The Local Algorithm: Relevance, Proximity, and Prominence

Posted by MaryBowling

How does Google decide what goes into the local pack? It doesn’t have to be a black box — there’s logic behind the order. In this week’s Whiteboard Friday, renowned local SEO expert Mary Bowling lays out the three factors that drive Google’s local algorithm and local rankings in a simple and concise way anyone can understand.

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

Video Transcription

Hi, Moz fans. This is Mary Bowling from Ignitor Digital, and today I want to talk to you about the local algorithm. I’d like to make this as simple as possible for people to understand, because I think it’s a very confusing thing for a lot of SEOs who don’t do this every day.

The local algorithm has always been based on relevance, prominence, and proximity

1. Relevance

For relevance, what the algorithm is asking is, “Does this business do or sell or have the attributes that the searcher is looking for?” That’s pretty simple. So that gives us all these businesses over here that might be relevant. For prominence, the algorithm is asking, “Which businesses are the most popular and the most well regarded in their local market area?”

2. Proximity

For proximity, the question really is, “Is the business close enough to the searcher to be considered to be a good answer for this query?” This is what trips people up. This is what really defines the local algorithm — proximity. So I’m going to try to explain that in very simple terms here today.

Let’s say we have a searcher in a particular location, and she’s really hungry today and she wants some egg rolls. So her query is egg rolls. If she were to ask for egg rolls near me, these businesses are the ones that the algorithm would favor.

3. Prominence

They are the closest to her, and Google would rank them most likely by their prominence. If she were to ask for something in a particular place, let’s say this is a downtown area and she asked for egg rolls downtown because she didn’t want to be away from work too long, then the algorithm is actually going to favor the businesses that sell egg rolls in the downtown area even though that’s further away from where the searcher is.

If she were to ask for egg rolls open now, there might be a business here and a business here and a business here that are open now, and they would be the ones that the algorithm would consider. So relevance is kicking in on the query. If she were to ask for the cheapest egg rolls, that might be here and here.

If she were to ask for the best egg rolls, that might be very, very far away, or it could be a combination of all kinds of locations. So you really need to think of proximity as a fluid thing. It’s like a rubber band, and depending on… 

  • the query
  • the searcher’s location
  • the relevance to the query
  • and the prominence of the business 

….is what Google is going to show in that local pack.

I hope that makes it much clearer to those of you who haven’t understood the Local Algorithm. If you have some comments or suggestions, please make them below and thanks for listening.

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!

[ccw-atrib-link]

Google’s Physical Web and its Impact on Search

Posted by Tom-Anthony

In early October, Google announced a new project called ”
The Physical Web,” which they explain like this:

The Physical Web is an approach to unleash the core superpower of the web: interaction on demand. People should be able to walk up to any smart device – a vending machine, a poster, a toy, a bus stop, a rental car – and not have to download an app first. Everything should be just a tap away.

At the moment this is an experimental project which is designed to promote establishing an open standard by which this mechanism could work. The two key elements of this initiative are:

URLs: The project proposes that all ‘smart devices’ should advertise a URL by which you can interact with that device. The device broadcasts its URL to anyone in the vicinity, who can detect it via their smartphone (with the eventual goal being this functionality is built into the smart phone operating systems rather than needing third-party apps).


Beacons:
Not well known until Apple recently jumped on the bandwagon announcing iBeacons, beacon technology has been around for a couple of years now. Using a streamlined sibling of Bluetooth, called Bluetooth Low Energy (no pairing, range of ~70 metres / ~230 feet) it allows smartphones to detect the presence of nearby beacons and their approximate distance. Until now they’ve mostly been used to ‘hyper-local’ location based applications (check this blog post of mine for some thoughts on how this might impact SEO).

The project proposes adapting and augmenting the signal that Beacons send out to include a URL by which nearby users might interact with a smart device.

This post is about looking to the future at ways this could potentially impact search. It isn’t likely that any serious impact will happen within the next 18 months, and it is hard to predict exactly how things will pan out, but this post is designed to prompt you to think about things proactively.

Usage examples

To help wrap your head around this, lets look at a few examples of possible uses:

Bus times: This is one of the examples Google gives, where you walk up to a bus stop and on detecting the smart device embedded into the stop your phone allows you to pull the latest bus times and travel info.

Item finder: Imagine when you go to the store looking for a specific item. You could pull out your phone and check stock of the item, as well as being directed to the specific part of the store where you can find it.

Check in: Combined with using URLs that are only accessible on local wifi / intranet, you could make a flexible and consistent check in mechanism for people in a variety of situations.

I’m sure there are many many more applications that are yet to be thought up. One thing to notice is that there is no reason you can’t bookmark these advertised URLs and use them elsewhere, so you can’t be sure that someone accessing the URL is actually by the device in question. You can get some of the way there by using URLs that are only accessible within a certain network, but that isn’t going to be a general solution.

Also, note that these URLs don’t need to be constrained to just website URLs; they could just as well be
deep links into apps which you might have installed.

Parallels to the web and ranking

There are some obvious parallels to the web (which is likely why Google named it the way they did). There will be many smart devices which will map to URLs which anyone can go to. A corollary of this is that there will be similar issues to those we see in search engines today. Google already identified one such issue—ranking—on the page for the project:

At first, the nearby smart devices will be small, but if we’re successful, there will be many to choose from and that raises an important UX issue. This is where ranking comes in. Today, we are perfectly happy typing “tennis” into a search engine and getting millions of results back, we trust that the first 10 are the best ones. The same applies here. The phone agent can sort by both signal strength as well as personal preference and history, among many other possible factors. Clearly there is lots of work to be done here.

So there is immediately a parallel between with Google’s role on the world wide web and their potential role on this new physical web; there is a suggestion here that someone needs to rank beacons if they become so numerous that our phones or wearable devices are often picking up a variety of beacons. 

Google proposes proximity as the primary measure of ranking, but the proximity range of BLE technology is very imprecise, so I imagine in dense urban areas that just using proximity won’t be sufficient. Furthermore, given the beacons are cheap (in bulk, $5 per piece will get you standalone beacons with a year-long battery) I imagine there could be “smart device spam.”

At that point, you need some sort of ranking mechanism and that will inevitably lead to people trying to optimise (be it manipulative or a more white-hat approach).
However, I don’t think that will be the sole impact on search. There are several other possible outcomes.

Further impacts on the search industry

1. Locating out-of-range smart devices

Imagine that these smart devices became fairly widespread and were constantly advertising information to anyone nearby with a smart devices. I imagine, in a similar vein to schema.org actions which provide a standard way for websites to describe what they enable someone to do (“affordances,” for the academics), we could establish similar semantic standards for smart devices enabling them to advertise what services/goods they provide.

Now imagine you are looking for a specific product or service, which you want as quickly as possible (e.g “I need to pick up a charger for my phone,” or “I need to charge my phone on the move”). You could imagine that Google or some other search engine will have mapped these smart devices. If the above section was about “ranking,” then this is about “indexing.”

You could even imagine they could keep track of what is in stock at each of these places, enabling “environment-aware” searches. How might this work? Users in the vicinity whose devices have picked up the beacons, and read their (standardised) list of services could then record this into Google’s index. It sounds like a strange paradigm, but it is exactly how Google’s app indexing methodology works.

2. Added context

Context is becoming increasingly important for all searches that we do. Beyond your search phrase, Google look at what device you are on, where you are, what you have recently searched for, who you know, and quite a bit more. It makes our search experiences significantly better, and we should expect that they are going to continue to try to refine their understanding of our context ever more.

It is not hard to see that knowing what beacons people are near adds various facets of context. It can help refine location even further, giving indications to the environment you are in, what you are doing, and even what you might be looking for.

3. Passive searches

I’ve spoken a little bit about passive searches before; this is when Google runs searches for you based entirely off your context with no explicit search. Google Now is currently the embodiment of this technology, but I expect we’ll see it become more and more

I believe could even see see a more explicit element of this become a reality, with the rise of conversational search. Conversational search is already at a point where a search queries can have persistent aspects (“How old is Tom Cruise?”, then “How tall is he?” – the pronoun ‘he’ refers back to previous search). I expect we’ll see this expand more into multi-stage searches (“Sushi restaurant within 10 minutes of here.”, and then “Just those with 4 stars or more”).

So, I could easily imagine that these elements combine with “environment-aware” searches (whether they are powered in the fashion I described above or not) to enable multi-stage searches that result in explicit passive searches. For example, “nearby shops with iPhone 6 cables in stock,” to which Google fails to find a suitable result (“there are no suitable shops nearby”) and you might then answer “let me know when there is.”

Wrap up

It seems certain that embedded smart devices of some sort are coming, and this project from Google looks like a strong candidate to establish a standard. With the rise of smart devices, whichever form they end up taking and standard they end up using, it is certain this is going to impact the way people interact with their environments and use their smart phones and wearables.

It is hard to believe this won’t also have a heavy impact upon marketing and business. What remains less clear is the scale of impact that this will have on SEO. Hopefully this post has got your brain going a bit so as and industry, we can start to prepare ourselves for the rise of smart devices.

I’d love to hear in the comments what other ideas people have and how you guys think this stuff might affect us.

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!

[ccw-atrib-link]

Announcing the 2014 Local Search Ranking Factors Results

Posted by David-Mihm

Many of you have been tweeting, emailing, asking in conference Q&As, or just generally awaiting this year’s Local Search Ranking Factors survey results.
Here they are!

Hard to believe, but this is the seventh year I’ve conducted this survey—local search has come a long way since the early days of the 10-pack way back in 2008! As always, a massive thanks to all of the expert panelists who in many cases gave up a weekend or a date night in order to fill out the survey.

New this year

As the complexity of the local search results has increased, I’ve tried to keep the survey as manageable as possible for the participants, and the presentation of results as actionable as possible for the community. So to that end, I’ve made a couple of tweaks this year.

Combination of desktop and mobile results

Very few participants last year perceived any noticeable difference between ranking criteria on desktop and mobile devices, so this year I simply asked that they rate localized organic results, and pack/carousel results, across both result types.

Results limited to top 50 factors in each category

Again, the goal here was to simplify some of the complexity and help readers focus on the factors that really matter. Let me know in the comments if you think this decision detracts significantly from the results, and I’ll revisit it in 2015.

Factors influenced by Pigeon

If you were at Matt McGee’s Pigeon session at SMX East a couple of weeks ago, you got an early look at these results in my presentation. The big winners were domain authority and proximity to searcher, while the big losers were proximity to centroid and having an address in the city of search. (For those who weren’t at my presentation, the latter assessment may have to do with larger radii of relevant results for geomodified phrases).

My own takeaways

Overall, the
algorithmic model that Mike Blumenthal developed (with help from some of the same contributors to this survey) way back in 2008 continues to stand up. Nonetheless, there were a few clear shifts this year that I’ll highlight below:

  • Behavioral signals—especially clickthrough rate from search results—seem to be increasing in importance. Darren Shaw in particular noted Rand’s IMEC Labs research, saying “I think factors like click through rate, driving directions, and “pogo sticking” are valuable quality signals that Google has cranked up the dial on.”
  • Domain authority seems to be on its way up—particularly since the Pigeon rollout here in the U.S. Indeed, even in clear instances of post-Pigeon spam, the poor results seem to relate to Google’s inability to reliably separate “brands” from “spam” in Local. I expect Google to get better at this, and the importance of brand signals to remain high.
  • Initially, I was surprised to see authority and consistency of citations rated so highly for localized organic results. But then I thought to myself, “if Google is increasingly looking for brand signals, then why shouldn’t citations help in the organic algorithm as well?” And while the quantity of structured citations still rated highly for pack and carousel results, consistent citations from quality sources continue to carry the day across both major result types.
  • Proximity to searcher saw one of the biggest moves in this year’s survey. Google is getting better at detecting location at a more granular level—even on the desktop. The user is the new Centroid.
  • For markets where Pigeon has not rolled out yet (i.e. everywhere besides the U.S.), I’d encourage business owners and marketers to start taking as many screenshots of their primary keywords as possible. With the benefit of knowing that Pigeon will eventually roll out in your countries, the ability to compare before-and-after results for the same keywords will yield great insight for you in discerning the direction of the algorithm.

As with every year, though, it’s the comments from the experts and community (that’s you, below!) that I find most interesting to read.  So I think at this point I’ll sign off, crack open a
GABF Gold-Medal-Winning Breakside IPA from Portland, and watch them roll in!

2014 Local Search Ranking Factors

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!

[ccw-atrib-link]