Innovation through 20% Time and The R&D Gauntlet

By Hayden on 63

Innovation through 20% Time and The R&D Gauntlet
4 votes, 5.00 avg. tacos (93% full)

As our mission states, being on the cutting edge of online marketing is one of our core goals.  Innovation has to be worked into  the company’s blood, which is why, like Google, Facebook, 3M and others, we’re implementing an R&D program.  The program’s goal is to test out new ways of making money online, to experiment with new and existing forms of traffic generation, monetization and conversion.  For some this will mean testing out exactly how certain parts of Google or Facebook work, for others it will mean testing a business model that NoHatDigital is yet to try.

 

Here are some details:

  • Everyone in the core team of 13 has has up to 20% of their time to work on an experiment of their choice.
  • The experiment must have a clearly defined hypothesis, with clearly defined success and failure criterium.
  • The experiment should have a high likelihood to succeed or fail within 45 days.
  • The experiment must be approved by at least 3 people in the Founding Team.
  • Previous interns are invited to assist with the experiments in exchange for exclusive access to tools.

 

The best performing experiments will be incubated and act as part of a future internship so we can scale the testing.  Like all our internships, we will have at least one of the week’s devoted to testing an experiment at scale.

 

is-such-a-thing-even-possible

 

So, what would you like to test?  Leave a comment below!  If any of the NoHatters like a particular test, they may take charge of it.  We will follow up later this week with info on Nate’s test, as well as a reader poll taking some of the best ideas you volunteer here, and then case studying the winning test.

The NoHatDigital Internship.

Learn the Exact Skills and Processes the NoHatDigital Team use to Dominate Online.

Innovation through 20% Time and The R&D Gauntlet
4 votes, 5.00 avg. tacos (93% full)

63 thoughts on “Innovation through 20% Time and The R&D Gauntlet”

  1. Armando Saenz says:

    Thanks for the post. I’d like to see how long it takes you to rank a local site with just PBN links, How long with just social links, and how long with just no follow links. Proper on page of course.

  2. Zac says:

    A test I’ve been curious about is using a country specific TLD and hosting from that country to see if it gets any local boost.

    So for example, a .ca EMD/PMD local site with Canadian hosting versus the same domain, another TLD, on a non-Canadian host.

    • GregNunan says:

      Hi Zac, we usually do not go with country specific TLDs, largely because we’ve found that you can rank sites in foreign markets with .coms and some TLDs require certain registration requirements which lessens the pool of potential buyers down the track.

  3. Jeff M says:

    I think you might have mentioned this in a video, but I’d like to see a case study done on Youtube videos. What keywords trigger videos, what niches trigger videos, scaling it out for cities or zipcodes, etc.. and most importantly, how to monetize it.

    Also, I’ve thrown away dozens of “free $100 adwords” coupons. I’m by no means an expert on adsense, but would it be possible to inflate the bids by using hundreds of these coupons for a niche at 10x the suggested bid? Combine this with something viral, scraping/email blasting, etc…

    • GregNunan says:

      Interesting tactic for inflating the bids. That’s not where we want to be heading, but interesting none-the-less! Ranking YT videos, yes certainly something for us to consider thank you.

  4. Taylor says:

    I would love to see a case study on one-page sites (~7,500 words) vs siloed, multiple page sites. Also from my own testing, the ultra-performing sites (98+/100 Google Page Load Insights, and 0.4 second load times) are really dominating compared to normal sites, but I would love to see some comparisons on that from others. From what I’ve found, those sites are taking roughly 1/4 of the links needed to rank.

      • Taylor says:

        The main high-performance site I was testing on is stuck in the top 3 in a foreign financial niche, has been for months with only about 3-4 solid PBN links built at the start, and only has a homepage with 1700 words (and a tiny interior page that does nothing with 500 words). Would love to see others have success with it too!

  5. Kapil says:

    Hi Hayden and Team, I have been a regular follower of your blog from a long time now. This is an amazing concept that you have come up with. Since SEO and Internet Marketing is a very wide field, a lot of ideas cross our minds daily but we do not have the expertise/capability to bring them into existence and they get lost somehow. With this initiative of yours, at least some of the innovative ideas will get highlighted and who knows which one makes millions of dollars. So here are a few which I really wanted to test out:

    1. I’d like to explore in the direction of creating a SAAS based model which helps enterprises or local businesses in a certain niche to minimize or simplify their daily tasks. For example, if we create a SAAS based SEO Monitoring and Reporting tool for SEO agencies and clients which will help them keep a record for every website for which they are doing SEO in terms of each link added, date added, the progress in terms of PA, DA, CF, TF and an overall progress graphs for the site. This can also be integrated with the webmaster tools and by using that data, we can graphically or pictorially represent the exact amount of links discovered by google and their vital stats such as their authority quotient, their anchor text % which is not accurately shown by the tools like ahrefs and majestic. So overall, this product will help the SEO agencies to keep all the data in one place for a site as well as help their clients to see what sort of progress is being made on their site and what links are being built or simple get the reports from time to time. Small things like rank changes for crucial sites can be monitored through this system and then an automated message/email can be generated to the SEO managers/clients regarding this information.

    2. I have seen a lot of internal pages of authority sites ranking with 2-3 backlinks for even tough keywords. For example, I am targeting a 8k per month search term and I see one internal page of a site with 3 backlinks standing number one for it. Maybe you could exploit this partiality that google is doing in favor of authority sites and buy or rent a page in an authority domain in a specific niche. Then put an article focusing a tough keyword and see how it ranks. If it ranks well, the affiliate commissions will be a good way to earn from that page. I am not talking about the generic press release sites here because they have stopped ranking now. The idea is to rent or buy a page from an authority site in a specific niche.

    3. I also wanted to test out dropshipping vs the amazon commissions. I believe that amazon’s trust factor motivates people to buy from them but on the other hand, dropshipping has good margins. So will the people buy from a non brand site as compared to amazon. The answer obviously is no but I wanted to test how are the commissions and profit margins different and what is the best way to move forward in future.
    Also, it would be good to explore if a dropshipping product can be created and then shipped. If such a product can be created, it can also be sold on amazon and then those amazon pages can also be SEOed to increase sales.

    4. Facebook Ads – I have seen a product on clickbank which is called Manifestation Miracle. Its about manifestation as the name suggests. I believe that it would apply to all the audience who have read the book The Secret since I guess its pretty much about the same concept. The facebook group of The Secret is really big. So the idea is to get all the facebook UID’s of the audience that is in The Secret group and related groups and show them the ads for this product. We can target people of ages 35 to 65 since they have the most buying capacity.
    Also, since facebook doesnt allow to take people to clickbank offers directly, we can buy a domain which will open an iframe for our clickbank link which eventually will open the affiliate site on our site’s url. So the person will be directed to our site but he will see the affiliate site inside.
    This technique of ads and offers can be generalized and applied to a lot of offers on clickbank.

    5. In the course of ranking our sites in google, we almost forget that yahoo and bing also get a lot of searches on a daily basis and their algorithms are too old. Why not exploit them the way we used to do seo 5 years back and rank some tough keywords by spamming the sites with automated software like GSA and see how it goes.

    6. I am not seeing any videos ranking for tough keywords these days. Youtube’s algorithm has changed and so has google’s. Maybe you could try some ways to rank videos again. I’d like to experiment with niche relevant backlinks for a video ranking, maybe they will add a relevancy factor in videos too.
    Apologies for the big comment but there were a lot of ideas in my mind to explore.

    Thanks a lot.
    Kapil

  6. Jim says:

    I would like to see a test where you identify a low competition keyword, register 10 PMD’s, host each site on a different IP, and… see if you can Rank “ALL 10” of the Top 10 positions for a Goggle search for that keyword.

    Why?

    Because if the goal of a niche site is to “get to Page 1” for a low competition (well paying) keyword… if worth doing… why not overdo… and rank for all 10 of the Top 10 and TRULY OWN that keyword.

    Your thoughts?

    -Jim

    ps: I’m somewhat competitive, so would love to see if the NoHatSEO team can not only just grab a parking lot, but the entire city! 92% of all clicks are on Page 1… why settle for just Ranking #1 (35%)… why not develop a game changing model and TRULY OWN IT!

    • Hayden says:

      I’ve actually done this, with my old furniture business I held all of the top 5 positions, and 8 of the top 10. You can’t really do it with just PMDs anymore, as Google likes to diversify the SERPs. You could however buy a presell page on an authority site in the niche and link build that, write a wikipedia page if one doesn’t exist, create and link-build a youtube video and set up a couple of your own sites, while renting ads on other sites that already rank…

  7. Sam says:

    Hello,
    I would like to begin by saying that, nohatseo (hayden and team) teachings has helped me in taking my seo skills to a new level.. I have had the honor and privilege to complete the june internship successfully..

    As it is getting harder and harder to find good expired domains with a DA above 25 with the xenu method, we need to find a different way to get them ..I have yet to come across a domain DA >30 which is not spammed..

    My suggestion would be to try to find all DA > 30 which are not available to buy but are dead sites, scrape their emails using a bulk whois scraper and send them a general email that we are ready to buy their domain for x$..

    It would be a gr8 idea to test them out and if it works then sky is the limit..

    Thanks
    Regards
    Sam…

  8. NicheGirl says:

    I’m pretty new at SEO, so a bit daunting to add to the great ideas mentioned above.
    What does spring to mind immediately though is more of a conversion subject.
    I saw a mention of targeting traffic for non-us sites with a .com extension. And that got me thinking :)

    Sites can be structured very differently for different countries. A chinese site for instance is more layered and busy than a dutch site, which tends to be more structured in columns and less cluttered.

    There are some hypotheses that state that the conversion can depend on the way these cultural differences are applied to web design. One of the studies that you can use to determine which structure and content converts better is by Geert Hofstede. I would be interested to see if that can be proven bij having a split test for two regions, based on the cultural dimensions that Hofstede describes here: http://bit.ly/CulturalDimensions

    So, not really SEO, but interesting!

    • GregNunan says:

      Hey NicheGirl, that’s some deep stuff! We haven’t gone to the extent of testing conversions with different site layouts, but from a ranking perspective we build all our sites the same way and see no difference in rankings.

  9. Gary Kirwan says:

    Hi guys,

    My experiment test is from this article https://affiliate-program.amazon.co.uk/gp/associates/promo/a20140811

    The idea is that if you implement the time event tracker code in a site using google analytics that it could dramatically improve the bounce rate. Apparently this in turn would make your site “more interesting for advertisers like Google AdSense or affiliate networks”. So could that mean better CPC earnings for adsense sites.

  10. Julien says:

    Hey guys,

    I’ve been flirting with the idea of “knowing” how many links should be required to rank a mid-range competitive/local keyword.

    We could try to create a matrix based on the KW difficulty, PA/DA of the Top ten (Maybe Trust/Citation flow as well) on about a 100 of keywords to see if we could accurately “predict” what’s needed in term of links to achieve top 5 rankings.

    It’b be amazing to be able to says that: with a KW diff. of +-30, we usually need 20 links from 20 domains between DA 25-30 to be able to rank. Since you guys have a lot of data, maybe you’re best suited for this type of analysis?

    Cheers,
    Julien

    • GregNunan says:

      Julien, we actually have a metric for this that we call Thirst™. We use it to prioritize which keywords to build sites around and which keywords deserve our link building attention. We teach this in our Internships and Private Training Courses.

  11. chris kent says:

    Hey Guys,

    My experiment is based on monetizing the visitors that sites are already getting more effectively. This would be by using autoresponders.

    The basic overview is to collect email addresses on each individual post & then dropping the person into an auto responder that directs people to several different pages on the site, increasing the likelihood that they would click more than one ad. My hypothesis is that this would increase the ad revenue of the site, but also the overall value because now you have an email list & a higher revenue per visitor.

    Thanks guys & I hope you like it!
    -Chris

  12. Goran says:

    What I’m really interested in is to see how a localized PBN would affect local rankings of a global site.

    The experiment would start with a site that isn’t geo-targeted. Let’s say our homepage main keywords are rowing machine reviews and best rowing machine, which you guys here seem to love to use as examples. Usually the target of such sites is primarily US, all the rest, like UK and AU, is more or less a side effect. For one of my sites more than 50% of its earnings have come from outside US – CA and Europe with UK on top. But I also get some sales from Germany. My theory is that with a local (in this example German) PBN would help me get a much higher ranking in that country and also stabilize it there for much less effort and at a lower cost than just expanding a primarily US based PBN.

    So in short, test of several smaller local PBNs for a global site vs. one global PBN in local rankings. The more cost effective (ROI) solution would win.

    • GregNunan says:

      We’ve not documented this or case studied it in detail, but local links certainly help. Would make a very useful case study for sure.

  13. Ted says:

    Hi guys,

    I’m just wondering if you’ve ever done any experiments around site load times and how they affect not only bounce rates but also where you appear in the SERPs?
    Also, I know that well over half of my traffic comes from mobile devices. Is there a way to optimize sites strictly for mobile users to see if it makes any difference in traffic or bounce rates?
    I have one site that I used a popup form on entry, and no one was really interested in it anyway. When I removed it the time on site went up a bit. I know some forms can be difficult to close on a mobile device.

    Just wondering…

    Thanks!
    Ted

  14. Mohit says:

    I would be interested to see if a test could be done where we have a three way fight between
    a) Expired Domain
    b) New Domain
    c) Parasite Pages (Already high PR blogs like Storify, Medium, Pinterest, Facebook, Youtube etc.)

    The content on all 3 of these should target same KW and same kind of backlinks are sent to each of them to finally see which type wins the battle of the serps. It will be interesting to see which kind ranks faster and higher in Top 10 Serps? To make it interesting we can choose multiple niches and provide results for each niche as results might vary depending on niche we choose.

  15. Kevin says:

    Apps anyone? Big market & Big money to be made – and i’m sure its been mentioned before at NoHat Hq but obviously not the ideal module for the internship course due to capital restrictions. Nevertheless its something that doesn’t rely on seo and just a little on the viral aspect. What do you think?

  16. Steve wyman says:

    Hi

    PR Train – the idea being to redirect expired domains into serial config with domain 1 301’d to number 2, 2 to 3 and then 3 to money site. Using lower da cheap sites.

    Some juice is lost in the train but would it build up to create a greater than its parts effect?

    Needles bulk testing rather than just a few experiments

  17. Garrick Dee says:

    Lots of awesome ideas here, I’m interested in finding out how to improve conversions in an Amazon affiliate site (also an Adsense site). I’ve been tweaking my sites a bit but I’ve not found an exact solution for it as results have been mixed so it could be a good case study. I also understand that conversions will be niche dependent but hopefully you’ll consider this.

    If you’re site getting lots of traffic, improving it even by just 1% can do wonders :)

  19. Mark Jenner says:

    The whole world of SEO is mad on PBNs right now, with a lot of experimental data showing they work…especially from your good selves.

    Myself and a friend were having a chat some time back about the whole concept and a couple of interesting thoughts came up.

    I find it odd that a company the size of Google with the computational clout and intellectual minds at their disposal aren’t able to ‘reset’ the linking power of a domain once it has expired / been dropped.

    Yes, some sites may unintentionally expire and then be re-instated by the owner. Or perhaps a webmaster lets a site expire, rebuilds and re-brands on the same domain with new content. So the same company or service may be using the same domain and in this case the authority (and hence linking power) should remain.

    But in the vast majority of cases when a domain expires, it now pops up with entirely new content, a new ‘voice’, and is very easy to distinguish from the old as a ‘completely new site’.

    So my thinking is…and I’ve read others have this point of view too…surely Google can quite easily tag that site in it’s algorithm as ‘starting from new’ and all links pointing to it and any authority it has are basically tagged as being zero and are discounted. This wouldn’t be at all hard to do.

    Google can quite easily ‘show a PR’ while the algorithm tags the domain as zero to fool the greater world. And tools such as Moz, Majestic etc. wouldn’t know what Google does internally and will still seek out and find the links pointing to a dropped domain and show the same metrics (DA, PA, trust flow or whatever) as if that domain was never dropped. So Google can internally adapt to the drop with other tools still showing great authority.

    So maybe the climbs in serps a money site has when linked to from dropped domains isn’t in fact coming from believed authority of dropped domains, but is instead merely from the relevancy, volume of links etc. that are coming from such domains.

    Maybe 20 links from dropped domains is just the same as 20 links from new domains. Maybe putting up the ‘best page in the world’ for a particular keyword and all other on page SEO metrics is enough to rank the money site when 20 links are thrown at it as long as those 20 links have relevancy. Regardless of if they come from a dropped domain PBN or brand new sites.

    So my experiment idea: Make a money site targeting a low-med comp keyword and instead of throwing lots of dropped domain links at it from a PBN, see if a site can be ranked with a similar number of links from NEW domains.

    I mean, instead of buying up and building 20 dropped domains to link to a site, try creating 20 brand new sites and link those to a new money site and see if it ranks.

    But to make it a fair test, build two VERY similar sites, making them as close to one another as possible, and link build one with dropped domains and link build the other with new domains. See if there’s much difference.

    Will a money site keyword with THE BEST content targeting it and perfect on page SEO + links from brand new domains rank as well as a money site keyword with THE BEST content targeting it and perfect on page SEO + links from dropped domains.

    I’ve never seen anybody tackle this concept before.

    If nothing else, if there isn’t something in the Google algo to consider dropped domain PBNS currently, the one linked to from new domains would certainly be more ‘future proof’ if Google ever does do something to tackle dropped domain PBNs.

    Thoughts?

    • Nate Tsang says:

      Hey Mark,

      Thanks for the well thought out comment, lots of interesting thoughts and you make some good points!

      But about new domains having the same effect as dropped domains with a strong backlink profile, I’m confident that link juice is still a key factor in the strength of PBNs, there are just too many data points indicating that is the case to believe otherwise.

      I agree 100% that we should be looking to future proof our SEO tactics, but I don’t think linking from fresh domains with no backlinks would be the way to do it.

      However, I would definitely be interested in running (or seeing the results of) a modified version of your experiment that looked specifically at the effect of relevancy on linking power.

  20. Alan says:

    Hello,
    I have been a regular follower of your blog for over a year now and I just wanted to say thank you for sharing so much incredible and actionable information for building niche sites.

    I was recently accepted into the September Internship, so I am looking forward to learning how to take my niche site portfolio to the next level.

    I would like to see an experiment on building multiple niche sites using sub-domains and see if you can rank near the top on Page 1. If you use a brandable domain name, then the site could eventually turn into an authority site.

    Thanks,
    Alan

  21. Bryce says:

    I want to see an experiment which tests what is the optimal number of words per page for a given budget. This will be tested in relation to Google Traffic and social media traffic. Let’s say I have budget that allows me to publish 10,000 words per week on an authority magazine style website. Should I publish one 10,000 word article per week or twenty 500 word articles? Ideally there will be a broad range of words per post tested. Here are some suggested configurations to test:

    Forty 250 word articles
    Twenty 500 word articles
    Ten 1000 word articles
    Seven 1,428 word articles
    Five 2,000 word articles
    Three 3,333 word articles
    Two 5,000 word articles
    One 10,000 word article

    1) Which configuration produces the most traffic from Google?
    2) Which configuration produces the most traffic from social media?
    3) Which configuration produces the most total traffic (when adding 1 + 2)?
    4) I want to know if the optimal number of words per article is the same for social media and Google traffic.

    I think the NoHat group has done something similar to this in regards to optimizing Google traffic and used RPM. I want to test social media versus Google traffic in a head to head competition. I am interested to know if there is a big difference in the optimal configuration for Google traffic versus social media traffic. Now that NoHatDigital is foucsed on traffic from all sources not just SEO this seems like a worthwhile experiment.

    To control for as many confounding variables, existing sites with similar traffic and rankings in similar niches should be used.

  22. Bryce says:

    PBN Link Relevancy – I have a website that I developed with a lot of well written content but very few links. The site sat stagnant for a long time (thanks sandbox!) and I pretty much gave up on it until it started ranking. Everything was done with quality in mind; on page SEO, PBN links and content. The site had only 3 PBN links and 10 pages of content but after about 6 months it ranked in the low 20s for two competitive key words with about 7500 searches per month each. The PBN links were very relevant.

    Therefore I want to know can just a couple of good ultra relevant links replace 25 to 50 unrelated PBN links? There should be some way to score the relevancy of the PBN site to the money site. Perhaps you can have some interns blindly rank how related the PBN sites are to the money sites and then use this information to conduct the experiment? Of course the PA/DA numbers for the PBN sites need to be similar for the results to be valid.

    This experiment can be done on existing sites where you want to give the money site a boost from page 3 to page 1.

  23. Ark says:

    I’d like to see the difference in ranking for sites developed with different platforms (wordpress, joomla, drupal, ghost, static HTML, etc.) and different hosting services, assuming their backlinks and content use the same methods.Then, compare them to sites developed the same way but hosted on Google’s app engine. It would be interesting to see what difference if any platforms, hosting services and Google’s app engine can make in a sites ranking.

  24. Jason says:

    I would like to see an experiment on recovering from negative seo which seems to be more and more commonplace. Some facets of the experiment could include:
    1. How much the authority (a DA 30 vs a DA 50, for example) of the page lends itself to a bit of resistance to the effects, if at all.
    2. How long does it take for the negative seo to affect rankings?
    3. How well does G’s disavow work? How long does it take to work? Does the site fully recover?
    4. It could even include spamming web 2.0 sites which link to the test target. How do these links affect rank? If negative, how long does it take for the ‘money site’ to recover once the spammed 2.0 links are removed?

    • Roman B says:

      I’ve recovered a site several months ago from a negative SEO attack.

      The site was around 50 DA
      Took about 2-3 months before the neg seo screwed it however*
      Disavow did absolutely nothing

      There are many different ways to negative SEO someone. In particular I recovered the site utilizing on-page adjustments to negate the Exact match anchor he pushed at me.

  25. Roman B says:

    I would like to see specifically a SEO experiment utilizing Keyword Density.

    In particular I would like to see how Keyword Density relates with anchor text. I know for a fact that google has an algorithm that multiplies KW Density with Anchor text.

    I would also like to figure out what ideal Keyword Density is on-page. I have an idea on where to start but I’ve seen cases where reducing Keyword Density has increased rankings.

    I would also like to see if Keyword Density in articles linking from a PBN have a measurable effect on the sites they are linking to.

  26. Jack says:

    I’d love to see you go after google images. My hypothesis is that it’s easier and quicker to rank in google images than it would be to rank for the same or similar terms in a standard google search.

    Maybe you could have a set budget for SEO for each group (images and normal search), then see which has a greater ROI. I assume that images will produce lower CTR but possibly higher page views because you’ll be able to rank higher much quicker. I also think because you can see a lot more results on the first page of google images, it would be easier to access the large jump in traffic that being on the first page offers.

  27. TED LAMBERT says:

    I would like to see a test between curated vs article builder type articles vs plr and what helps the rankings fastest. Or are these all bad ideas for adding content to a site or do they have some use as backlink products. Ted

  28. Eric Arrington says:

    I would like to see a test of linking out to authority sites affects ranking. Also how many it takes to make a difference. Is three links to Wikipedia or government articles going to improve the onpage SEO of my pregnancy site or will a list style post of the 10 best resources be better?

  29. JanisG says:

    I would like to see an experiment proving/disproving the existence of ‘Google random ranking factor’.

    There is a theory from several SEOs (e.g. Terry Kyle, google this snippet to find his thoughts on this subject): “In order to keep SEOs off balance and their algorithm from being completely reverse engineered, Google has built in an element of natural variation and unpredictability into their search results. That’s why you often see sites with lousy SEO ranking well (assuming we’re seeing their real backlink profile which could be masked) and beating sites with much better quality SEO.”

    I am wondering what would happen if you set up 3-5 similar sites (same niche and kws, similar kw density and site structure) and then build similar PA/DA PBN links to these sites. Possible variation – compare performance of brandable and kw rich domains.

  31. Jason McCuen says:

    Hayden and Crew,

    My experiment ultimately comes down to traffic, but has to do with indexation – specifically learning more about how and when Google chooses to crawl a page… and if it can be manipulated…

    I Present for your consideration ‘The Great GoogleBot Puppeteer Experiment’

    I have always been intrigued by GoogleBot, and the way that Google indexes sites in general. I know in the past this has been thought to be based largely on PR, but I wonder if times have changed.
    Here is my experiment:

    Part 1:

    • We register a brand new domain – long gibberish is great!
    • We forward the ‘A’ Record DNS to a self-hosted server with a single IP address assigned
    • We upload a html file with a single page that has unique content.
    • We wait and see if it indexes by itself for a week or two.
    – It would probably be a good idea to check index on http:// as opposed to https:// – although I honestly don’t know if the SSL would make a difference
    • We wait again
    • If it still hasn’t indexed – we simply directly access the site via Chrome. I am pretty sure that that will get just about any site to index by itself. If that still doesn’t work – we can add a sitemap and ping Google.

    Part 2:
    Once the site has indexed – We can upload a new html site – a real behemoth!

    Basically – Think of It is one big pyramid, or even a maze – that is set up in a way in which we can track where Googlebot enters the site, and how far he crawls.
    We can set up and run different experiments to see how Googlebot reacts i.e. have a page that is not linked from any other page and bombard it with social signals and see how long it takes GoogleBot to crawl.

    We can track how deep and wide he goes, and then we can see what we can do to control where he goes and what he crawls.

    At the same time we will need to monitor exactly what is in the index and how it matches up to the pages he has crawled.

    This testing will give us a great understanding on how to quickly get pages indexed – as well as a more accurate understanding of how, why and where sites get crawled – and if it can be influenced.

    Thanks for your time!
    Jason

  32. Zach says:

    One thing I have been thinking about, similar to one of the approaches you have mentioned of building a number of small sites with 7500 word articles and waiting 6 months just to see how these will do with google (and picking the ones that seem most promising to build out). I would like to see a similar approach tried with one page coming soon pages with an email opt-in, such as those offered by lead pages (optimized for email list conversions). Apparently there has been some sucess with collecting a good number of email addresses just with a coming soon page. This would allow you to measure and compare both the google rankings and email opt ins of different topics (and would be less initial work to get each site up to measure the interest). Thanks

  33. Richard Johnson says:

    My idea is to develop experiments around the kind and quantity of content placed on a page to test on-page SEO.

    Specifically:

    1. Does an increasing amount of quality text increase the placement of an article in the SERPs? Would a 1,500 word article rank as high/quickly as a 3,000 word article?

    2. Does the addition of photos increase the placement in the SERPs? Would a 1,500 word article without photos rank as well as a 1,500 word article with one photo? Does an article with 10 photos outrank an article with just two?

    3. If photos affect SERPs, do slideshow plugins work as well/better/worse than simply adding the same photos to the article?

    4. Do lists increase the placement of an article in the SERPs? Would a 1,500 word article without a list rank as well/as fast as a 1,500 word article with a list? Do multiple lists have an increased affect on ranking?

    5. Does a site with the WordPress SEO plugin from Yoast rank better/worse/the same as a site using the All in One SEO Pack plugin?

    Thanks,

    Richard

  34. Travis says:

    My experiment is related to finding expired domains for building a PBN. My thought was can you find domains of bankrupt companies such as banks (thinking 2008 financial crisis) and find more sites in that way. I am not an expert on valuing domains so couldn’t do this quickly or efficiently, but thought you guys could tell pretty quickly whether or not it is a viable strategy.

  35. Brett says:

    Testing a 301 on a new domain.
    Build two niche sites targeting the same keyword with similar on-page and similar amounts of content. After that, find an expired domain and 301 redirect it to one of the new domains to see the ranking effect. You can either do this while you continue to do your regular PBN link building or do it without SEO at all.

    Testing the effect (Long-term or Short-term) of SAPE links
    SAPE links were a huge thing not too long ago, and after Google addressed it, people have pretty much completely given up on SAPE links. I have heard from a few people that SAPE links are just as strong as they ever were. Obviously this is more of a short-term, rank and bank type experiment, but still interesting none the less.

    I’m enrolled in the September Internship, and what you said about “Thirst” has got me excited!

  36. Pieter Le Roux says:

    An experiment that doubles your money-sites as a PBN site. Authority is brought to your money site via PBN (and other) links, so why not publish a blog or some content on your money-site that also doubles as a lower-tiered PBN? Obviously you won’t link from your money-site to other money-sites you own, but to others not on your PBN linked by your network already. I theorize that this will spread out your linking profile even more, looking even more natural and give you more bang for your buck while gaining decent rankings.

  37. Clarence says:

    I am interested in seeing how to earn money using PPC on lesser knowns sites such as bing and yahoo network,stumbleupon ppc and others.It might be a good experiement for people who want to try PPC but with lower entry level.

  38. Dan says:

    I would like to see an experiment on very high Cpc keywords (100-1000$ range).

    Sometimes those keywords have a low search volume (less than 1000 per month local search) and relatively low competition metrics. I would like to see what happens if you’re able to send good targeted traffic volume to the niche site who’s targeted the high Cpc keywords.

    Even if Google gives you only 25-30% of the Cpc value, it can be interesting to get a few clicks (a day or a week) generating 30% of let say 300$.

    Thanks.

  39. Adriaan says:

    Similar to the following suggestion:

    Bryce says:
    August 30, 2014 at 9:48 pm

    I would like to see the value of relevant niche sites vs general niche sites.
    start 2 new sites, similar content and setup
    then link out from similar pbns with the only difference being that site 1 receives links from a pbn that is in the same niche and site 2 receives links from non relevant pbn.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes:

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>