Tag Archive | "Using"

How LinkedIn is Using Machine Learning to Determine Skills

One of the more interesting reveals that Dan Francis, Senior Product Manager for LinkedIn Talent Insights, provided in a recent talk about the Talent Insights tool is how LinkedIn is using machine learning to determine skills of people. He says that there are now over 575 million members in the LinkedIn database and there are 35,000 standardized skills in LinkedIn’s skills taxonomy. The way LinkedIn is figuring out what skills a member has is via machine learning technology.

Dan Francis, Senior Product Manager, LinkedIn Talent Insights, discussed Talent Insights in a recent LinkedIn video embedded below:

LinkedIn Using Machine Learning to Determine Skills

The skills data in Talent Insights comes from a variety of sources, mainly from a member’s profile. There are over 35,000 standardized skills that we have in LinkedIn’s skills taxonomy, and the way we’re figuring out what skills a member has is using machine learning. We can identify skills that a member has that’s based on things that they explicitly added to their profile.

The other thing that we’ll do is look at the text of the profile. There’s a field of machine learning called natural language processing and we’re basically using that. It’s scanning through all the words that are on a member’s profile, and when we can determine that it’s pertaining to the member, as oppose the company or another subject, we’ll say okay, we think that this member has this skill. We also look at other attributes, like their title or the company, to make sure they actually are very likely to have that skill.

The last thing that we’ll do is look at the skills a member has and figure out what are skill relationships. So as an example, let’s say that a member has Ember, which is a type of JavaScript framework, since we know that they know Ember, they also know JavaScript. So if somebody’s running a search like that, we’ll surface them in the results. I think that the most important reason why this is helpful and the real benefit to users of the platform is when you’re searching, you want to get as accurate a view of the population as possible. What we’re trying to do is look at all the different signals that we possibly have to represent that view.  

575 Million People on LinkedIn Globally and Adding 2 Per Second

Today, LinkedIn has over 575 million members that are on the platform globally. This is actually growing at a pretty rapid clip, so we’re adding about two members per second. One of the great things about LinkedIn is that we’re actually very well represented in terms of the professional workforce globally. If you look at the top 30 economies around the world, we actually have the majority of professionals in all of those economies.

LinkedIn is the World’s Largest Aggregator of Jobs

I think there’s often a perception that most of the data’s directly from LinkedIn, stuff that’s posted on LinkedIn and job status is one notable exception to that. Plenty of companies and people will post jobs on LinkedIn, and that’s information that does get surfaced. However, we’re also the world’s largest aggregator of jobs. At this point there are over 20 million jobs that are on LinkedIn.

The way that we’re getting that information is we’re working with over 40,000 partners. These are job boards, ATS’s, and direct customer relationships. We’re collecting all of those jobs, standardizing them, and showing them on our platform. The benefit is not just for displaying the data in Talent Insights, the benefit is also when members are searching on LinkedIn.com, we’re giving them as representative a view of the job market as possible.

The post How LinkedIn is Using Machine Learning to Determine Skills appeared first on WebProNews.

WebProNews

Posted in Latest NewsComments Off

Sparrho CEO: Using Augmented Intelligence to Build Trust in Brands

Many companies are working to build authentic and trusted brands with consumers. This is especially true with pharmaceuticals, biotech, and med-tech companies. The CEO of Sparrho, Dr. Vivian Chan, says that their approach combines artificial intelligence and 400,000 Ph.D.’s to deliver scientific data to companies. This data helps companies back up their marketing messages which enables them to more effectively build that vital trust with their customers.

Dr. Vivian Chan, Sparrho CEO, recently discussed on CNBC their unique hybrid AI approach to helping companies use science and information to back up their brands messaging:

AI Enables Humans to Make Better-Informed Decisions

Artificial intelligence is really about algorithms and how we can use data that we collect to enable humans to make better-informed decisions. I not at all about having computers make decisions on behalf of humans. In a way, I think it’s machines that will be helping evolve the tasks and not actually replacing the human roles. Human roles themselves will be evolving also as the technology improves. This allows humans to have more headspace to be thinking about things that machines can’t do right now.

Machines can’t necessarily summarize a lot of pieces of contextual analysis very well yet to a 100 percent accuracy and humans are still better at making nonlinear connection points. For example, being able to say that this mathematical equation is super relevant to an agricultural problem. If we don’t have the tagging and reference and citations humans are still better at making those nonlinear new connection points than machines.

Humans are still good at coming up with the questions. If you actually pose the right question and you train the data and the algorithms you might actually get the right answer. However, you still need to have the humans to be thinking about what the questions are in order to ultimately get the answers.

It’s About Using AI as a Means to an End

I  think the angle is really thinking about using AI as a means to an end and not just the end. Ultimately, this is a hybrid approach and various different people are calling it differently. Even MIT professors are calling it a hybrid approach. We’re calling it augmented intelligence. We need to come up with a good relationship between humans and machines. Marketing is about building relationships. It’s about building relationships between brands and consumers and now how do we build that relationship digitally?

Using Science to Build an Authenticated Brand

In this digital age, consumers are a lot more tech savvy but are also information savvy. They want to know what the is science behind certain things. Even if you’re talking about CPG, consumer packaged goods, what is the science behind a shampoo product right now when it claims 98 percent prevention of hair loss? What is the real science behind that and how do we actually bring that simplified science-oriented message to the consumer? How can consumers educate themselves and make informed decisions based on the products and thereby build a stronger brand relationship?

Ultimately what we’re trying to do at Sparrow is simplify science to build trust in brands. Especially for marketing departments and brands, it’s really allowing them to have the evidence-based science and the facts because building a very authenticated brand is what is meaningful to consumers. Research says that about 71 percent of consumers immediately reject content that looks like a sales pitch. Building a relationship and having an authenticated brand and content is super important in building that relationship between brand and consumers.

Sparrho Provides Content as a Service On Demand

We’re going even wider with that by providing what we call content as a service or relevant content on demand. We then integrate that into the digital platforms or the brands. We have what we call augmented intelligence with over 16 million pieces of content that is augmented by a network of more than 400,000 monthly active PhDs in a150 countries. They curate and summarize what’s actually happening in the latest of science.

We know that in about 60 percent of pharmaceuticals, biotech, and even med-tech companies, are spending more than $ 50 million per year just in content. Content has been the major driver for a lot of their marketing. In pharmaceuticals, they’re trying to really bring that relationship that they have offline to online. It’s at the heart of this digital transformation age that we are going through. This is really helping bring that relationship online by using the right engaging content. Our goal with Sparrow is to drive more engagement and ultimately more sales.

The post Sparrho CEO: Using Augmented Intelligence to Build Trust in Brands appeared first on WebProNews.

WebProNews

Related Articles

Posted in Latest NewsComments Off

How to Stop Drowning in Data and Begin Using Your Metrics Wisely

Digital marketers have a problem: We’ve got too much data. It sounds like a ridiculous complaint coming from a data…

The post How to Stop Drowning in Data and Begin Using Your Metrics Wisely appeared first on Copyblogger.


Copyblogger

Posted in Latest NewsComments Off

Using a New Correlation Model to Predict Future Rankings with Page Authority

Posted by rjonesx.

Correlation studies have been a staple of the search engine optimization community for many years. Each time a new study is released, a chorus of naysayers seem to come magically out of the woodwork to remind us of the one thing they remember from high school statistics — that “correlation doesn’t mean causation.” They are, of course, right in their protestations and, to their credit, and unfortunate number of times it seems that those conducting the correlation studies have forgotten this simple aphorism.

We collect a search result. We then order the results based on different metrics like the number of links. Finally, we compare the orders of the original search results with those produced by the different metrics. The closer they are, the higher the correlation between the two.

That being said, correlation studies are not altogether fruitless simply because they don’t necessarily uncover causal relationships (ie: actual ranking factors). What correlation studies discover or confirm are correlates.

Correlates are simply measurements that share some relationship with the independent variable (in this case, the order of search results on a page). For example, we know that backlink counts are correlates of rank order. We also know that social shares are correlates of rank order.

Correlation studies also provide us with direction of the relationship. For example, ice cream sales are positive correlates with temperature and winter jackets are negative correlates with temperature — that is to say, when the temperature goes up, ice cream sales go up but winter jacket sales go down.

Finally, correlation studies can help us rule out proposed ranking factors. This is often overlooked, but it is an incredibly important part of correlation studies. Research that provides a negative result is often just as valuable as research that yields a positive result. We’ve been able to rule out many types of potential factors — like keyword density and the meta keywords tag — using correlation studies.

Unfortunately, the value of correlation studies tends to end there. In particular, we still want to know whether a correlate causes the rankings or is spurious. Spurious is just a fancy sounding word for “false” or “fake.” A good example of a spurious relationship would be that ice cream sales cause an increase in drownings. In reality, the heat of the summer increases both ice cream sales and people who go for a swim. That swimming can cause drownings. So while ice cream sales is a correlate of drowning, it is *spurious.* It does not cause the drowning.

How might we go about teasing out the difference between causal and spurious relationships? One thing we know is that a cause happens before its effect, which means that a causal variable should predict a future change.

An alternative model for correlation studies

I propose an alternate methodology for conducting correlation studies. Rather than measure the correlation between a factor (like links or shares) and a SERP, we can measure the correlation between a factor and changes in the SERP over time.

The process works like this:

  1. Collect a SERP on day 1
  2. Collect the link counts for each of the URLs in that SERP
  3. Look for any URLs are out of order with respect to links; for example, if position 2 has fewer links than position 3
  4. Record that anomaly
  5. Collect the same SERP in 14 days
  6. Record if the anomaly has been corrected (ie: position 3 now out-ranks position 2)
  7. Repeat across ten thousand keywords and test a variety of factors (backlinks, social shares, etc.)

So what are the benefits of this methodology? By looking at change over time, we can see whether the ranking factor (correlate) is a leading or lagging feature. A lagging feature can automatically be ruled out as causal. A leading factor has the potential to be a causal factor.

We collect a search result. We record where the search result differs from the expected predictions of a particular variable (like links or social shares). We then collect the same search result 2 weeks later to see if the search engine has corrected the out-of-order results.

Following this methodology, we tested 3 different common correlates produced by ranking factors studies: Facebook shares, number of root linking domains, and Page Authority. The first step involved collecting 10,000 SERPs from randomly selected keywords in our Keyword Explorer corpus. We then recorded Facebook Shares, Root Linking Domains, and Page Authority for every URL. We noted every example where 2 adjacent URLs (like positions 2 and 3 or 7 and 8) were flipped with respect to the expected order predicted by the correlating factor. For example, if the #2 position had 30 shares while the #3 position had 50 shares, we noted that pair. Finally, 2 weeks later, we captured the same SERPs and identified the percent of times that Google rearranged the pair of URLs to match the expected correlation. We also randomly selected pairs of URLs to get a baseline percent likelihood that any 2 adjacent URLs would switch positions. Here were the results…

The outcome

It’s important to note that it is incredibly rare to expect a leading factor to show up strongly in an analysis like this. While the experimental method is sound, it’s not as simple as a factor predicting future — it assumes that in some cases we will know about a factor before Google does. The underlying assumption is that in some cases we have seen a ranking factor (like an increase in links or social shares) before Googlebot has and that in the 2 week period, Google will catch up and correct the incorrectly ordered results. As you can expect, this is a rare occasion. However, with a sufficient number of observations, we should be able to see a statistically significant difference between lagging and leading results. However, the methodology only detects when a factor is both leading and Moz Link Explorer discovered the relevant factor before Google.

Factor Percent Corrected P-Value 95% Min 95% Max
Control 18.93% 0
Facebook Shares Controlled for PA 18.31% 0.00001 -0.6849 -0.5551
Root Linking Domains 20.58% 0.00001 0.016268 0.016732
Page Authority 20.98% 0.00001 0.026202 0.026398

Control:

In order to create a control, we randomly selected adjacent URL pairs in the first SERP collection and determined the likelihood that the second will outrank the first in the final SERP collection. Approximately 18.93% of the time the worse ranking URL would overtake the better ranking URL. By setting this control, we can determine if any of the potential correlates are leading factors – that is to say that they are potential causes of improved rankings.

Facebook Shares:

Facebook Shares performed the worst of the three tested variables. Facebook Shares actually performed worse than random (18.31% vs 18.93%), meaning that randomly selected pairs would be more likely to switch than those where shares of the second were higher than the first. This is not altogether surprising as it is the general industry consensus that social signals are lagging factors — that is to say the traffic from higher rankings drives higher social shares, not social shares drive higher rankings. Subsequently, we would expect to see the ranking change first before we would see the increase in social shares.

RLDs

Raw root linking domain counts performed substantially better than shares at ~20.5%. As I indicated before, this type of analysis is incredibly subtle because it only detects when a factor is both leading and Moz Link Explorer discovered the relevant factor before Google. Nevertheless, this result was statistically significant with a P value <0.0001 and a 95% confidence interval that RLDs will predict future ranking changes around 1.5% greater than random.

Page Authority

By far, the highest performing factor was Page Authority. At 21.5%, PA correctly predicted changes in SERPs 2.6% better than random. This is a strong indication of a leading factor, greatly outperforming social shares and outperforming the best predictive raw metric, root linking domains.This is not unsurprising. Page Authority is built to predict rankings, so we should expect that it would outperform raw metrics in identifying when a shift in rankings might occur. Now, this is not to say that Google uses Moz Page Authority to rank sites, but rather that Moz Page Authority is a relatively good approximation of whatever link metrics Google is using to determine ranking sites.

Concluding thoughts

There are so many different experimental designs we can use to help improve our research industry-wide, and this is just one of the methods that can help us tease out the differences between causal ranking factors and lagging correlates. Experimental design does not need to be elaborate and the statistics to determine reliability do not need to be cutting edge. While machine learning offers much promise for improving our predictive models, simple statistics can do the trick when we’re establishing the fundamentals.

Now, get out there and do some great research!

Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don’t have time to hunt down but want to read!


Moz Blog

More Articles

Posted in Latest NewsComments Off

Using Empathy and Connection to Craft More Powerful Content

I recently heard our friend Joanna Wiebe say something that blew my mind. I didn’t get it down word for…

The post Using Empathy and Connection to Craft More Powerful Content appeared first on Copyblogger.


Copyblogger

Posted in Latest NewsComments Off

Augmented reality artist creates sculptures using Bing search

Using the Bing Search API and a proprietary AR platform, an artist created sculptures composed entirely of dynamic search images customized in real time.



Please visit Search Engine Land for the full article.


Search Engine Land: News & Info About SEO, PPC, SEM, Search Engines & Search Marketing

Posted in Latest NewsComments Off

If you’re not using Remarketing Lists and Similar Audiences for Search, you’re leaving money on the table

Tapping into Similar Audiences in conjunction with RLSA can be a recipe for success for retail marketers. Here’s how.



Please visit Search Engine Land for the full article.


Search Engine Land: News & Info About SEO, PPC, SEM, Search Engines & Search Marketing

Posted in Latest NewsComments Off

Visualizing Time: A Project Management How-To Using Google Sheets

Posted by R0bin_L0rd

The short version of this post: Project management is a vital part of our job as marketers, but planning and visualizing projects over time is hard, so I’ve created a set of Google Sheets to make that work easier for you.

I’ve found this system helpful in a number of ways, so I’m sharing my templates here in case it’ll make your day a bit shorter. I’ll start off with a brief overview of what the sheets do, but in the latter section of this post I’ll also go into greater depth about how they work so you can change them to suit your own needs.

If you’d like to skip this post and get straight to the templates, you can access them here (but I’d recommend reading a bit about how they work first):

It’s worth mentioning: I don’t consider these sheets to be the only solution. They are a free solution that I’ve found pretty useful, but I have colleagues who swear by the likes of Smartsheet and Teamwork.

It’s also worth noting that different tools work better or worse with different styles. My aim with these sheets is to have a fairly concrete plan for the next three or four months, then a looser set of ideas for further down the line. When I’m filling out these sheets, I also focus on outcomes rather than processes – that helps cut down the time I spend updating sheets, and makes everything clearer for people to read.

The long version of this post is a lot like the short version above, but I talk more about some principles I try to stick to and how this setup fulfills them (shocker, eh?). As promised, the final section will describe how the sheets work, for anyone who runs into problems or wants to make something of their own.

Contents (for if you just want to jump to a specific section):

The 3 principles (which are about people as much as using the sheets)
An early conclusion
Appendices & instructions
How to add tasks to the list
Splitting tasks across multiple time periods
Working with the Month View tab (Planner and Stakeholder Versions)
How to make the Gantt charts work (and add categories)
How to make the Category-Filterable Forward-Facing Gantt Charts work
How to create the Stakeholder View
How to update the God’s-I Version


The principles (which are about people as much as using the sheets)

Principle 1: We shouldn’t need to store all our information in our heads.

This is a simple one — if we have to regularly understand something complex, particularly if it changes over time, that information has to be on the page. For example, if I’m trying to plan a marketing strategy and I have to constantly look at the information on the screen and then shuffle it around in my head to work out what we have time for month to month, I’m going to lose the thread and, eventually, my mind.

The Planner Version sheet aims to solve this in a few ways. First, you write all the tasks down in the Task View tab, the time period you’re completing them in is on the far left (in my example, it’s the month the task is planned for), and there are other columns like status and category — but initially, it can just be a brain dump of what needs to happen. The idea here is that when you’re first writing everything out, you don’t have to think too much about it — you can easily change the dates and add other information later.

The Month View tab takes the information in the Task List tab and reorders it by the months listed in column A of the Task View (it could be other time periods, as long as it’s consistent).

This way you can look at a time period, see how much resource is left, and read everything you currently have planned (the remaining resource calculation will also take into account recurring tasks you don’t always want to write out, like meetings).

While the Month View tab can help you focus on specific time periods, it doesn’t give you a long-term view of the plan or task dependencies, so we have the two Gantt views. The Gantt View tab contains everything from sixty days ago and into the future, as long as you haven’t just marked the task as “Later.” The Category-Filterable Gantt only focuses on things that are planned for the next six months.

As the name suggests, you can filter this second Gantt to only show specific categories (you label tasks with categories in the Task View tab). This filter is to help with broader trends that are harder to notice — for instance, if the most important part of the project is a social campaign or a site change and you don’t get to it for six months, you may need to make sure everyone is aware of that and agrees. Likewise, if you need to be showing impact but spend most of your time reporting, you may want to change your plan or make sure everyone understands why things are planned that way.

Principle 2: No one knows everything (and they shouldn’t).

If you’re working on a project where you have all the information, then one of two things is likely happening:

  1. You’ve really doubled down on that neuroticism we share
  2. You’re carrying this thing — you should just quit and start your own company selling beads* or something.

We can trust that our clients/bosses have more context than we do about wider plans and pressures. They may know more about wider strategies, that their boss tenses up every time a certain project is mentioned, or that a colleague hasn’t yet announced their resignation. While a Google Sheet is never an acceptable substitute for actual communication, our clients or bosses may also have an idea of where they want the project to go which they haven’t communicated, or which we haven’t understood.

We can also trust that people working on individual tasks have a good idea of whether things are going to be a problem — for instance, if we’re allowing far too little time for a task. We can try to be as informed as possible, but they’re still likely to know something we don’t.

Even if we disagree that certain things should be priorities or issues, having a transparent, shared plan helps us kick off difficult conversations with a shared understanding of what the plan currently is. The less everyone has to reprocess information to understand it (see Principle 1), the more likely we are to weed out problems early.

This is all well and good, but expecting someone to absorb everything about a project is likely to have the opposite effect. We need a source of data that everyone can refer to, without crowding their thoughts or our conversations with things that only we as project managers have to worry about.

That’s why we have the Stakeholder Version of our sheets. When we write everything in the Planner Version, the Planned tab is populated with just the things that are relevant for people who aren’t us (i.e, all the tasks where the status isn’t “unpitched,” “cancelled,” “forgotten,” or blank) with none of the resource or project identifier information.

We never have to fill out the Stakeholder Version sheet — it just grabs that information from the Planned tab using importrange() and creates all the same Gantt charts and monthly views — so we don’t have to worry about different plans showing different information.

*Bees?

Principle 3: I’m going to miss stuff (less is more).

I’ll be honest: I’ve spent a bunch of time in the past putting together tracking systems that I don’t check enough. I keep filling them out but I don’t spend enough time figuring out what’s needed where. If we have a Stakeholder Version which takes out the stuff that is irrelevant to other people, we need the same for us. After all, this isn’t the only thing we’re thinking about, either.

The What-in-God’s-name-have-I-missed Version (God’s-I from now on) pulls in data from all of your individual project management sheets and gives you one place to go to be reminded about all the things you’ve forgotten and messed up. It’s like dinner with your parents in a Google Sheet. You’re welcome.

The three places to check in this version are:

  1. Alerts Dashboard tab, which shows you the numbers of deadlines upcoming or missed, the work you need to budget for or brief, and how much unplanned budget you have per project, per month (where budget could just be internal people-hours, as that is still finite).
  2. Task Issues tab, which gives a filterable view of everything over the next three months (so you can dig in to the alerts you see in step one).
  3. Deadlines This Week tab so you have a quick reminder of what you need to complete soon.


An early conclusion:

Often, when I’m making a point, people tell me they hope I’ll wrap up early. This section is mainly proof of personal growth.

It’s also because everything after this is specific to using, changing, or understanding the project management sheets I’ve shared, so you need only read what follows if you’re interested in how to use the sheets or how I made them (I really do recommend dabbling with some uses of filter() and query(), particularly in conjunction with RegEx formulas).

Aside from that, I hope you find these resources useful. I’ve been getting a lot of value from them as a way to plan with people collaboratively and separate the concept of “project manager” from “person who needs to know all the things,” but I would be really interested in any thoughts you have about how to improve them or anything you think I’ve missed. Feel free to comment below!

Access the template sheets here:


Appendices & instructions

Some general notes

Quick notes on avoiding problems:

  1. Make sure that when you copy the sheets, the sharing permissions for the Planner View is email- or at least organization-based (anyone with access to the Stakeholder View will see the Planner View URL). It’s a good idea to keep the God’s-I Version permissions email-based, too.
  2. Try to follow the existing format of words and numbers as closely as possible when creating new information.
  3. If you want a new row, I’d insert a row, select the one above, copy it down into the new row, then change the information — that way, the formulas in the hidden columns should still work for you.
  4. If you want a new column, it might break one of the query() functions; once you’ve added it, have a quick look for formulas using =query() and consider changing the columns they reference that will have been affected by your change.

Quick notes on fixing problems:

Here’s a list of things to check for if you’ve changed something and it isn’t being reflected in the sheet:

  1. Go through all the tabs in the stakeholder view and unhide any hidden columns
    1. They usually just contain a formula that reformats text so our lookups work. See if any of those are missing or broken.
  2. Try copying the formulas from the row above or next to the cell that isn’t working.
  3. Try removing the =iferror portion of formulas.
    1. A lot of the cells are set up to be blank if they break. It makes it easier to read the sheet, but can make it harder to know whether something is actually empty or just looks empty.
  4. If one sheet isn’t properly pulling through data from another, look for the =importrange() formulas and make sure there is one that matches the URL of the sheet you’re trying to reference and that you’ve given permission for the formula to work — you’ll need to click a button.
    1. Check the Task View tab in the Stakeholder Version and Project URLs tab in the God’s-I Version
  5. Have you just called a task “Part 4” or similar? There is a RegEx formula which will strip that out.
  6. Have you forgotten to give a task a type? If so, the Gantt view will warn you in the Status column.

The query function

The =query() function in Google Sheets is awesome — it makes tons of things tons easier, particularly in terms of automating data manipulation. Most of what these sheets do could be achieved with =query, but I’ve often used =filter (which is also very powerful) because =filter is apparently quicker in Google Sheets and at times these sheets have a lot to process.

RegEx

You shouldn’t need to know any RegEx for this sheet, but it is useful in general. Here the RegEx is mainly used to remove the “Part #” in multi-part tasks (see below) and look for anything that matches multiple options — for instance, when selecting multiple categories in the Category-specific forward-facing Gantt tab (see below). RegEx is only used here in RegExmatch(), RegExextract(), RegExreplace(), or as part of the query function where we say “matches.”

Query/filter and isblank

A lot of the formulas in these sheets are either filter() or query() or are wrapped in =if(isblank() — that’s basically because filter and query functions can fill more cells than just the one you put the formula in. For example, they can fill a whole row, column, or sheet. That means that other cells are calculating or looking up against cells which may or may not be empty, so I’ve added the isblank() check so that the cells don’t break when there isn’t information somewhere, but as you add information you don’t have to do as much copying and pasting of formulas.

Tick boxes

The tick boxes are relatively new in Google Sheets. If you need another one, just copy it from an existing cell or select from the “Insert” menu. Where I’ve used tick boxes, I often have another formula in the sheet which filters rows based on what boxes are ticked, then creates a RegEx based on the values that have a tick next to them.

You don’t need to understand this to use the sheets, but you can see it in the rows I’ve unhidden in the Category-specific forward-facing Gantt tab of the Stakeholder Version sheet.

Quick tip — if you want to change all the boxes to ticked/unticked and don’t want to have to do so one by one, you can copy a ticked or unticked checkbox across all the other cells.

How to add tasks to the list

In the task view, the most important things to include are the task name, time period it’s planned for, cost, and type.

For ease, when creating a new task I recommend inserting a row, copying the row above into it, and then changing the information, that way you know you’re not missing any hidden formulas.

Again, don’t bother changing the Stakeholder Version. Once you’ve added the URL of the Planner Version to the =importrange() function, it will pull automatically from the Planner Version.

Splitting tasks across multiple time periods

You can put more than one thing in the time period for a task, just by separating it with “, “ (comma space). That’s because when we get the full list of months, we join all the individual cells together with “, “ then split them apart by “, “ and then dedupe the list — so multiple months in one cell are treated the same as all the other months.

=unique(transpose(split(JOIN(", ",'Task view'!A:A),", ",0)))

The cost-per-month formula in the Task List tab counts how many commas are present in the month column for that row, then divides the planned cost by that number — meaning the cost is split equally across all of the months listed.

=H2/(len(REGEXREPLACE(A2,"[^\,]*",""))+1)

If you don’t want the task to be completely equally split between different time periods, you can write “Part 1” or “Part 2” next to a task. As long as you write just “Part” and then numbers at the end of the name, that’ll be stripped out in column O of the task list tab so the different parts of a task will be combined into one record in things like the Gantt chart.

=REGEXREPLACE(B2,"Part \d+$  ","")

Working with the Month View tab (Planner and Stakeholder version)

A few key things are going on in the Month View tab. First, we’re getting all of the time periods we have listed in the Task View.

Because the months don’t always show up in the right format (meaning later filters don’t work), we then use a =text() formula in the hidden column B to make sure the months stay in the format we need.

Then, in the “deliverables” section of this tab, we use the below formula:

=if(not(isblank(A12)), iferror(TRANSPOSE(FILTER('Task view'!B:B,RegExmatch('Task view'!A:A,B12))),""),"")

What we’re doing above is checking if the “month” cell of this row is has anything in it. If there is a month there, we filter the tasks in the Task View to only those that contain that month in the text month column. Then we use the transpose() function to change our filtered tasks from a vertical list to the horizontal list we see in the sheet.

Finally, we use the below formula to filter the costs we’ve listed in the Task View tab, the same way we filtered the task names above. Then we add together all the costs for the month (plus the standing monthly costs) and subtract them from the total amount of time/hours we have to spend. That way we calculate how much we have left to play with, or if we’re running over.

=if(isblank(A12),"",((D12-SUM(FILTER('Task view'!I:I,RegExmatch('Task view'!A:A,B12))))-sum($  D$  6:$  F$  8)))

We also pull this value through to our God’s-I Version to see at a glance if we’ve over/under-planned.

How to make the Gantt charts work (and add categories)

Column C in the Task View tab is the category; you also need to fill this out for the Gantt charts to work. I haven’t forced the kind of categories you have to use because each project is different, but it’s worth using consistent categories (down to the capital letter) because we deduplicate the task categories, and that relies on all of the names being consistent.

What’s happening in the Gantt chart is each cell is a combination of a filter and vlookup (the below looks more complicated than it is).

=iferror(if(not(or(isblank($  D6),ISBLANK(F$  1))),vlookup(filter('Task view'!$  C:$  C,'Task view'!$  O:$  O=$  D5,REGEXMATCH('Task view'!$  A:$  A,F$  2)),'Status and colour code'!$  C:$  E,3,0),""),"")

The formula first checks if the task or month cells are blank. If not, it looks in the month cell in its column and cross-references with the task cell in its row. Where the intersection of a month and task matches a task in our Task View (as in the task in that row is taking place during the month in that column), the filter formula will return the category. For those interested, this might also have been achieved with index-match, but filter lets us match with RegEx so we can give multiple matching options and they don’t have to match exactly. Because we split tasks across multiple months, we need to be flexible in our matching.

The reason we check whether the task or month cells are empty, as mentioned above is so we can paste the above formula in all the cells of the Gantt chart and have them fill out as we add more months and tasks, rather than having to copy and paste the formula each time.

When our filter formula returns the specific category of our task, we take that value and run a vlookup in the Status and color code tab. (That’s only necessary so I could set up the conditional formatting for you so it won’t break when you change the specific category names.)

At the moment, the Gantt charts are set up to color-code the first 7 categories, plus a Deadline category if needed. If you want to add more, they’ll show up initially in the Gantt chart as a black block and you’ll need to set up conditional formatting to color-code them.

To add automatic color formatting for more categories, repeat the below process for each of the Gantt chart views in the Planner and Stakeholder sheets:

  1. Select all the cells in the tab
  2. Select “Conditional Formatting” from the Format menu
  3. Find the rule with the black box next to it and make a note of what number it’s currently targeting from
  4. Create a new rule for anything which equals the number in step 3, then set the same color for both the background and text of that rule
  5. Change the rule that’s got a black block next to it to target one number higher

How to make the Category-Filterable Gantt Charts work

This tab uses our old friends, the =filter() and =query() functions. First we use filter to grab the full list of categories from the Status and color code tab we mentioned before:

=FILTER('Gantt view'!A6:B,RegExmatch('Gantt view'!A6:A,".*[a-zA-Z].*"))

Then we put Google Sheets’ shiny new checkboxes next to them (that’ll help us filter our data easily).

Normally we’ll hide row one, but it’s visible to show you a formula that looks at all of the categories and filters them to just those where the tick-box next to them is ticked. If there are none, it returns “(\d|Deadline)” meaning “either a number, or the word Deadline” in RegEx-speak (so anything in our list), because the vertical pipe “|” means “or” and “\d” means “number.”

If there is a tick next to one or more of the categories, the formula will return those things, separated with the “|” that, again, means “or” in RegEx.

=if(countif(C3:C,True)>0,CONCATENATE("(",JOIN("|",FILTER(B3:B,C3:C=True,not(isblank(B3:B)))),")"),"(\d|Deadline)")

Then in cell E3 we have a query formula. The reason we’re using =query and not =filter here is that we need to look for things in more than one column; filter can only really handle one column at a time.

The query function then checks the first six columns of our original Gantt chart, each time looking for any of the category numbers we’ve ticked (what the conditional formatting hides is that the category numbers are in that original Gantt, they’re just the same color as the cell shading). When no tick-boxes are checked, it returns anything that has falls in to any category over the next six planned months. Once we start ticking checkboxes, this will return only the things over the next six planned months that are in one of the categories we’ve selected.

=query('Gantt view'!D1:1056,"Select D, E, F, G, H, I, J where D <> '' and (E matches '"&B1&"' or F matches '"&B1&"' or G matches '"&B1&"' or H matches '"&B1&"' or I matches '"&B1&"' or J matches '"&B1&"')",1)

How to create the Stakeholder View

The Planner Version sheet has a tab called Planned. You don’t need to fill out this tab — it has a query which extracts information from the Task View tab using a =query() function:

=QUERY({'Task view'!A1:F,'Task view'!O1:P},"Select * where not (Col6 contains 'pitched' or Col6 contains 'cancelled' or Col6 = '' or Col6 = 'Forgotten')")

All the formula above is doing is taking the Month, Task, Description, Blocker, Status, Category, and Full task columns, then showing every record where the status isn’t “unpitched,” “cancelled,” “forgotten,” or empty. That gives us a tab with the information we’re ready to share. We could also achieve this with =filter() if we reordered the data in the Task View tab, but this ordering of data is easier to work with, so we just use =query() and select only the columns we want here, combining the ranges horizontally by listing them between {} at the start of this formula.

Then, the Task View tab in our Stakeholder Version sheet file uses =importrange() to target that cleaned list we’ve created. To make sure the Stakeholder Version keeps functioning when you create copies of both of these files, all you need to do is go to the new Planner Version sheet and copy the URL of the page, then go to the Stakeholder Version, find the Task View tab, and update the importrange() formula in cell A1 to have the new URL of your Planner Version sheet. The cell will recalculate, you’ll need to grant permission, then it should work as normal.

How to update the God’s-I Version

This view gives you the following:

A quick look at the total number of tasks in any project which:

  1. Have a deadline within 10 days of now
  2. Have passed a deadline (with the task not completed)
  3. Don’t have a deadline set
  4. Aren’t briefed or aren’t budgeted for the next three months

It’ll also give you a quick look at the amount of unplanned budget per project, per month, to make sure you haven’t forgotten to plan a month and haven’t overplanned a month.

The God’s-I Version works in a similar way to the Stakeholder Version in that it pulls in information using =importrange(), but a key difference is that we want to pull from multiple sheets. Rewriting the formula could get to be a pain, so instead we can generate the formula we need in the Project URLs tab.

The only things you need to do are:

  1. Add the URL of the new Planner View sheet you want to include in the Project URLs tab of the God’s-I Version
  2. Grant permission for this sheet to access that sheet (you can click on the alert that appears in column A)
  3. Copy the value in Cell B1, go to the All Imported Task Views tab and select cell A2, then paste the value into the top bar. It’s important that we don’t paste straight into the cell or the sheet will run the concatenate formula rather than the query formula we’re making.

It’s worth noting that this sheet will have all the information about every project you’re managing. Once it’s set up, you shouldn’t share access to anyone unless you’re happy with them seeing all the budgeting details for each of the sheets.


A late conclusion:

Why are you looking for a conclusion down here? It’s in the middle of the post under the title of An early conclusion,” of course. Have a nice day!

Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don’t have time to hunt down but want to read!


Moz Blog

Posted in Latest NewsComments Off

Using the Flowchart Method for Diagnosing Ranking Drops – Whiteboard Friday

Posted by KameronJenkins

Being able to pinpoint the reason for a ranking drop is one of our most perennial and potentially frustrating tasks as SEOs. There are an unknowable number of factors that go into ranking these days, but luckily the methodology for diagnosing those fluctuations is readily at hand. In today’s Whiteboard Friday, we welcome the wonderful Kameron Jenkins to show us a structured way to diagnose ranking drops using a flowchart method and critical thinking.

Flowchart method for diagnosing ranking drops

Click on the whiteboard image above to open a high-resolution version in a new tab!


Video Transcription

Hey, everyone. Welcome to this week’s edition of Whiteboard Friday. My name is Kameron Jenkins. I am the new SEO Wordsmith here at Moz, and I’m so excited to be here. Before this, I worked at an agency for about six and a half years. I worked in the SEO department, and really a common thing we encountered was a client’s rankings dropped. What do we do?

This flowchart was kind of built out of that mentality of we need a logical workflow to be able to diagnose exactly what happened so we can make really pointed recommendations for how to fix it, how to get our client’s rankings back. So let’s dive right in. It’s going to be a flowchart, so it’s a little nonlinear, but hopefully this makes sense and helps you work smarter rather than harder.

Was it a major ranking drop?: No

The first question I’d want to ask is: Was their rankings drop major? By major, I would say that’s something like page 1 to page 5 overnight. Minor would be something like it just fell a couple positions, like position 3 to position 5.

We’re going to take this path first. It was minor.

Has there been a pattern of decline lasting about a month or greater?

That’s not a magic number. A month is something that you can use as a benchmark. But if there’s been a steady decline and it’s been one week it’s position 3 and then it’s position 5 and then position 7, and it just keeps dropping over time, I would consider that a pattern of decline.

So if no, I would actually say wait.

  • Volatility is normal, especially if you’re at the bottom of page 1, maybe page 2 plus. There’s going to be a lot more shifting of the search results in those positions. So volatility is normal.
  • Keep your eyes on it, though. It’s really good to just take note of it like, “Hey, we dropped. Okay, I’m going to check that again next week and see if it continues to drop, then maybe we’ll take action.”
  • Wait it out. At this point, I would just caution against making big website updates if it hasn’t really been warranted yet. So volatility is normal. Expect that. Keep your finger on the pulse, but just wait it out at this point.

If there has been a pattern of decline though, I’m going to have you jump to the algorithm update section. We’re going to get there in a second. But for now, we’re going to go take the major rankings drop path.

Was it a major ranking drop?: Yes

The first question on this path that I’d want to ask is:

Was there a rank tracking issue?

Now, some of these are going seem pretty basic, like how would that ever happen, but believe me it happens every once in a while. So just before we make major updates to the website, I’d want to check the rank tracking.

I. The wrong domain or URL.

That can be something that happens a lot. A site maybe you change domains or maybe you move a page and that old page of that old domain is still listed in your ranking tracker. If that’s the case, then the rank tracking tool doesn’t know which URL to judge the rankings off of. So it’s going to look like maybe you dropped to position 10 overnight from position 1, and that’s like, whoa, that’s a huge update. But it’s actually just that you have the wrong URL in there. So just check that. If there’s been a page update, a domain update, check to make sure that you’ve updated your rank tracker.

II. Glitches.

So it’s software, it can break. There are things that could cause it to be off for whatever reason. I don’t know how common that is. It probably is totally dependent on which kind of software you use. But glitches do happen, so I would manually check your rankings.

III. Manually check rankings.

One way I would do that is…

  • Go to incognito in Google and make sure you’re logged out so it’s not personalized. I would search the term that you’re wanting to rank for and see where you’re actually ranking.
  • Google’s Ad Preview tool. That one is really good too if you want to search where you’re ranking locally so you can set your geolocation. You could do mobile versus desktop rankings. So it could be really good for things like that.
  • Crosscheck with another tool, like Moz’s tool for rank tracking. You can pop in your URLs, see where you’re ranking, and cross-check that with your own tool.

So back to this. Rank tracking issues. Yes, you found your problem. If it was just a rank tracking tool issue, that’s actually great, because it means you don’t have to make a lot of changes. Your rankings actually haven’t dropped. But if that’s not the issue, if there is no rank tracking issue that you can pinpoint, then I would move on to Google Search Console.

Problems in Google Search Console?

So Google Search Console is really helpful for checking site health matters. One of the main things I would want to check in there, if you experience a major drop especially, is…

I. Manual actions.

If you navigate to Manual Actions, you could see notes in there like unnatural links pointing to your site. Or maybe you have thin or low-quality content on your site. If those things are present in your Manual Actions, then you have a reference point. You have something to go off of. There’s a lot of work involved in lifting a manual penalty that we can’t get into here unfortunately. Some things that you can do to focus on manual penalty lifting…

  • Moz’s Link Explorer. You can check your inbound links and see their spam score. You could look at things like anchor text to see if maybe the links pointing to your site are keyword stuffed. So you can use tools like that.
  • There are a lot of good articles too, in the industry, just on getting penalties lifted. Marie Haynes especially has some really good ones. So I would check that out.

But you have found your problem if there’s a manual action in there. So focus on getting that penalty lifted.

II. Indexation issues.

Before you move out of Search Console, though, I would check indexation issues as well. Maybe you don’t have a manual penalty. But go to your index coverage report and you can see if anything you submitted in your sitemap is maybe experiencing issues. Maybe it’s blocked by robots.txt, or maybe you accidentally no indexed it. You could probably see that in the index coverage report. Search Console, okay. So yes, you found your problem. No, you’re going to move on to algorithm updates.

Algorithm updates

Algorithm updates happen all the time. Google says that maybe one to two happen per day. Not all of those are going to be major. The major ones, though, are listed. They’re documented in multiple different places. Moz has a really good list of algorithm updates over time. You can for sure reference that. There are going to be a lot of good ones. You can navigate to the exact year and month that your site experienced a rankings drop and see if it maybe correlates with any algorithm update.

For example, say your site lost rankings in about January 2017. That’s about the time that Google released its Intrusive Interstitials Update, and so I would look on my site, if that was the issue, and say, “Do I have intrusive interstitials? Is this something that’s affecting my website?”

If you can match up an algorithm update with the time that your rankings started to drop, you have direction. You found an issue. If you can’t match it up to any algorithm updates, it’s finally time to move on to site updates.

Site updates

What changes happened to your website recently? There are a lot of different things that could have happened to your website. Just keep in mind too that maybe you’re not the only one who has access to your website. You’re the SEO, but maybe tech support has access. Maybe even your paid ad manager has access. There are a lot of different people who could be making changes to the website. So just keep that in mind when you’re looking into it. It’s not just the changes that you made, but changes that anyone made could affect the website’s ranking. Just look into all possible factors.

Other factors that can impact rankings

A lot of different things, like I said, can influence your site’s rankings. A lot of things can inadvertently happen that you can pinpoint and say, “Oh, that’s definitely the cause.”

Some examples of things that I’ve personally experienced on my clients’ websites…

I. Renaming pages and letting them 404 without updating with a 301 redirect.

There was one situation where a client had a blog. They had hundreds of really good blog posts. They were all ranking for nice, long tail terms. A client emailed into tech support to change the name of the blog. Unfortunately, all of the posts lived under the blog, and when he did that, he didn’t update it with a 301 redirect, so all of those pages, that were ranking really nicely, they started to fall out of the index. The rankings went with it. There’s your problem. It was unfortunate, but at least we were able to diagnose what happened.

II. Content cutting.

Maybe you’re working with a UX team, a design team, someone who is looking at the website from a visual, a user experience perspective. A lot of times in these situations they might take a page that’s full of really good, valuable content and they might say, “Oh, this is too clunky. It’s too bulky. It has too many words. So we’re going to replace it with an image, or we’re going to take some of the content out.”

When this happens, if the content was the thing that was making your page rank and you cut that, that’s probably something that’s going to affect your rankings negatively. By the way, if that’s happening to you, Rand has a really good Whiteboard Friday on kind of how to marry user experience and SEO. You should definitely check that out if that’s an issue for you.

III. Valuable backlinks lost.

Another situation I was diagnosing a client and one of their backlinks dropped. It just so happened to be like the only thing that changed over this course of time. It was a really valuable backlink, and we found out that they just dropped it for whatever reason, and the client’s rankings started to decline after that time. Things like Moz’s tools, Link Explorer, you can go in there and see gained and lost backlinks over time. So I would check that out if maybe that might be an issue for you.

IV. Accidental no index.

Depending on what type of CMS you work with, it might be really, really easy to accidentally check No Index on this page. If you no index a really important page, Google takes it out of its index. That could happen. Your rankings could drop.So those are just some examples of things that can happen. Like I said, hundreds and hundreds of things could have been changed on your site, but it’s just really important to try to pinpoint exactly what those changes were and if they coincided with when your rankings started to drop.

SERP landscape

So we got all the way to the bottom. If you’re at the point where you’ve looked at all of the site updates and you still haven’t found anything that would have caused a rankings drop, I would say finally look at the SERP landscape.

What I mean by that is just Google your keyword that you want to rank for or your group of keywords that you want to rank for and see which websites are ranking on page 1. I would get a lay of the land and just see:

  • What are these pages doing?
  • How many backlinks do they have?
  • How much content do they have?
  • Do they load fast?
  • What’s the experience?

Then make content better than that. To rank, so many people just think avoid being spammy and avoid having things broken on your site. But that’s not SEO. That’s really just helping you be able to compete. You have to have content that’s the best answer to searchers’ questions, and that’s going to get you ranking.

I hope that was helpful. This is a really good way to just kind of work through a ranking drop diagnosis. If you have methods, by the way, that work for you, I’d love to hear from you and see what worked for you in the past. Let me know, drop it in the comments below.

Thanks, everyone. Come back next week for another edition of Whiteboard Friday.

Video transcription by Speechpad.com

Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don’t have time to hunt down but want to read!


Moz Blog

Posted in Latest NewsComments Off

Tracking Your Link Prospecting Using Lists in Link Explorer

Posted by Dr-Pete

I’m a lazy marketer some days — I’ll admit it. I don’t do a lot of manual link prospecting, because it’s a ton of work, outreach, and follow-up. There are plenty of times, though, where I’ve got a good piece of content (well, at least I hope it’s good) and I want to know if it’s getting attention from specific sites, whether they’re in the search industry or the broader marketing or PR world. Luckily, we’ve made that question a lot easier to answer in Link Explorer, so today’s post is for all of you curious but occasionally lazy marketers. Hop into the tool if you want to follow along:

Open Link Explorer

(1) Track your content the lazy way

When you first visit Link Explorer, you’ll see that it defaults to “root domain”:

Some days, you don’t want to wade through your entire domain, but just want to target a single piece of content. Just enter or paste that URL, and select “exact page” (once you start typing a full path, we’ll even auto-select that option for you):

Now I can see just the link data for that page (note: screenshots have been edited for size):

Good news — my Whiteboard Friday already has a decent link profile. That’s already a fair amount to sort through, and as the link profile grows, it’s only going to get tougher. So, how can I pinpoint just the sites I’m interested in and track those sites over time?

(2) Make a list of link prospects

This is the one part we can’t automate for you. Make a list of prospects in whatever tool you please. Here’s an imaginary list I created in Excel:

Obviously, this list is on the short side, but let’s say I decide to pull a few of the usual suspects from the search marketing world, plus one from the broader marketing world, and a couple of aspirational sites (I’m probably not going to get that New York Times link, but let’s dream big).

(3) Create a tracking list in Link Explorer

Obviously, I could individually search for these domains in my full list of inbound links, but even with six prospects, that’s going to take some time. So, let’s do this the lazy way. Back in Link Explorer, look at the very bottom of the left-hand navigation and you’ll see “Link Targeting Lists”:

Keep scrolling — I promise it’s down there. Click on it, and you’ll see something like this:

On the far-right, under the main header, click on “[+] Create new list.” You’ll get an overlay with a three-step form like the one below. Just give your list a name, provide a target URL (the page you want to track links to), and copy-and-paste in your list of prospects. Here’s an example:

Click “Save,” and you should immediately get back some data.

Alas, no link from the New York Times. The blue icons show me that the prospects are currently linking to Moz.com, but not to my target page. The green icon shows me that I’ve already got a head-start — Search Engine Land is apparently linking to this post (thanks, Barry!).

Click on any arrow in the “Notes” column, and you can add a note to that entry, like so:

Don’t forget to hit “Save.” Congratulations, you’ve created your first list! Well, I’ve created your first list for you. Geez, you really are lazy.

(4) Check in to track your progress

Of course, the real magic is that the list just keeps working for you. At any time, you can return to “Link Tracking Lists” on the Link Explorer menu, and now you’ll see a master list of all your lists:

Just click on the list name you’re interested in, and you can see your latest-and-greatest data. We can’t build the links for you, but we can at least make keeping track of them a lot easier.

Bonus video: Now in electrifying Link-o-Vision!

Ok, it’s just a regular video, although it does require electricity. If you’re too lazy to read (in which case, let’s be honest, you probably didn’t get this far), I’ve put this whole workflow into an enchanting collection of words and sounds for you:

I hope you’ll put your newfound powers to good. Let us know how you’re using Tracking Lists (or how you plan to use them) in the comments, and where you’d like to see us take them next!

Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don’t have time to hunt down but want to read!


Moz Blog

Posted in Latest NewsComments Off

Advert