Creating a Programmatic SEO Operation with No Code

0:00so uh today i will be sharing with you how um we have used no code tools at failurey to run a programmatic seo content strategy and uh to create content at scale and grow our organic traffic to eleven 000 users per month by 11 000 users per month our current organic traffic per month is around 80 000 users i'm the founder of failure this is failure here you can check it out at failurey.com and this is a scalp uh you can check it out at getscalp.com uh so it all started uh early this year when i was defining failures goals for 2021 and one of these

0:40goals was to grow our organic traffic from 2000 users approximately to 5000 users per day right so as part of this goal i started working on uh some keyword research and writing down dozens of startup related content ideas to publish on failure slow which is this one that you can see over here after a few days i had a huge list of interesting keyword ideas along with their monthly search volume and keyword difficulty but i didn't really know where to start that's when i realized an interesting pattern in this list which was that i have written a few content ideas that were related to pitch text this content

1:25ideas were vintage startup peace texts food and beverages starts at peach sex and media started peach sex that's when i came up with the idea of creating dozens of articles of the style industry startup pitch text i will be referring to industry startup pitch sex as a content type besides i will call all of the articles that form part of this content type as content topics for example vintage startup pizzex is a content topic these are some terms defined by pat walsh in his online course lean seo which i think that will be helpful when explaining this story so i have this content type idea but i

2:08still had two big questions that i wanted to figure out before moving on with the project the first question was how can i evaluate this content type without spending a lot of resources on it and the second question was how can i scale this content type without spending a lot of resources on it first i'm one with validating the content type to do so i applied the lean startup mindset and created three really simple articles on vintage startup pitch text food and beverages startup pixels and media startup pitch sex each of these contains 5 to 10 pitch sets from startups in the corresponding industry along with some information

2:46about these startups i think i spent around two hours to write and publish the three of these posts one month later i checked how these articles were performing i only checked will search console because i was only interested in seeing their organic traffic and not the traffic coming from other sources such as i don't know social networks the results were pretty good as you can see here um the articles had gotten over 2000 impressions and they were getting they had got 58 50 and seven clicks this was enough variation for me to start working on scaling the project but in that month that i was carrying out

3:27the experiment i came up with a few related content types that i also wanted to validate before moving on with the project these content types were company pitch tech for example airbnb pitch deck stage pitch text such as series a pitch text customer type start at pitch 6 such as b2b starts at pitch 6 and business model start at pitch 6 such as b2b starts at p6 or a sorry such as sas starts at pitch 6 or marketplace it starts startup pitch text in order to find if these content types were also a good idea i wrote and published three new simple articles titled shopify pitch tech pre-seed beep

4:10sex and seed pitch text and i wait one more month one month later the results were really impressive all of the new articles had gotten a few clicks even the the pc one that you can see in the in the third row of the table had gotten over 50 clicks and all of the articles that i had published initially were getting a few clicks per day and another positive thing was that both the impressions which are indicated by the purple line here and the clicks which are indicated by the blue line we're showing a really clear growing trend so i consider the content types as validated and it was now time to start

4:52scaling the project this is where programmatic seo comes in there are thousands of companies business models industries and i wanted to have an article for all of which were interesting from an seo perspective i need to find a way therefore to identify which content topics were worth it and automate the process of creating articles for these topics right so i came up with this four step scaling process that you can see here the first step was collection this means collecting on another table the largest database of pitch decks possible along with information about the startup including their industry stage customer type and business model i would

5:36later on use this information to filter which pitch text will go on each article the second step was enrichment which consisted on enriching the database with information about the startups and their pitch text which might be useful for the articles readers the third step was preparation which consisted on preparing everything to upload the databases to workflow and make the creation of the articles faster and the fourth step was creation which consisted on uploading the files to workflow and seeing the articles get created as if it was magic through programmatic uh programmatically created right so yeah let's get deeper into each of these steps the first step was a

6:18collection we combined some scrapping tools and manual work to build the largest database of pitch decks are possible we use no code tools particularly one that is called simple scrapper to get pitch text from collections made by other websites that you can see uh here these were some collections of pitch decks that other websites had we complemented this with manual work to find pitch text that these collections didn't include we use google and search queries like site column slideshare.net pitch tech to find pitch text host in slicer which is the preferred tool used by founders when scrapping and manually collecting the pitch text we also extracted as much

7:03information as possible from for the startup including its name logo website url description industry stage demand breaks the year of the pitch deck among other types of information after a few days we have this huge article that you can see over here with 352 rows of pitch sets and a lot of information about each pitch day the second step was enrichment we proceeded to enrich each row with information about the pitch check and the startup that readers could find useful first we manually checked each startup to identify what was their business model and type of customer second with some scrapping tools we found our twitter and ancient list

7:48profiles urls and then with some ai writing tools particularly one that is called wilbot we rewrote the startups descriptions that we had got from our websites we did this in order to avoid having duplicated content on failure which can be penalized by google right so so yeah this was probably the most time consuming part of the of the process i think that it took around one week but after a few days we had this huge air table with uh 352 pizzas and a lot more of information for each of these pitch sets the first step was uh preparation before being able to programmatically create the listicle articles of pitch

8:32sex for different industries business models customer types and stages we need to do some more preparation work let me first talk about the content type pitch a company pizzagate each of the rows in their table was one startup and a speech set each of these rows had to therefore have a specific page about it on failure i will call these pages an individual pitch check page for example the row that contains information about airbnb and speech deck will be turned into a page focused on ranking for the keyword airbnb pitch deck and that will later on look something like this so you can see here that it only contains um the pitch tech

9:16from airbnb and some information about the startup and a description etc within the air table we already had all the data that we wanted to display on these pages such as the pitch deck and the startup's name industry business small stage customer type twitter crown trace financial list urls and more but we need some more data for the meta title meta description and social media sharing image of these pages we wanted these pages to rank on google for keywords of the style company pitch tech so we need to set up all the seo basics correctly we first created the meta title and meta descriptions using templates and search

9:57and changing certain words by the information of the startup for the meta title we use the formula startup's name stage pitch the in here vertical bar failure for example for airbnb the metadata the meta title is airbnb seed pitch tech in 2008 vertical bar failure then we created a social media sharing title using a template quite similar to the metadata one you can see that it is slightly well it is different uh this is the meta title and this is over here the social media sharing title this was only because we thought that a for for social media sharing purposes it made more sense a title like this one than one

10:40like this one and we combine this social media sharing title with the startups logo using a tool called plasiv that allow us to programmatically create images and oh yeah we created these social media sharing images this is the example for airbnb so now we have this huge our table uh with a lot of information for um for each startup and the meta title meta description social media sharing images that you can see over here now let's talk a little bit about the preparation of what i will call p tech listicles these were the articles focused on their content types that were not the company one these were industry startup pitch

11:24sex stage pitch text customer type starter pitch sex and business small standard p6 a bridge stack listicle for example is crypto startup speech text here is how this article looks like nowadays um here's like the title uh and the top of the article and this is how the information is presented within the article from our earth table of beach sex we extract all the possible industries stages customer types and business models and wrote them down on a worksheet along with the number of times that they appeared in the air table this means the number of pitch texts that each of these topics will have on the

12:06article we only stay with those topics that we have that we have more than a two pitch sets for a actually it was three page six or four yeah so we then created a meta title for each of the topics combining different uh variants of text and the topic itself for example for the cryptocurrency article it was five page sets from from big cryptocurrency startups following a similar process we created a meta description for each topic for example here is for the cryptocurrency one uh it says here is a list of five speech sex examples from well-known startups in the cryptocurrency industry including coinbase and google coin

12:47we then move on with creating a blog introduction for each topic using templates of text that we created with another ad writing tool that is called copy ei and we changed some words on these templates by um the topic name no right so for example in the case of cryptocurrencies the vlog interaction is cryptocurrencies are currently the hostess trend in the market and it is clear that they are here to stay here are five speech techs from well-known startups in the cryptocurrency sector that should provide inspiration on how to put together a good one unlike the individual pitch tech pages all of our blog posts have a cover image

13:26so we created a really simple image for each topic using a background and an icon from icons eight here is the example of cryptocurrency cover image and finally using placid once again we created a social media sharing image that included the meta title and the cover image that we had just created this is the example for the cryptocurrency article not how we have used air writing tools no code tools and google sheets features and functions to automate this process a little bit and make it faster so once we have finished this preparation step we had the air table that i showed you before and this huge wall sheet with all

14:10of the industries business models customer types and um and stages that we had a pitch text for and a lot of metadata for each of them for example here you can see the meta title meta description blog interaction and social sharing um images um okay so you can produce you are referring to yeah at the end of the sorry at the end of the talk i will i will go over the no code tools that i have used uh as a quick recap and explain how we use each of them the fourth step was uh the fourth step of the process was the creation of the

14:59articles when scripting this talk at first i was going to explain this step from a perspective considering like how workflow works and using workflows terminology because we have used workflow to do all these uh steps that i will show you now but i realized that it was quite complex to understand particularly for those who haven't used webflow before that's why i decided to instead make use of some graphics and some sql or databases language to better illustrate the point the following explanation is not exactly how workflow works but we'll probably leave many more to an understanding of what we did and an easier option of the

15:45strategy to other local tools that are workflow first imagine a database where we uploaded all of the information that we had on the earth table and google sheets this database is divided into five tables one for each of the content types that we had so here's the database you can see that information from airtable and worksheets was was uploaded on this database and this database contains the five tables right here each of them are referring to one of our content types now stay with the company pizzeg content type that is at the bottom at the top sorry um because i will that that concept type require a special treatment than the

16:31other ones so i will talk about this one first so yeah that table corresponds to the company content type and it contains all the information on their table only information on their table each column on our table has a field to complete on this table you can see here that for example the column of slasher link has this um has this column the second column is the selection link and we have all the same information there and the information that we had on our table and each row of this article has an item or an entry in this table of the database each entry of this table has its own

17:12page which is what i earlier on called an individual pitch tech page where we show all the information related to the startup and its speech deck for example here is airbnb example once again you can see that for example here it says in the image it says airbnb and that information is um gotten from the database from the column startup name then in the case of the presentation this is an embeddable and it is it is embedding the slideshow that it is extracting from this slideshow link uh which is the second column of the of the database and then website url is the third column and uh we use it at the

17:57top at the bottom of the of the page to indicate the website of the startup so this way by uploading the airtable information we automatically created 352 pages that soon start to rank for keywords of the style company pitch tech such as airbnb bridgestep you can see here that we are ranking for this keyword this is failure is a page however these keywords don't have a huge monthly search volume in fact few people search for a specific big sec of a company they are more interested in pitch decks from one industry or pitch text from one business model so we need to uh we wanted to create listicle

18:42articles uh focus on the different business models and industries in order to increase the number of keywords that we rank for and the the monthly search volume of these keywords so uh going back to the um the graphic that i showed you before i have already covered the first time table of company spec now i will talk about these other four tables uh which uh are represent the type the content types of industries business models customer types and stages these tables contain all the information on the worksheet you can see here that for example this screenshot of the google sheet is only for the industries you can see that here we have all the

19:25industries uh that are acted drawn got edged right sharing cryptocurrencies each of these rows was an entry in the database in this table of industry startups and each column on the google sheets had a column on the table of industry standard page sex so yeah each of each of each entry of this table also has its own page but these pages were empty the they just contain like a title uh and a blog interaction you can see for example the meta title here then another column was uploading interaction um and they only had that information they didn't have any any pitch text on this on this database

20:09so the way that we filled these pages with corresponding bridge text was by relating each pixel on the first table the one of the company mistakes with one or more entries on each of these four tables this is the key slide of this presentation because it is what shows how we created a lot of articles without having to manually create each of them but instead by connecting information between the databases you can see here for example this is the company page table and one of the entries was for coinbase and two of the columns that we had in the case of the company pitch tech table

20:48were industry and customer type in the case of industry this cryptocurrency thing was related to the cryptocurrency item in the industry table and this b2c thing was connected with the p2c entry in the customer type table this i know it is a complex uh it is complex to understand this part but if you get the idea that we didn't manually create the articles um one by one by pasting the information for each a price tag and the startup but we did it automatically by connecting information within the databases that is completely fine and i want you to stay with that point only um so yeah um

21:35we in this way we filled um we automatically created a 77 listing articles with numerous speed sets each one here you can see the listicle article for cryptocurrencies uh you can see here in the table of content in the the the site of the blog cover image that there are five startups in this case which are well coinbase gold coin glue etc and then uh in their image you can see how we present the information we have the slicer then a description of the startup then some details of the startup then some details of the bitsec and the second slicer over here we faced various limitations when

22:19setting this up on workflow but we eventually managed to build it so if you want me just to to show you how our webflow cms settings uh look like i uh shoot me an email and i can show you but uh you need to know that it was super specific to our requirements in this specific case which was also why i thought that it didn't make a lot of sense to explain this fourth step uh from a point of view of workflow and our things uh well then we finally built a hoof of all the pitch decks uh the users here can find the individual pick pages using

22:58filters these filters we build them using a no code tool that is called chatboost that only works for workflow and it allows us to build dynamic filters and uh users can also find here the holistical articles so this is how we have programmatically created 352 pitch decks uh individual pitch set pages and a 77 pitch tech listicles all of which are nowadays ranking on google and bringing consistent traffic every month in the last three months these pages have brought an average of 381 clicks organic users per day or clicks per day and they have received over 600 000 impressions if you have a one percent conversion

23:47rate this means more than three users new customers per day or more than 100 per month most of the traffic comes from listicles but there are some individual pitch tech pages for example the airbnb one that are bringing some good amount of organic users besides that as you can see here the traffic is well distributed between pages and between queries and the country that is bringing the majority of the users is united states which is another opposing point because we are currently mostly focused on the us so some of you might be thinking that this programmatic seo strategy is only applicable to a content business like failure but there

24:32are a lot of businesses with different kind of business models and there are applying similar strategies first example is the live chat the live chat is a customer support software as a service or shopify and using webflow they have programmatically created more than 300 articles on the best shopify apps in different categories such as accounting and affiliate here in the first image you can see they have where they have all of the articles listing shopify apps and in the bottom image you can see the article focus on accounting apps for shopify this brings to our website shopify business owners some of who might be also looking for a customer support app

25:17and might therefore convert into the live chat customers mainly through this top banner that that all of their articles have where they talk a little bit about the live chat and show how the product looks like sapir is a well-known startup using this programmatic seo strategy you probably know what sofia is about it is connecting apps and automating workflows and they have created thousands of uh thousands um many hundred thousands probably of pages uh where they show their existing possible connections between two tools for example if i search on google webflow convertkit the third result is a sapir programmatically created page showing the existing connections between

26:07convertkit and workflow really positive point about this strategy employed by zapier is that it brings really target users to our website because the the people who are searching things like converkit webflow are people who have already shown a really clear interest in connecting these two applications because i i can't really think of other reasons why we'll you will search two tools with a more than a some interest in connecting those two tools so this is a really positive point about this appear programmatic seo strategy and then um you may know peter levels from from twitter he has various businesses and he he relies a lot on

26:52programmatic seo to get organic traffic for all of his businesses but particularly for nomad least which is an online community for digital nomads on the one side no at least has hundreds of spaces displaying cities that fit certain characteristics for example you can see here this page that is focused on latin american cities with fast internet these are the type of keywords that are not digital nomads might search on google when evaluating where to travel next on the second side normal list has various specific barrier various spaces for each city with a specific information about these cities such as the cost of living in buenos aires and the weather and website

27:39is these are keywords that digital nomads might search uh on google before traveling to a destination that they have already defined you can see how both of these uh types of programmatically created content that no matter that normal list has they are targeted to digital nomads who have a high a chance of being interested in nomad list subscription so yeah as you can see programmatic seo is applicable to all kinds of businesses with any type of customer so that was all for this talk i hope that you learned how to programmatically create content that runs on google and brings organic organic and target users and as a quick recap i wanted to

28:26to to go through the local tools that i have used in the first side i have used workflow to build and host websites then i use our table and google sheets to for the databases then a simple scrapper to scrap online stuff and the fourth tool was placid to programmatically create images and the fifth and last tool was shedboost to build dynamic dynamic filters on so yeah that was all uh thanks uh so much for listening um i guess that we can do now some question time yes indeed sir thanks for that we do have some good questions from the audience we had a couple around the

29:11tools that you're using but you just you just answered that with that last slide um spencer uh oh no the question right before that how do you come up with ideas for your seo campaigns okay um so there are basically and seeing the examples of um that i have just shared i basically identified two uh really clear uh ways of coming up with programmatic seo ideas the first uh kind of strategy is like focusing on the business and the and your solution and that is exactly what sapir is doing uh you can see that they have created thousands of pages really focus on their solution showing how

29:54sapir can help you connect two applications so that is uh that is something that i mentioned that was super positive about these pages because they are getting really target users and the second strategy that i identified is thinking what kind of uh things might uh your target customers search on google which aren't strictly related to your product in the case of the live chat for example they are a customer support sas for shopify and they haven't created any uh programmatically created pages for customer support but they have created programmatically pages created places for shopify apps which is something that many shopify business owners might search on google

30:41and um and only when they come across delighted they might be interested in uh such a customer app but they weren't initially looking for to pay for a tool like the item so those are the basically the two kind of approach that i have seen the most uh when coming up with this programmatic seo ideas very good our next question is from spencer and he says did you do anything to advertise the article advertise or promote the articles when released or did you just let google find them on its own no at the end uh it is a good idea to promote them if you have some budget to

31:23do that but no within we have been publishing content for the last three years so we are quite we have some of authority um in google we have a big domain and page authority on both pages so it wasn't hard to run for these keywords and know that these are really a specific long-tail keywords so they have a low keyword difficulty so so yeah it wasn't really hard to to to rank well on google despite we didn't do any promotion or any interlinking um efforts there very good next question is from esprit and she says how do you monetize failurey yeah that's a good question we monetize

32:09it through sponsorships so growing our traffic is strictly related to a higher amount of revenue we also make money for digital products we have a a new book about product market fee and we have recently launched an online course about validation of ideas focus on pre-sales and we also make money through affiliate commissions we promote products and courses and we keep a commission if we we make a sale so those are the three main uh sources of income for failure matt says great job nicholas thank you for the presentation could you explain a bit about how you created those pages in webflow yeah yeah i can um i it is quite complex

32:56so i wasn't i'm i was trying to avoid getting too much into maybe on that summary yeah yeah so weblog has this um it doesn't have like a database but it has what is called a cms connection and each pitch deck we have a cns collection for pitch text and there we place all the pitch text that we have on our table we upload each each row on our table is a needle on this cms collection a cms item they call it and we have another cms collection uh for the articles of the speech text the listicle articles and uh there we uploaded all the industries business small customers

33:38types and stages that we are the 77 topics that we have [Music] on our table and um we used a future that webflow has that is called multi-reference field to connect one cms collection with the other one and uh then in the in the items of the of the business small industries customer types and stages pages we used uh an item that i can't remember how it is called but basically we called we we yeah we call all of the items from the pizztech cms collection that i had for example on industry a link to that specific topic for example in the case of cryptocurrency uh there were five

34:31startups that in the field of industry in the multi-reference field of industry had a connection with the cryptocurrency them so in the cryptocurrency page we had an um we call all of those pitch texts that were uh connected to the cryptocurrency article it is quite hard to explain but if anyone wants to see the the behind-the-scenes of the cms settings i kind of i can record a video and send them to anyone very cool um and matt has another question and you may have answered this but he says which no code tool did you use for content creation so uh in the case of the content yeah the

35:14content creation was um made from from the data that we had on our table um so we had like the whole layout of the article and there were some empty spaces that were filled with information for each uh startup and pizza so we didn't really need to have to do any a content writing work besides rewriting the descriptions of the startups for that we used a tool that is called wilbot that i didn't enter a put it there because it is not a no code tool it is an ai writing tool it is called wilbot and then we use another ar writing tool for the blog interactions which is a

35:59copy ai which is well known um and yeah those were the writing tools that we used very cool thank you sir [Music] you