roadmap. So, I try to be patient, waiting for the SOTA (State of the Art) to improve enough to include in our immediate plans.

How much automation do you think ML would take over in the content creation pipeline in the near future?

I’d say definitely over 50%. Check out the amazing work by MarketMuse already and the NLP industry is progressing exponentially. We live in an amazing time!

Do you have any recommendations (courses or techniques) for brushing up your public speaking skills?

Absolutely, I took a public speaking course in Coursera a couple of years ago. It depends if you prefer in-person too. But, I definitely recommend professional training it makes a big difference. This one is really good. It teaches a specific technique that worked well for me.

Since Google started using BERT in its search engine, how much weight do you think it has on ranking by now?

I wouldn’t think about it as a weight in their rankings. It is more of a way to interpret queries better. Before they would not match queries and documents as precisely and now they are better at that. I’d say it would lead to completely different rankings. Remember that rankings are tied to queries and probably semantics.

What are some of the possibilities/challenges you envision in the future when we talk about deploying the state-of-the-art NLP models like T5, BART, etc. to solve tasks in SEO?

I am excited as I see more menial work getting done and faster. There are many SEO tasks that don’t get done because they are boring and time-consuming. I’m also seeing this as raising the bar across the board in the industry, which is something I’d love to see more respect and appreciation. I also think SEOs will enjoy doing more higher level, strategic work.

What resources better explain the role played by structured data into the establishment of entities and the knowledge graph?

The structured data you add to a website specifies the entities and relationships. A knowledge graph binds all site data together into a bigger data structure. Here is a simple tutorial that you can follow to build a knowledge graph in Python from scratch.

There are relational/tabular databases more similar to spreadsheets and then there are hierarchal/graph databases. The main difference is they make it easy to connect relevant entities. A JSON-LD is better represented using a graph. Building a graph means adding and connecting relevant entities, for example, people, places, etc. My talk at SMX Advanced will focus on a specific use case: an internal search engine.

Summary of Translate Model for Knowledge Graph Embedding should be a good starting point, but the math might be a bit intimidating. Take advantage of these popular papers with the code to try the ideas. Knowledge Graph Embedding is a complex topic overall, and one of the things I will do in my talk is to simplify and make practical. Google recently released an article on generating structured data with Javascript. The developer resources should be a must-have for tech SEOs. Try this codelab too.

I also wrote an intro to JS that includes a structured data example.

Do you have any simple scripts that are must-haves (ie, identifying 404s, etc)?

Absolutely. I shared many in How to Use Python to Analyze SEO Data: A Reference Guide.

Do you know of any good resources for analyzing file logs?

There are many resources on this topic: 

Is there a specific methodology to follow for specific NLP tasks? Any recommended resources on that topic?

I’d recommend trying the Hugging Face pipelines first, so you can see which ones are more practical for your use cases. This notebook should be a good starting point. Then you can investigate how to create the pipeline steps directly as you will have more control.

Resources

Marketing AI Profile: Hamlet Batista.

Paul Shapiro

How to Start a Movement

How to Use Python to Analyze SEO Data: A Reference Guide

Ruth Everett

DeepCrawl

Ludwig

Automating Intent Classification

Transformers

Pandas

JR Oakes from Locomotive

Using the Apriori algorithm to visualize change in search console rankings

Vincent Terrasi from OnCrawl

How to Use BERT to Generate Meta Descriptions at Scale

Andrea Volpini

How to Write Meta Descriptions Using BERT

Pythia

How to Generate Text from Images with Python

Papers with Code

Webinar Transcript

[00:00:00] Jeff Coyle: Hello, and welcome to another MarketMuse content strategy webinar. I’m your host and the co-founder and chief product officer for MarketMuse Jeff Coyle. Today’s discussion is entitled “What I Learned Teaching Python to Marketers and SEOs. And it’s going to be discussion about building a personal brand with content, with community management, with a focus on engagement, but with the twist that the brand is something very technical.

That’s bleeding out of current communities of experts. Before I intro our special guest today, a few housekeeping notes, ask us questions in the Goto webinar sidebar on the right side. If you haven’t, you’re not familiar with Goto webinar ask them as much as you want about almost anything.

Any, we either get to them in line, or if they’re relevant to our discussion or we’ll get to as many as we can at the end. If there’s any overflow, we’ll get to them directly via email personal follow-ups. And we’d like to get that. The recording and notes are available after the webinar. So there’ll be sent to you via email.

Look for those. Please download the rest of our recordings. We have an amazing collection of content strategy webinars from startup founders like Brennan white from Cortex, sales enablement experts like Pam Didner. Just did an amazing presentation at B2B exchange on AI and sales enablement. SEO superstars like Andy Crestodina, all around brilliant people like Kevin Indig.

If you download multiple recordings this week, shoot me an email on top of that, and I’ll set you up with a free content advisory team session. Cool. All right. Now I’ll intro today’s special guest. He’s an expert on technical SEO has some really unique abilities between with regards to assessing and resolving site architecture issues.

All kinds of fun stuff. There’ll be a lot of migrations and all that good stuff. He’s the CEO and founder of Rank Sense. And he’s frankly been on fire in the last year or two in our circles. Hamlet Batista 

[00:01:46] Hamlet Batista: Thank you for having me, Jeff. Really happy to be here. 

[00:01:49] Jeff Coyle: For sure, I am. Tell the audience a little bit about yourself, your business.

[00:01:55] Hamlet Batista: Yeah, absolutely. Yeah. So I’m the CEO and founder of Rank Sense. And we specialize on technical SEO implementation, Agile SEO, which is doing SEO incrementally in the cloud and specifically in CloudFlare for a lot of our clients are agencies and brands with teams that have a hard time getting their stuff done.

And part of the work that we do in relationship to AI is time, Python, which is an easy way to get involved in automating, unique use cases, including generating meta descriptions, titles image alt tags in a way that, leverage the latest advances in AI, machine learning and AI, and then deploying those changes.

Once you have them in an experimental fashion in CloudFlare, and I’ve been building a movement around this and I’m really excited to share it, what I’ve learned about it. 

[00:02:43] Jeff Coyle: Excellent. And you get the tail end of today’s discussion. We’re going to get into the details of those projects and I’m really pumped to have a bunch of questions about.

But I wanted to start the conversation, just to talk about the journey for you. And I thought I’ve been lucky enough. We’re a member of a mastermind or a Slack community on technical SEO. So I’ve been able to see this happen. But what’s that journey been from not being on the speaking circuit to, not doing these types of methods all the time.

As many of us have been for a really long time to now having your entire year effectively booked ahead and. 

[00:03:21] Hamlet Batista: Yeah, no, that’s, it’s been a, an incredible and a crazy journey. And I like to explain, what happened, right? What was the catalyst of this? And it started with working behind the scenes last year, around June.

And we were working on an implementation of the merchant account for ourselves. It was using a brain tree from PayPal, worked for about three weeks on the implementation and we, and then, oh, we have to apply for the merchant account. We go apply for the merchant account and they rejected it.

Why did you reject in our, why? Oh, they said, because we were SEO’s we were a high risk. And then I said, why do you mean high risk category? I looked up on their website, the list of what are the high risk categories. And they listed mail-order brides. They listed, penny loans, they listed all these horrible scams.

And I’m like, what? That’s what you’re just saying that we’re, bucketing and us with the same group, I try to appeal to them. They say, no, we’re not going to give you the merchant account. You’re high risk. So that kind of fired me up because I was really pissed because we put the time and investment and I also had the embarrassment say, look, I know I can understand now why a lot of people didn’t want to call themselves SEOs because there’s a really bad group of people that are damaging the reputation of the industry. So I said, okay, I have to do one thing, I either give up change careers, move to another different, industry. A lot of people now call them inbound marketers. Let me call myself something else or get a different industry or do something about it.

And it happened that the, the year before I, Paul Shapiro from Catalyst invited me to do a talk about Python, you know, you write code. And I said, okay. And then that talk turned out to be somewhat, popular when I was talking to developers through SEOs, what they liked about it.

And I said, maybe that’s something that I could, I know I enjoy writing code. I don’t enjoy. at all talking and speaking or traveling. As an introvert that’s not something that I really look for, but I enjoy writing code. I said, okay, maybe I can connect these two things and do something that I can do and enjoy it.

I said, okay, let me teach the new generations of SEO data science so that they can, they can improve the reputation. My generation, we can’t do anything because we let this stuff happen and we have this terrible reputation that is now preventing me from getting the merchant account. But if I give the tools for the new generation, maybe this can happen and I’ll say, wow, but I only have a few hundred followers.

How am I going to even do this? Then I Google it. Jeff, I did this literally, how do we start a movement? And do that, go to Google and try that. I saw these video from Derek Silvers in, from a tech talk, how to start a movement. It’s like a three minute video. But so insightful.

I certainly, he started with some people, a guy dancing, like crazy. Nobody was paying attention, but then he got the first guy to follow him and he nurtured that guy. And then at the end it was just, everybody was dancing like crazy. It was like, wow, that’s such a powerful thing. So I said, that’s what I’m going to do.

The same thing. I’m going to just ride this Python stuff. Nobody’s going to think I’m crazy. Even the first person that follows my lead and do something right. I’m going to just praise them and show them. Look, it’s not just Hamlet that has all these years of experience. These guys learned Python. It was happened. Guy in two weeks.

He said, Hey, I just got inspired by your article in two weeks, I built this amazing thing and. I blasted it. And I continue the same idea that I learning the video and built a crazy movement. And that’s pretty much the genesis of what’s happened so far. I took that movement idea. I’m writing code.

I enjoy it. And I’ve been with. Highlighting people that have taken my idea that learn if, you know, spend a little bit of time and be known amazing job so far. 

[00:07:14] Jeff Coyle: Yeah, I think it’s really inspirational. From my background, I was a computer scientist first, and I’ve been in the search space for 21 years.

I haven’t seen too many people do what you’ve done. And it’s really, like you said, high risk category, I’d say more high drama category sometimes. High risk category, that’s got a been really a smack to the head. But to take that in and speak, community management practices, I think that’s something that, we’re not always thinking about when we’re thinking about content marketing and how that contributed to our personal brand, but, you said the original plan was focusing on it being a movement. How did you make decisions about what to put out there? As far as materials and what formats and things like that. 

[00:07:57] Hamlet Batista: Yeah. So that’s an excellent question. So what I did is this. Looking retrospectively because when I started doing it, I said, I need to get my name out there.

And I don’t enjoy this writing thing or this speaking, but I love writing code. I love figuring out stuff. And I, because I’m running a business, I don’t touch our products. I would love to be writing code, but I don’t trust myself writing production code. I have a team to do that. So I don’t have a reason to write code because I have a team to do that.

And I said, Brian, I crave it. I’m missing that part, I must have the opportunity to do it. And I said, it initially was an escape to have a reason to be able to write code that will actually also give me a business benefit because I can get into visibility. And because when I was writing the Python thing, the first one that I wrote for Search Engine Journal, I said, look, I know, I don’t know that I can commit to do it, on a regular basis, I’m just offer them to do it atad hoc, you got 20,000 views and he said, what? They offer, “No we want you to write a regular column”and I’ve never done a regular column and it’s writing code. I don’t know I can do that, but I said, okay, I’ll take the challenge. I’ll do it. And the way that I chose the topics, this is the cool thing.

And I think it’s a big learning here is. I and when you write for for Search Engine Journal they gave you topics and different writers choose the topics. I said, no, I don’t want to be writing about topics that you came up with. I want to write about stuff that I’m interested. He said, “No, you don’t need topics, write whatever you think.”

And I said, no, this is great. So I chose things that I wanted to learn. Think about that. So I said, I want to learn about this, right? I don’t know what you know, and I don’t have the time to spend it, but now I have an excuse because I have to write a column and I have to write the code to discover something new.

When I discover it with a goal, I ha I’m getting really excited. And then I share it. And that’s like the self motivation to continue doing it because I am not just scheduling with a content plan. I’m saying, look, there are things that are happening that are new, Microsoft for this, there is new API or something that was released that I want to have an excuse to test and play with.

And now I want to learn about it. And then I discovered something really interesting in the process. And when you’re writing code, you don’t know exactly, the problem, but when you start writing it or you feel, wow, I discovered something really interesting. If I learned something with all the years of experience that I have, I’m sure a lot of people will learn as well.

And as simple as that, Jeff. That has been my technique. I just say, I want to learn about this or that. I see that, there was this new model or something that was published and said, I might be able to use it for this. Let me see if that works. I try it. Write the code, learn something.

And it’s and because I’m writing the code, that’s the motivation for me to do the work. Both for the speaking and the writing, I write the code first, learn something, and then I do the write up. Oh, this is what I learned. Or do the deck for the presentation. This is what I’ve learned, right? 

[00:10:53] Jeff Coyle: I love that.

When you say, when you said that, why write up, what you learned? Are you thinking about that while you were working with what you’re going to write? Is that you just, you wait, that’s your.

[00:11:02] Hamlet Batista: I don’t know. 

[00:11:03] Jeff Coyle: It’s your. 

[00:11:04] Hamlet Batista: No, I tell you that’s the worst. The crazy thing about it. I don’t know exactly what I said that I want to write.

I just have a blur idea. Both what I want to learn. And sometimes some articles, I have an initial idea and it doesn’t work and I have to try it a few times until I get to something that I feel, wow. I really learned something. And that’s why sometimes I have to trust it started, way beforehand.

But now if you think, wow, that’s like a lot of work. The thing is that I, there are a few things that I. The benefit, right? The main benefit is that I’m learning something new that I can apply to the business. The second benefit is that when I share it, and when I write, for example, for Search Engine Journal, they tell me how many visitors they tell me all the other metrics.

I see the engagement and I gauge which ideas are more powerful. Some ideas that I’ve shared. I’ve shared it before. Even we write them in the code or put it in our product. And someone say, wow, people really love this one. And then we put it in the sprint for developers to put in the product. Because I see not just that the idea was good, theoretically, but people, their feedback that they provided us, it was a valuable thing for us to embed into our technology or our service.

And let me tell you the, even the better one I came up with, it’s happened to me at least three times. I hear one idea that I thought it was really good. Somebody from the community took it a step further and came up with even something better and that’s even better for me. Cause now I say, wow, they even refine my idea.

So now I can even deliver even better service or even better features in my product because they show me ways that I even think about. 

[00:12:33] Jeff Coyle: Nothing’s better than that coming from a product perspective. It’s like you said. Another question I had is, taking feedback and, being able to spotlight the person like in your community that, “Hey, you influenced me.”

And I was the first person. You did the follow up was fantastic. When you, if you’re going to share the output as well as the code, does it cause you to code differently? Does it cause you to plan differently than when you’re working on a black ops for intern? 

Not necessarily because one of the things that I realized right, is, oh, and you’re in the SAS business, you’re in the IP business, like how much should I share?

How much should I keep internally? And you have to be strategic about deciding, what is going to be your competitive advantage. And one of the things that I see is that the technology part. That seems to be traditionally the best, the most, the competitive advantage is not that it’s been blurred because of how fast things are moving and because how fast things are moving, if you’re thinking about, oh, I came up with this idea and that’s my IP.

I don’t want to share it because that’s where my defensive is on that. By next month, something better came up that it’s going to make it obsolete. So that’s what happened is the best ideas are, have very limited time. So I think that the community that you build in our case is a stronger, competitive advantage, building an engaged and committed that trust the trust that you build around the community that wants to work with you now because you have a secret or something but because you are consistently.

Improving and innovating at a faster pace and giving them what they want and listening to their feedback. I think those are things are stronger. So that’s what I, don’t what I realized that because before I was thinking the other way, I say, wow, I don’t really, it doesn’t really matter whether I’m sharing the best stuff because by next month or by next year, there’s always going to be something better.

Yeah. 

That’s a, that’s the role of a great product leader is to say what took to make that justification to say that this is our secret sauce. Maybe that isn’t, and maybe this is going to help people in a way that could grow our audience and grow our engagement. So I think, it takes a level of maturity to be able to separate.

To separate and make those decisions. You gotta be at a particular level with your business. You gotta be at a particular stage with your business. Some businesses are venture backed. Don’t have that luxury. Some businesses are profitable, others are not, you gotta really take a lot of things into consideration.

When you’re deciding what, what might be shared, what might not be in, how truly ahead of the curve, you really are. I think a lot of. Think that they are way farther ahead than they actually are. 

[00:15:04] Hamlet Batista: Yeah no, absolutely. That’s what I said. You have to consider your specific situation.

And also Jeff, there are all the things that are more that I stand necessarily in the knowledge, but on the experience. So you talked to, a surgeon. Oh yeah. He can write a book about everything. Hey, and that he knows you’re not going to be able to replicate why nobody’s going to trust you even.

Oh, I read the book at a pub, a surgeon published and you can trust me. We’d be doing the surgery on your own, the actual practical experience. It’s it’s invaluable. So being able to say, no, I’ve been doing this with all these different sides or lead these different clients. And we’ve seen this in this success and I’ll write a book and tissue, everything that I’ve learned and you, who are you going to hire?

Who’s going to hire, you’re going to hire the person that has done it because they have. Experience of doing it. They can innovate when they face new challenges than the new person. That’s also the other advantage is that 

[00:15:56] Jeff Coyle: yeah. And it can, it connects to empathy. And you mentioned it in your audience building too.

And so other than, just lifting up anyone that bobs their head up and, and shows themselves in your community, when you’re talking technical to, traditionally non-technical audience timid people or someone who might believe that they’re behind with the community in their knowledge, how do you get them, to participate, raise their hand, which is, who might think, man, I’m behind the trend.

I couldn’t write a line. I learned basic in grammar school and I haven’t done anything on this, but I kinda wants it. How do you get them to speak up and be part of the. 

[00:16:32] Hamlet Batista: Yeah. So the way that you do that is that you show people similar to them that have already crossed the line.

Look, this and I mentioned, for example, Ruth Everett, that she learned from the CHRO. And she said, Hey, I was scared about these spikes and things. I did a women are with them and she sold the cool stuff that you could do. And even with simple steps, cause I show, I first show the outcome and I like to use this analogy of weight loss, right?

Nobody is selling you. Weight loss shows you, oh, look how beautiful the weights are or the sweat or the hard work. Nobody shows you the work. They show you the outcome. They show you the bodies, that’s what you want. As. And then that’s the similar to what I did. Look, this stuff you’re doing manually and step-by-step.

Look how that is possible. And now let me show you simple steps that you can do it and try to show the simplest thing, simplify things so that they use cases. They can see that it’s accessible. And then when somebody, you just need one person that didn’t know anything about to do it and say, look, this guy did it, this guy right now.

The other thing is that the older learning to tell about the older learning about taking this approach is that I saw. The power of getting the best out of people, right? A lot of people, they underestimate their capabilities, they underestimate what they can do. And marketers, typically we fear coding, we’re technophobic.

We don’t think we can do this or that. And then when it said. You think you can’t do it, but look, this guy thought the same thing and look what they did. And now what’s happening is that as more people started doing, and that’s one strategy that I’m using my articles. If you look at the articles that I wrote from last year, I’d have a section of the end community.

These are the community projects. And before I’m going to publish the next one. Hey, who’s doing some really cool stuff too. We’ll give you a shout out. The bigger my audience, the more visibility they want to, they wanted the PR they want the visibility. So I said, look, I’m going to shout out and sorry, I can give you visibility.

So they get distracted, 

[00:18:37] Jeff Coyle: right? Yeah. Yeah. I love that strategy and it works. It’s worked in this space so well for you. The nuance here, I think also that, the technical search engine optimization community, some of us, some folks are getting it. Some folks are not that traditional developers or people that are developers by.

They sometimes look at these projects as having less credibility as well. So they’ll roll their eyes. A CEO who is writing some code on the side or the, the fact that there’s an avenue of a community who’s doing, develop, trying to figure out coding on their own.

But when you show the impact. With these projects and you get, some moments, they have to begin to take notice that this is a thing, it’s not just that, we do this is our job, we’re developers or we’re data scientists. It really illustrates that you can come together and build great things.

And. I like to think hard business as an example of that, where we’ve got, an amazing data science team, amazing development teams, but our inspiration comes from marketers and our inspiration comes from marketing and content strategy. But you’re a great example of this. I just think it manifested.

So what before we get into some fun Python tech, SEO stuff what would you do if you. No other than, knowing what your movement is that you want to inspire. What would you, what advice would you give, like a quick snackable list, four ways to do this for your oneself, for their own personal brand.

[00:20:05] Hamlet Batista:

And you have, a little boat and you’ll have a gas model and the, and then you’re trying to push your business, getting the attention that you need. And then you have another boat that is just taking advantage of the winds. Finding the strongest winds and just moving the boat through the winds.

So it’s a lot less effort and you make a lot faster progress. The equivalent is that you find waves in your business that are already happening in your industry and finding an emotional connection that you. Connect your business with that with that Trent, right? Yeah. There’s a lot of trends, right?

There’s global warming. There is, the, a lot of, the millennials don’t believe, fake pig, fake ass, fake people. And they S there’s a lot of really interesting movement. They have a social economic impact that you can align your business with those so that your business is not everybody is in the business to make money in and try and trade goods for services.

But when you make it, you create a greater purpose for your business and you align your business with aggravated purpose. Like samples are in e-commerce stores that are using, recycle Debris from the ocean, right? It’s an ocean cleaning project. I show it’s really interesting or people that are trying to take care of the areas of the, there’s a lot of different ways that you can tap into existing movements, that they will propel you a lot faster than you trying to do that.

And the example where my business is that I showed the data science and they eye movement. Taking, I could have do this with R a lot of this stuff, and I’m doing this another programming language called R by them. I could have done that. But that one is growing the fastest. So Python is growing the fastest growing like crazy. So I exactly tapped into something that I was already seen taking, a strong current, and you’ll find that in your industry. And then I tap into that because I said, eventually this is going to also happen in my community. Let me be the one that brings it first.

[00:22:30] Jeff Coyle: And there was already a crisp. And is the Hamlet of of our, or you can say you’re the the Chris pan of Python, 

[00:22:38] Hamlet Batista: right? Yeah, exactly. So the idea was that right, so that you find. A movement occurring that is to really happen and find that connection, find that emotional tie. And in my case, the tide is that marketers are taking a fallback, right?

And it’s people, there’s a reason why people are not doing it or teaching marketers code because that’s not something that marketers want to do, but that’s one way to do it, but to think about it, but it said, look, yeah, They don’t want to do it, but if I show them the outcome of doing it, if I show that their reward is higher than the effort, then you can, you don’t need to convert everybody, but enough people can create enough if enough noise to do that Naturally

[00:23:20]And the one key add on when you said there is it’s how severe and known is. And I think naturally agency EDC team members, or they growing through an agency or being in a larger business, you tend to default to manual tasks and rip it’s that repetitive task. And you don’t even know if there’s a better.

And I think if you inform that there’s a better way, even for mundane tasks. I, like you said, I don’t care if you’re programming in R if you’re in Python, if you’re programming and visual, basic to sit on top of an Excel spreadsheet, you can get more work done if you do that. And if you’re doing repetitive tasks, you’re hurting your business, not even just their visit, you’re hurting yourself and you’re probably driving yourself.

[00:24:04] Hamlet Batista: Exactly. Exactly. So they will see Jeff that’s why it wasn’t that hard of a pitch because he’s already a non pain that they have that they even know existed. Exactly. And then the older, the other thing that I did is that I show agency people from agency that Twitter and Hey, for example, my Cuesta.

You said, oh, we have to set a new way, a new hire that just a few months. She’s reading Hamlet’s articles and did this amazing visualization from people. Also ask you, scrape the surface. And created an amazing visualization. I should retweet it. So it’s also the compounding effect of when you empower, when you push people to do their best.

And I bet it on the new generation. I don’t know the new generation purposely and they’re excited. They’re doing some work that, and then they’re getting praise, it’s a compounding effect because if they also become your ambassadors, they started doing the word of mouth and that’s how I keep getting more opportunities to speak and present as well, because not just because of me, but because all these people that I’m empowering giving the good war about, oh know, Hey, how many, Hey, we learned dish.

And we learned that it’s a virtual cycle. That’s how, that’s 

[00:25:10] Jeff Coyle: how you get to the point where, there’s an event at the digital analytics association in New York. And. I’ll give you a call and that seemed more typically the 17th. They said we’ll do some plugs at the end, but let’s talk about icon.

Really? What makes it special? This is interesting. And so my hot take would be, the libraries make it compelling, the mass usage, the community, the momentum, there isn’t anything special or unique, but the ecosystem around it makes it very special right now. What’s your take on that kind of, am I, is that blasphemous or how do you think about why Python?

[00:25:50] Hamlet Batista: I will tell you it’s is, I think is more a UX thing, right? So I’ll tell you because I program in a lot of different languages. ’cause I like you, I have an engineering background, right? So I’ve done assembly. I don’t job up, all these different languages right now. What happened is I was thought I was thinking back, why am I drawn back into Python when I was doing coding and other different things.

I was like, why am I drawing back to Python? I say, wow, it’s just that Python. They D they say UX. They designed the language in a way that, for example, when I had to pick up a script that I wrote in. Years ago. I wouldn’t understand it. I will read it. And I said, I don’t know what I did here. I don’t remember details about the variables Kobes.

That is a fundamental thing in parallel. I have to look it up, so I have to look up for basic things to give an extreme example with peril. I don’t know art, to give an extreme example with Python, I’ll pick it up immediately. I will read the code. I know what I was doing. So Python was designed to be.

So you don’t need people write comments because but a lot of languages, you need to be writing comments. You’re writing, coding. See if you don’t write the comments, you wouldn’t understand what is being done, but Python, you should just read it. And the same code is that it’s a, what am I the developer says, oh, I don’t need to write code.

I don’t need to write comments. The language is the code is a, is the comments is explaining exactly what he said that he’s doing. There are exceptions, but I think it’s the readable. He does the spacing thing, which is super frustrating, but it forces you to write in a clear format that it makes us super easy to read.

Right? 

[00:27:23] Jeff Coyle: Very elegant. It’s an elegant, and readable some old in the past and, it’s. Platforms in the 90 some languages that were made in the nineties and early two thousands tried to do, whether you’re talking about eight or small talk or, there’s some of those, but it just, it didn’t have any actual applications that were meaningful.

And it wasn’t powerful. This combines that because I feel like the libraries are the, to get there so beautiful.

[00:27:48] Hamlet Batista: And all, and if you think about it, right? One of the weaknesses that I had when it was, years ago that performance and all that stuff was eventually a very minor thing as the computers improve the performance and the capability.

So the weakness, it was no longer a weakness, but the strength of the readability becomes a really powerful benefit that I have. 

[00:28:10] Jeff Coyle: Cool. So a couple of questions you know, so for a For, we can start naming them, but I’ve got a couple of questions about specific Python libraries for the specific use of SEO, technical, SEO, natural language processing.

What are your, what’s your top top pool there. 

[00:28:26] Hamlet Batista: Yeah in terms of practical applications, Uber has it, my favorite and the simplest one is a tool from Uber that is called Lu. They release it. It’s like a A knife of deep learning models that you can just define a Jamel file which is very simple configuration file.

You said, okay, this is the input that I’m going to provide in a CSV. And this is the one of the models, and this is the task that I want to do. I want to do a classification. I want to do, an image, a captioner, you define what you want to. And it will do the whole thing that you typically have to have a, an expert deep learning program or to code, it will do the coding of the input in the right format.

It will do the split of the datasets and they will do the cross, our relation. It will do everything at the end. You get a model that you can start doing prediction. So it’s incredibly fast because you can do deep learning at scale advanced, deep learning. With bait, virtually writing note code. And that’s the easiest one that you can use. And there’s. Powerful applications of this. And if you want to check some examples that I’ve done in search engine journal, one of them for example, is intent classification, which is something that for agencies is it takes a lot of time to do manually. They can all these millions or hundreds of thousands of keywords and grouping into the right buckets based on intention.

That’s something that takes weeks. And I said, I wrote a one on article. I said, look, guys, you’re doing this spending weeks, doing this stuff manually. And with Excel formulas and stuff like that, I look at this, I can do this in minutes and here’s the code. You can actually copy that. And they were blown away by that 

[00:30:02] Jeff Coyle: a product releasing into the market this year that we’ll do that.

[00:30:17] Hamlet Batista: Yeah, exactly. Yeah. So that’s my library, right? Yeah. So that’s my library. The second one is a transformer. So this one needs a little bit more. From that the great team at Hoggins hugging face, which is it accomplishes similar, more advanced things you can do with transformers, but it requires a little bit more code and more knowledge, but it’s still far simpler than, doing, direct TensorFlow code or or PI torch code that you have to do.

And something that you needed is pandas is like Excel in a sterile. And does it, is it is something to say yes. Yeah. And talk a little bit, Jeff, what are more higher level for content marketers, why, because why would they say, why do I have to code? Why do I code when I can just hire a developer to the live work?

I know you didn’t ask me that, but I wanted to mention that because. I get that often, right? Oh, I can’t just harder to be able to do that stuff. And here is the thing, assignments yet we all are trying to get attention. We’re trying to break out and find out idea like I did with the spices and stuff to break out and get all this attention.

And let me tell you the good news. He said we are, there’s so much data and so much information that your idea is definitely there. All right, but now let me tell you the bad news. The problem is that if you hire a developer to look for that idea, they don’t have the context to be able to find it in.

Here’s the reason. Let me tell you the recent and tell you how marketers, because we do the Mar we are both marketers and engineers, right? How, and you relate to this as an engineer. The first thing Jeff, that we were taught to do Y is to remove the context, right? So we’re talking, math, the first thing you do, here’s the problem.

And then you’ll turn everything into wearables because you are, your job is as an engineer is to generalize. So equations math, formulas algorithms is all about generalization and you have to remove the context, right? And that’s the context of an engineer. So because, but as a marketer, you need to connect the emotional.

To your TA to your target, persona, to the target customer, right? So you have to add the context is the most important things. You need to know who it is. What is the job? What is the person who is the family? You need to know everything that you can to make a relevant and personalize, pitch and message, right?

And content connect to them emotionally. So you have to do the opposite as a marketer. You are need the context, but as an engineer, you need to remove it. So when. When you switch from your problem to an engineer, they are not able to make that emotional connection because they don’t have the context. But when you are both a marketer and you learn a little bit of coding to dive into the data and slice it and dice it, you can ask the right questions and you ask questions.

And when you get the data, you can continue to drill down and follow up with the questions until you get. Wow. I really found something really powerful and unique. That nobody else knows. And I can capitalize on this to to define the story that I’m going to bring to the market that is going to push them into the spotlight.

That is how that. That is why you need to learn a little bit of code as a marketer so that you can capitalize on that gap. 

[00:33:40] Jeff Coyle: I’d like you to work. They let you work with developers much more effectively as well. Yeah, that has always been my life, and, work with many development teams and just being able to, it’s not just to call BS on things it’s so that you can respect the work that has to go.

And really you can focus on training, even your development teams on the art of search engine optimization on the Marta content strategy. The value of being a subject matter expert on something, because a lot of times, like you said, the soul comes out of it. When it gets turned into a spec, you can bring the soul back in.

And when when you can show that, I think the way that you said it, I think it’s so appropriate for this discussion because. And like you said, if you initially, threw this out to someone who has no idea, anything about content marketing, content, strategy, search you know what you’re going to get, it’s going to be like explicitly what you put together in those requirements documents.

And if you’re a marketer, You likely, aren’t going to be very good at reading requirements documents. So you’re going to be a lot of stuff lost in translation. It’s just happens. And I know it from trial and error and hope from what you said is reality. I think it’s a really awesome example for sure.

For sure. Yeah, for sure. Tell me, tell some examples, like you mentioned the user intent classification, what are some other projects that. You have seen done we’ve named dropped enough this week. That’s very unusual for this this show, but if you want to give some credit to recent projects that you’ve seen that are really interesting that people have done with Python or related technologies.

And by the way, libraries we love to mess around with Spacey is really great and LP natural language processing. And and obviously you said all the corridors, but there’s a couple other ones that, for NLP is another one, but there’s a pile, really dig in.

There’s probably a, there’s probably elaborate for what you like, at least adjacently or near what you’re trying to accomplish. But as far as projects you’ve seen implemented that connect to SEO and content, what have you seen. 

[00:35:42] Hamlet Batista: Yeah. It’s the work that the community is doing is it’s incredible.

For example, Jr. From from locomotive incredible work, a big guy. Yeah. He’s a big fan of you guys. He published a search engine journal article about class clustering keywords using an algorithm called. It, you go to search any line, look for Jr. He polished an incredible project with a notebook and everything that he’s pulling data from search console, and he got the project translated into other languages.

So Vincent Harare from on took his code and wrote it for German and all the different languages. So that was really cool. So that was that, that, that was a great project generating meta descriptions. So I wrote an article for search engine journal about S by using abstractive summarization is completely noble meta description by, within the content from the page and generating a really compelling one.

So I did one for search engine journal using Burt using and what I show people is like, there is this cycle papers with code that you can keep track of the latest, the state of the art, and they actually gave you the code. So you just take the project, clone it, and you can make, and you can do applications because these are academics.

They don’t know the business value of this, and you can apply your accounts. To use this to things that are going to solve your problem. So I did that. Andrea will, penny took the same concept and he did it with with the library from hugging face transformers as well. The generating of those Christians with summary.

Really cool stuff. There is also another project too, from Facebook called that uses computer vision to take, images and captioning them. So a lot of the stomps, you have all these images that don’t have descriptions. Amen. All text. I show in a, in an article as well, how to use that project for that.

And there’s there’s one article in the queue that I sent over to waiting to get published, that he’s going to be using question and answering. And I like that you mentioned about Spacey because I, I had to use in order to educate people about what embeddings are, what is an embedding and stuff.

And I used a GPS analogy to use, and I use Spacey to show the difference between. Context free embeddings and context and contextual embeddings, which are the ones that you get with Burt. One of the big, as Bert as being one of the biggest breakthroughs in the deep learning community.

So I show, look, this is one visually with Spacey, I show you loading. I take a sentence showing link. And the sentence said, look, I gave you a pre-phase of the article without the, with Albert Euston Spacey look at when I compare the similarity of Lincoln and the sentence and why it was the Lincoln tunnel.

And other was the Lincoln university. The first one, without the context embeddings, it was think they’re the same that they mean the same. But when I use the Bert embeddings and do the same comparison he sees is that they’re different. So bascially, you can see. And I was showing that with writing code, you get a better understanding.

You get a more powerful understanding of what are these things, the matins and the code and all that stuff. So it was really cool. 

[00:38:43] Jeff Coyle: No, that’s it, those are some great examples of, you mentioned the attractive summarization and there are still businesses that, for, the journal industry, generally industry, people, white paper promotion, that’s where I came from was, generating leads for B2B tech for a long time re handwriting abstracts, and doing writing promotions for emails.

By hand at scale with teams of teams and, the, there are so many applications that relate to natural language processing. And, obviously I’m a bit bias, says, that’s the foundation of what we do, but we’re, one of the big things that we’re focused on is generations and being able to take outlines and build, drafts of content.

And that’s something that I think is going to be moving in. Few months a lot. 

[00:39:28] Hamlet Batista: Yeah, for sure. I’ve seen your content brief. It’s incredible. It’s amazing. It’s actually 

[00:39:36] Jeff Coyle: excellent night. So we had a few other questions, but we’re running them. I’ll get it.

[00:39:41] Jeff Coyle: One more question. And then we’ll talk about some events that we’re doing as well as As well as a few other points of reference.

So the one question was you had said that you’re running an agency. How have you used this to generate leads for your agency?

[00:39:59] Hamlet Batista: Yeah. Yeah, so we run a software business and and through. To generate leads. We primarily use partnerships for, because we have a software with a freemium model and the freemium model. It generates a lot of leads just because you have a fee offering. With the writing that I do get leads, but it’s not as many as we get from the freemium model and the partnerships, but we do get good and quality ones from the writing and the speaking.

And. So I wouldn’t say that the primary focus of us is lead generation with a content strategy is more to build up the trust that leads to those partnerships that we can. Because the partners, they not, people want to partner with people that, they trust. And I it’s going to make them look good to their audience. So this is more a long-term play of, Hey, look, this guy can consistently deliver, are these companies going to consistently deliver? Yeah. They could make for a good partner. We done not going to worry about our reputation when we refer customers to them.

[00:40:56] Jeff Coyle: So really building up your brand. Focus on, partnership development. Those are gonna be great channel partners because they’re going to get to know your business. Those are going to be great connections, your content, you may get stuff in the front door, but it’s going to be just like anything else.

Not, I think that’s really great advice, frankly, as it goes. 

[00:41:14] Hamlet Batista: Let me mess you up. Sorry. I want to mention related to that, right? That I think the biggest takeaway that’s the biggest thing that I’ve learned from being doing this for over a year is that I think that consistency is underrated. So all the times is, oh, I’ve had to write content.

Like you see, like when you have a schedule that the webinars are going on a specific cadence, that consistency that when you are writing content on a consistent basis, That builds up over time. And the main reason I was thinking, why is this consistency so powerful? And it is because you are training people to expect great stuff from you, and then, especially when you’re consistently exceptional, when you’re trying to consistently get better. Yes. Like you’re having a you’re creating like a channel. You’re creating like a channel for people to, I got a comment from somebody pricing, one article say, oh, every time I see, one of Hammer’s articles on their home page.

So search engine journal, I know it’s going to, I’m going to learn something that it’s not going to happen with uterus writing to two articles. It’s not going to happen with writing one or two articles or writing them in consistently when you’re doing things on a consistent basis and hammering. A theme, a movement and you gain that benefit that people will attend when they synchronize, because they know that you’re going to be better, you’re going to be delivered. 

[00:42:35] Jeff Coyle: If you can commit to making sure that everything you publish is going to be novel, more comprehensive. Every time you hit the button every time you hit the publish button, really commit to that. And yet that has such an impact. That’s every client I work with.

It’s. Really discussed that it’s like, you can make sure everything you publish is equal to or better than your competitors every time, the people who were thinking about this. Look at the example, look at the example of, in our space, look at Kevin indig. I mentioned him before everything, everything he publishes is, has a lightsaber attached to it.

And you get that, you get trained and to say, wow, this is going to be good. And you, that’s what you got to focus on is to say I might have some like Musk, I might have some things that we’re going to, like I have to do, I’m not terribly inspired, but at least make it great.

Make it as great as it can be even if the topic isn’t inspirational. So I think that’s really a great note to finish on it’s building your personal brand is consistency. It’s making sure it has a, the focus of a known pain it’s using the momentum that already exists in the channel. Using you in the Wednesday and then, just staying focused and you keep going.

[00:43:48] Hamlet Batista: inconsistent. Yeah. Let me interject also, too. So that people also think about from the other side, right? Why would a publisher, accept your writing or apple conference, want to hear your speak right? Is because when you are consistent, and you’re delivering in a consistent basis, you remove the.

So they know, oh, if we take Hamlet to speak about this, we know that he’s going to live. So that is why you have to fake thinking when you’re consistently you eliminate the rest and they want you because they know that it’s the chances of you making them look bad. It’s really low.

Totally. Totally. Yeah. 

[00:44:24] Jeff Coyle: It’s straight forward. Yeah. You put yourself in their shoes too. As an event manager. One little nuance secret to his tone Pitt. When you get those speaking opportunities, make it about the audience. We could have a whole, we could have a whole webinar about 

[00:44:37] Hamlet Batista: what not to do when you’re pitching, 

[00:44:40] Jeff Coyle: but that’s a really good one don’t pitch.

I remember it was a walking up to me saying, thank goodness. We finally got somebody who goes in pitch. It’s really good. It’s really cool. So now I’m that note? Okay.  What are the free tools?

[00:44:58] Hamlet Batista: Yeah, so we have a Chrome extension, right? So you got to rank sense free tools that allows you to do, we call it a 15 minute SEO audit.

So make sure that you install it with your Chrome browser. And I write, I wrote an article for search engine land. If you search for my name about a common problem that I see on enterprise sites. You add a tag manager on the head and then it will push it, affect your SEO and you wouldn’t even know that was a problem.

So check that out, check their article because our extension, as far as I know, our Chrome extension is the only ones that the texts is problem. And in the article, I explain why that happens. And in the article you also find the link to the extension. It’s really interesting. And I gave a talk at.

West last week that I was showing tasks that I did with Google running their, your inspection tool in Google. And I was able to confirm that those are real problems that affect SEO. Places that third party scripts incorrectly in the head when they belong in the body really powerful stuff.

So that’s a free tool you can get from our website and other free tool is if you use CloudFlare or considering a cloud for ACEs, CDN to speed up your site, we have a free SEO monitoring tool in CloudFlare. So you can also find it on our site. You can install and it will be monitoring your site on real time.

And then from that tool, you learn a little bit about our premium offering that allows you to implement changes quickly to fix issues that you find. And we have a Python meetup next month, 11th group M offices in New York and the beautiful offices in the empire. And Portray center ma save the date.

It’s going to be for people that know, marketers don’t know anything, we welcome bring your laptops, your questions, your problems. We welcome. Any question, any challenges, any mistakes you make, we’re going to be there. Hands-on to help you March 11th, six to eight. Search for that.

And meetup is called the Python for a CEO’s meetup group in New York. And we have that going to have our first event. I’m doing it in collaboration with catalyst part of the group and Paul Shapiro, which is the head of innovation is going to be with me, helping me without events. So it’s really excited about that.

[00:47:13] Jeff Coyle: Paul is the best. Bombed than I am in New York the week after. But that’d be a great event. I’m sure I’ll hear all about it. I will be at content tech in in April, we’ll be talking a lot about these types of topics, natural language processing, natural language generation. And also the findability conferencing Raleigh.

You mentioned Dara, so I wanted to call that out. You guys really some really good events coming up. So set your calendars for sure. And then on our side we have a trial almost equal don’t know that we have a trial now on our site and the self-service offering MarketMuse pro it’s certainly much more for an individual practitioner, the protocol.

But if you’re interested in just getting a taste of what we offer get on the site, sign up for the trial. If you want to reach out and talk about some of our more premium offerings. Shoot a note to anybody who to know to the bot shouldn’t know to me and we’ll make it happen. But I can look thank you for joining us today.

I think this was a great session thinking critically about how to set yourself up to be successful and showing them with real stories. And that like you have. And so I appreciate the time and I will see you in a. 

[00:48:23] Hamlet Batista: Awesome. Thank you for having me and thank you everybody. Thanks.

[00:48:27] Jeff Coyle: Bye-bye.

Stephen leads the content strategy blog for MarketMuse, an AI-powered Content Intelligence and Strategy Platform. You can connect with him on social or his personal blog.

Tweet
Share
Share