[MUSIC]

Well, thank you for that overly generous introduction Sirus.

I'm not sure who got a better one, Matt Brown or me.

So thank you very much.

I am indeed here to talk to you guys for a few minutes today about Facebook search.

And before I do, I do want to lead with a brief disclaimer,

which is I am by no means a Facebook marketing expert.

That's the domain of friends of mine like Marty Weintraub, and Will Scott, and

Matt Siltala.

But, as Cyrus said, I have been spending a lot of time experimenting, and

researching a little bit into Facebook search product, and had my interest piqued

by a number of stories that have come out about Facebook search this spring.

And so I wanted to share some things with you and

hopefully by the end of this 30 minutes, I've convinced you that it's something

that you too should think about paying more attention to.

So that's really the goal.

Brief outline of what I plan to cover here today.

Little bit of context, again, about why you should care.

Some of the history involved with Facebook search,

which I think is incredibly important to look at for where they might be going.

Then I'll dive into some tactual ways to construct queries.

Especially structured search queries using the graph Piece of Facebook search.

And use cases for

those queries before getting into tactical implications for the future.

So with that a little bit about why I think you should care and

why you should be paying attention to Facebook search.

So I'm sure all of you have seen some variation of this graphic,

basically Facebook's user growth in the last seven years.

The main point to me is despite we hear about Instagram and Whatsapp, and

Snapchat, and all these things bleeding Facebook users

they're still growing at a pretty good clip.

The percentage growth rate has obviously slowed down but

it's a linear growth pattern.

They have almost 1.5 billion users on Facebook today so

it's a huge number of people.

And more importantly, something that many of you probably didn't know.

I certainly didn't know it before I put this presentation together.

As early as September of 2012, Facebook was already getting

a billion search queries a day, which is kind of a crazy number.

And in fact, I wanted to see how crazy a number it was, so I went back and

I looked at the Com Square data.

From this time period, September of 2012.

And I know Marshall's presentation yesterday was spot on.

This was not perfectly precise numbers probably, but

I think it's directionally accurate.

And if you look at this time period, September of 2012,

you see Google's at about 110 billion searches a month.

And Facebook's 30 billion searches a month already in September of 2012 would

have made it a very strong number two search engine at the time.

And I think that a lot of us just don't realize how much search activity

was already happening on Facebook three years ago, and

has continued to increase in the last few years.

And I think this becomes even more important when you look at slides like

this and the one that Ryan shared from the RKG studies in his initial presentation.

The overall growth rate of referrals from Google search, and

especially from Google organic search, thinking about Dr. Pete's presentation.

The number of those referrals is continuing to level off,

at the very least.

And so we as marketers, if we want to continue to try to grow our audience

through search, I think we need to start thinking about different opportunities.

And Jeremy Stoppleman, Yelp's CEO,

actually echoed this sentiment a few months ago in one of their earnings calls.

Basically saying that the age of users going to Google, and typing a query, and

clicking a link, and

visiting a website,and hopefully converting on that website.

The traditional funnel that I think a lot of us have kind of built our businesses

on and are generally evaluated on, that era has kind of peaked,

we kind of maxed out that.

And so if we want to continue to grow our audience through search,

I think Facebook represents a really good opportunity to do that.

The picture becomes even clearer when you look at what's happening on mobile.

So this is a slide from Flurry Analytics.

Time spent on mobile phones.

We probably, hopefully you guys already know this but

86% of the time is spent in app and only 14% in mobile browsers.

Obviously it's a huge threat to Google [COUGH], and one of the reasons

that they're pursuing all these alternative strategies in mobile.

And in particular, if you look at the specific apps that we're spending time on,

Facebook dominates.

17% of all time on mobile devices is spent in the Facebook app

compared to 12% in both browsers, basically Chrome and Safari.

That's browser time, that's not search time.

That's browser time.

So Facebook basically has become the homepage,

the home app for pretty much everyone on mobile phones.

And so as we look at, again,

the growth rate in search volume, declining at Google,

the share of organic search as a function of web visits, Declining from Google.

I think Facebook is generally very well-positioned

to help us make up some of that gap.

And for those of us who are paying attention in the early days of Facebook

search, I think it represents a great opportunity.

And yet, I don't know how many people are really aware

of the amazing search infrastructure that Facebook already has.

So my favorite Industry news sites, search engine land.

You can see here comparing the number of

articles they've written on Facebook search which is google search.

It's a dramatic difference.

Google is, we all have a sort of Google blinders on.

Even, especially, at Moz.

The same search for articles published on the Moz blog.

And even when we publish articles on Facebook Search on the Moz Blog,

two of the top three results are about Facebook influencing Google results.

Incidently, the top result here, the Marketer's Guide to Facebook Graph search,

an excellent guest post that I'll refer to at the end of the deck.

But at any rate, I think we all sort of have

these blinders on when it comes to search, that it's so Google focused.

And I think that Facebook is really poised

to present a great opportunity to those of us who are paying attention.

So let me walk you through a little bit about where Facebook Search has been,

where it's come from, and where I think it might be going.

As I said, Mark Zuckerberg was quoted three years ago that they're

already getting a billion daily queries.

This was even before they actually had a formal search product.

So Graph Search launched in January of 2013,

and Graph Search was and still is I think most useful and

kinda primarily focused on returning people results.

So this is from their initial press release.

People who like cycling and live in Seattle, Washington.

These kinds of queries are really what graph search was designed for.

And Mark Zuckerberg kind of backed that up in one of the initial articles that was

written about graph search on wired,

saying one of his primary use cases was really around recruiting.

Which is obviously very people focused,

but we do see that he has his eye on google search more broadly.

Whether you're looking for a top restaurant or a museum, or

let's just say the favorite tunes of art majors in Montreal.

So much more general queries that facebook search will eventually apply to.

So graph search was rolled out to all English users.

Soon after that Not much really happened

from a product standpoint in the interceding kind of year and a half, and

in December of last year I think something pretty important happened.

Graph Search became just Search, and really what that entailed was posts

as a result type, so instead of people as a result type kind of being prioritized,

I think posts now got a much bigger boost.

In terms of the types of results that Facebook was showing for searches.

And simultaneously to that or roughly simultaneously, Facebook dropped

Bing from back-filling it's web search results within their app.

And so I think that these two things combined

were a pretty important development.

In March of this year they acquired a company called TheFind,

which I've never heard of.

But I think directionally it's kind of important.

TheFind is a product search startup and

I think if you look at the markets that, anybody really, can monetize and search.

Product search is a big one.

So this strikes me as a real, so the start of a real move

into an actual search engine that they can monetize as I said.

And then, just recently,

these two I think are really interesting came out in May, just two months ago.

The adlink feature which only some users have on IOS.

This is essentially the ability to look for stories related to keywords that you

want to share with your audience without ever having to leave the Facebook ap.

Right. So you don't have to be in Google and

search for a story or

you don't have to be in Firefox or whatever and pull a story link.

You can do it right from within the Facebook ap.

And to this end, Facebook says it has indexed over a trillion posts

that reference web content that this feature is built on.

So, they already have a pretty amazing sort of library of content that

people can share, and it can be searched from directly within the Facebook app.

And just two days later,

they formally announce this partnership with a select group of publishers.

That you can now browse content from certain publishers,

again directly in the Facebook app.

And so this, I think, speaks to a more complete Facebook search ecosystem.

Where more and more of our behavior, search and.

Social related is all happening within Facebook.

And as far as where they're going I don't pretend to know the entire future.

But I do know that this is really kind of just the beginning.

We're sort of in the beginning to middle phase of the evolution of the product.

And how I know that is Mark Zuckerberg has come out and

said The graph search is really a five year thing.

And, so we're really in year two and

a half to three of Facebook's kind of really investment in search.

And, so I think the next two years,

we're going to see a tremendous amount of innovation and new features and

new opportunities provided by Facebook to us as marketers.

And so, although right now, and

a lot of this presentation is gonna focus on The potential and

the opportunities presented by this kind of people focused graph search.

I think in the future, Facebook search is going to be as, represents a tremendous

opportunity in terms of audience building, again to help us make up the loss of

traditional organic referral traffic that I think is kind of inevitable from Google.

Okay. A few slides about how

to construct some of these graph search type queries.

So, for all of you grammarians out there, my mom is grammarian,

so I had to include this slide.

Some parts of Facebook speech, you have subjects and objects Modifiers,

verbs and set relationships, and that's really complicated.

I'll explain to you an example of how this work.

But, Facebook graph search only interprets very

specifically structured types of queries, and so I'll show you how that works.

Here, you can see a lonely single entry

Matthew Brown as the only current Moz employee who likes the Portland Timbers.

Matt doesn't have a lot of friends generally, but it actually makes sense.

>> [LAUGH] >> In the context of this result,

because Moz of course is based in Seattle.

The Sounders are the Timbers' number one rival.

And so not very many Mozers like, Like the Timbers, but anyway, so

here's how this query is constructed.

You have the subject, you're looking for employees who like,

the object the Portland Timbers, and then the modifiers of current Moz employees.

And so, again, just an example, only certain types of structured queries

will work in terms of returning these people-focused results.

Here is a non-exhaustive List of parameters that you may find

useful as you do your own experiments.

And, I'll just leave it here for

you to refer to in the Slidedeck that you can download.

But a few ideas of things are visitors to a particular venue,

employees we already saw, professions, apps used, pages liked.

These are the kinds of Of subjects and objects that I think

provide a lot of interest or offer a lot of opportunity for marketers.

And professions in particular are one of my favorite areas that I've discovered.

Which makes sense given that Zuckerberg's,

one of his primary use cases for graph search early on was recruiting.

And if you want to know what professions are kind of possible to perform this crazy

round, just go and perform a fake edit to your own Facebook personal profile and

see the kinds of professions that are [INAUDIBLE] there in the drop downs.

This looks pretty similar to me to have free-based text automizes

their professional.

Sort of categories.

And some of the content that you see on pages about these professions

that Facebook has generated is pulled from Wikipedia.

So, those might also give you some alternative ideas for professions to try.

Next slide.

There we go.

Interests are also something that you can search around.

Basically.

Professions and interests are two node types within the Facebook graph.

Here's an example of a page, again,

that Facebook has created with Wikipedia content around this interest that's

accessible on a particular graph ID URL, small business.

There's a million of these things.

You can, again, look at Freebase, look at Wikipedia.

You could also potentially pretend to run a Facebook ad campaign and

see the interests against which you can advertise.

These are all really useful parameters around which to

perform these structured searches.

And sometimes people, people can also be nodes in the graph,

particularly if they're well-known figures or

Have some sort of authority within the overall Facebook graph.

And just one example of a very specific query that only returns

four results worldwide.

Political junkies will understand why that is.

But let me explain how this works.

Basically if you're looking at it from a URL perspective, every URL for

a Facebook search result starts with.

https, facebook.com/search, you the get the graph ID of senator Jim page.

and then likers of that page.

Append to that the ID of the interest page about environment and

then likers of that interest, and then you look at the intersection of those groups.

So for those of you that don't know,

the reason that there's only four very confused souls who show up for The result,

Senator Jim Inhofe is the head of the Senate Committee on the Environment

who happens to believe that global warming is a hoax.

So, if you are one of these four people, I'm sorry,

I'm not entirely sure how you reconcile those two differences.

So, I just put that up there as an illustration of the amazing

specificity that you can sometimes find with these Facebook search queries.

And who knows, if your target market is a very small subset of people,

those kinds of queries might be really helpful in identifying our audience.

Okay, you can also construct undefined queries.

A little less formal Less structure in the results using the STR parameter.

There's a couple of great tutorials I've linked to

here in the slide deck that shows you how to use this.

Essentially you're looking for keywords on a particular page type.

That as opposed to and sort of nodes in the graph.

And just an example of where this might be useful is here I'm looking at

professional profiles that people have created where they might mention blogging.

And you can see here these are three very well known bloggers that showed up for

this type of search.

And I'm looking for

the intersection of people who mention blogging on their employment profiles.

Even less structure.

You can try a URL that looks something like this with

what pages in Portland talk about float tanks.

Facebook, as you can see, doesn't really know how to interpret that in

terms of returning structured results for that query.

They think that means cities of people names Float Tank

that are in Portland, Oregon.

So, these kinds of things again don't always work.

Sometimes you don't get these structured results, but

if you just do this search more broadly as a keyword,

you can see the URL sort of down at the bottom of that screenshot.

Here, you can see the primary role that I think post

results are going to continue to play in the Facebook interface.

And you can see that the top result here within posts is something that was shared

by a friend of mine within my social graph, a piece of content about it happens

to be about a tank conference they are having in Portland but a piece of

content that is relevant to those key words from out there across the web so

this is where I think that intersection of that ad link capability and

the ability to browse content directly within Facebook.

This is where that, I think,

is really going to come together in a big way in the future.

But, for now, let's talk about a few use cases around which the structured

Facebook graph queries might be valuable to you as marketers today.

So one of them is around content marketing.

We heard Matt and Kristina in their earlier presentations

Talk about the importance of targeting your content to a specific persona.

And here's a query that can kind of help you figure out what kind of content

might be a good target as well as is talking about.

What do people aspire to, not just what they are searching for.

Here you can see what are the interests of people that like my page.

This is perch furniture,

a friend of mine who runs a small furniture store in Portland.

Interior design, you might expect, is a pretty good overlap for a furniture store,

but things like gardening, knitting, sewing.

If there's a creative way that you can get content out there in

front of your audience, you know they're already interested in these topics.

Probably gonna have a better chance of spreading virally among the key

audience you're interested in than just writing about furniture all day long.

Another thing, especially if you work with small businesses who have a hard time

creating content on their own, thinking about using photos as content,

specifically photos taken at your place of business.

Reach out to these folks who took a great photo at your business, first of all ask

them permission to re-use their photo in your newsletter, but maybe ask them for

a quick interview about what it was like when they were under the needle or

whatever they were For your orthodontic practice.

So I think thinking about content that your users have already created

is a good use case as well.

Audience intelligence.

So here's one, pages liked by people who like MozLocal.

And of course Search Engine Land shows up at the top of the list, not surprising.

But maybe you didn't know this if you didn't perform,

haven't performed this career before you might discover additional.

pages that you that are surprising that your fans are already liking and

so I think it's a great opportunity for

you to reach out and potentially get a guest article on some of these places.

For those of you who are lazy like me you can target these sites explicitly

in the google display network and that is a good way to not only capture

People that are evangelists for your brand, but

other sort of look alike folks that might be interested in what you have to offer.

You can even add a local lens to this.

Of course I always have local on the brain.

So here's an example of pages liked by Portland,

Oregon residents who like Perch Furniture and I see a calendar here so

I might Promote an event that I have on this calendar or

even put a banner ad at the top for fans of this

website, who also like my business.

Influencer research, this is where I think is maybe the most powerful area,

the graph search has to offer.

So If you're thinking about who do I talk to about helping me get the word out about

my content?

This professional search, journalists Who live in your town, Portland, Oregon.

He wanted to take it one step further and know where you have a soft intro already.

Friends of my friends, who are journalists, and live in Portland,

Oregon, kind of a great search.

And then journalists who live in your town and work at a particular outlet.

I mean, this kind of thing is amazing in terms of figuring out who to reach out to,

who to build a relationship with.

When you want your content to be spread in your community.

If you aren't locally focused, you can do a similar thing which using that STR

parameter for the less structured results.

Who's writing, who's a blogger or journalist or whatever.

Who's also interested in your topic.

So this URL is people interested in golf and

blogging on their Professional profiles.

You can go one step further and look at golf as an interest,

golfers who are journalists and live in Portland, Oregon.

Journalists who like Small Business Saturday, which is a page,

not an interest, and live in Portland, Oregon.

There's a lot of possibilities for these kinds of phrases.

This is another one that I think has a lot of potential, probably more for

big brands than small brands.

But, you can look at the profession of Presidents, or Vice Presidents, or owners,

or whatever decision-maker job title that you want,

who like your page and live in a given market.

So if you're doing a product launch, or you're hosting an event,

these are people that you wanna reach out to as VIPs, asking them to

spread the word about your event or product to their networks etc, etc.

I think this could be a really valuable thing for a lot of brands out there.

And then this one is a tip that I actually brought up

at our local up conference back in February that a lot of people liked.

So I thought I'll share with a broader audience.

People who use Yelp, who use the Yelp app and like perch furniture.

So Yelp obviously frowns upon asking people specifically to leave you a review

on Yelp, but you can do this in a white hat way.

These are people you know who are active Yelp users

who have all ready expressed that they like your business.

You can reach out and

ask them to actually check in the next time that they come into your shop.

Tell them you have some kind of special for people who check in, etc., etc.

Make sure they feel like a VIP.

Yelp is totally fine with you asking people to check in.

They actually came out and said that publicly about two months ago, or

three months ago.

And Yelp will then actually prompt people who check in to a given business to leave

a review through the Yelp app.

So it's a totally white hat way that you can kind of segment your audience and

identify who is a good possible reviewer and

get reviews that are really hard to get.

Some final use cases around business development.

So if you sell a product and you're trying to get it distributed in a new market,

one potential graph search you might do bars in Seattle,

Washington, liked by people who like Breakside Breweries.

So if you run the distribution operation at Breakside, and

you're trying to into Seattle, why not get your products out in stores that you are,

bars in this case, where you know you already have a built in fan base.

And I think that can really help you seed, potentially seed your presence in

additional markets beyond your kind of home base.

And if you're more on the B to B side this is a potentially valuable search.

People who like your business and work at another business.

If you're looking for an internal cheerleader to help you

with the proposal or getting executive buy-in.

A couple of locally focused examples here, just to kind of close out, you can get

really creative with targeted offers, so this is one based on relationship status.

Singles, who like one of my favorite coffee shops,

called The Coffee Roasters, and live in Portland,

Oregon you know, potentially you can reach out to them and say, hey you know,

here's a $5 coupon or here's a free coffee the next time you come in on a date.

You know, hope it works great for you.

Something like that.

Or if you are a day spa you know, engaged people who like your spa you know,

you can reach out to them and say, hey we noticed you just got engaged.

Congratulations, we'd like to offer your bridal party a you know,

20% discount on our bridal party spot package.

Something like that and this is all stuff that you can do organically that Facebook

helps you identify who these people are.

If you wanna take it to the next level there is a bonus couple of slides here.

Number one you can plug in Facebook profiles,

hopefully many of you know about this tactic already,

you can plug in individual Facebook profiles into the full contact API.

And FullContact will then tell you all the other social networks that they know about

that this person is on, which is really neat.

So if LinkedIn is a more appropriate venue for

you to actually reach out to this person.

Or if Twitter is a more appropriate venue to actually reach out to this person,

the FullContact API can inform you of those opportunities.

And again, for those of you who are lazy like me, and

don't want to learn about all of those grammar rules, and are frustrated by

the fact that Facebook doesn't always return structured results for

really creative searches that you though they should.

Intelligent software has built this amazing little widget.

I don't know these guys, just happened to come across it as I was doing research for

this presentation, where you can punch in all of the kind of

Various parameters that you're interested in, and

they will construct the most appropriate URL to return results for you.

So if you don't wanna spend a lot of time learning how this works and

just wanna do it, this is a great resource for you.

Okay, looks like I've got about five minutes left, more or less on time.

Let's look in the palantir.

I had to throw in a Christopher Lee reference, I'm glad Dr.

Pete did, what an amazing actor, RIP.

So what are the implication, the tactical implications for

maintaining a strong Facebook search presence as they grow that product?

Number one, I would encourage you guys to move beyond the page like.

Hopefully most of you who are bigger brands are already doing this.

But it kills me when I walk into a small business still to this day, and

see these big signs on the front counter that say like us on Facebook.

You know it doesn't really do you that much good these days.

Less than 2% of your page posts are actually going to be seen by your fans.

Meanwhile shared posts first of all get a lot of, get a lot

of visibility in people's newsfeeds among their friends and their friend circles.

So I think you're going to get, A,

that shared post much more visible on the initial share.

And, b, the shared post have this permanent life

in the kind of ad link ecosystem that Facebook is building.

so if I do this search best beer in Portland today in Facebook.

The number one result that pops up for me is this post by Ben Lloyd that he shared

about an article, wasn't even about the business itself.

It wasn't from the business itself,

an article that mentioned a friend of his' business Produce Row.

And so now Produce Row is at the top of my search results for best beer in Portland.

So, encourage your Fans encourage your audience to share your stuff much more

than like.

That's what's gonna give it,

I think will, this kind of permanent or longevity in Facebook search results.

These are basically plus ones that mean something.

And for those of you who are really old school SCOs you might remember

a website a product called cuil.

I think Facebook is kind of building what Cuil's idea was.

Cuil was a little bit ahead of its time.

But essentially Facebook has this amazing index of curated links

that your own social network has shared with you and with their circles.

This is completely independent of any kind

of link data other than that initial share, right, and so.

For us, as marketers, I think we should be really excited about this.

We don't have to worry about, oh, should we ask for this anchor text, do we pay for

this link, how much is that worth.

Is that going to get us in trouble with Google?

Who freaking cares.

Get people to, first of all, write great content.

Get people to share that content.

And that's what's gonna give you long term search presence in Facebook.

And Marshall pointed out yesterday in his presentation, sometimes that content

itself is paid, right, like Cathay Pacific article on the New York Times.

That was New York Times' most popular post.

There's a lot of opportunities here I think,

you know, in an algorithm that is not dependent on link relationships.

Number two, so in addition to making sure that your app is incredibly

indexable using the protocols I think Marshal and Cindy mentioned yesterday.

I would encourage you to drive as many app authentications of

people's Facebook accounts as possible,

from the initial wired coverage of the graph search launch you can see.

Another big agenda item is assimilating the mass amounts of data

generated by third party applications.

It's very clear that this is becoming a thing, not only for Facebook, but Google

now on tap is very much after the same kind of activity, the same kind of data.

And so I think that the extent, again, that your app can be indexable, and

that Facebook can see the activity of its users in your app

will probably help you in the long term as we see the mobile ecosystem evolving.

Okay, 90 seconds left for again, my favorite topic, local search.

Facebook did, basically, ignore local search in its last update in December.

Greg Sterling was not the only one to lament this.

Basically all of us in the local search community, we've kind of been waiting for

Facebook to make a real play in local.

But the reality is, they already have a pretty good set of results.

And this is maybe why it's so frustrating for us.

If you do a search, even a very long-tail search,

bankruptcy lawyers in my home town, Decatur, Illinois,

you get a pretty decent looking search result page, the one there on the left.

And golf courses in Portland, bigger metro area, sort of less long-tail query,

you get a great list of search results.

Those really are kind of the top public courses in Portland.

And what's very clear about this is that check ins and reviews are in

large measure, driving a big part of this algorithm, in terms of what's showing up.

And so I think, regardless of when or exactly how Facebook releases a local

search app, or embeds more local search results in standard search

results I think driving check ins and reviews now, is a great idea.

In fact, this is a 2012 slide here on the left, 50% of people way back in 2012,

we already actively looking for local businesses on social networks.

I think this behavior is happening More and more and more than we realize.

And in fact Mike Blumenthal did a study last year, that Facebook was the number

two preferred destination for people to leave reviews of local businesses.

So, people are already doing a lot of this behavior, make sure they know about it,

give them a call to action to do it on behalf of your business.

And in fact it can even help you get more attractive search results in Google now

that Facebook reviews show up as rich snippets in Google search results.

Facebook is getting much more serious about local I think,

it's pretty clear there's been a couple of additional innovations that have happened.

They basically replicated Google's right to call.

They're giving out free beacons to any SMB in the world who wants them.

I think we're going to see major movement from Facebook and

local in the next nine months to a year.

So, just to kind of sum up, promote sharing, not liking, not only for

immediate news feed visibility, but also for permanent archival search visibility.

Connect your apps, make sure they're indexable.

And promote, if you're a local business,

promote reviews and check-ins to stay ahead of the curve.

So I'm just gonna leave this up, I'm basically out of time and out of slides.

But this is a great post by Simon Penson that I mentioned earlier.

Might give you even more ideas on how to use graph search to build your business.

So hopefully I've convinced you, start paying more attention to Facebook.

Search, I think it's going to be really big and thanks very much for listening.

[APPLAUSE] >> Okay, so a year from now we're going

to be saying David Nim told us, about- >> Or you're going to be saying man,

that was a waste of 35 minutes.

>> [LAUGH] >> But hey we'll see,

I'll be back next year.

>> Okay, we're a little over time, so just a couple of questions here.

One on Twitter, James Loomstein, thank you for the tweet.

If I do Facebook graph search, are the results personalized to me or unfiltered?

>> Yes. >> Not signed in.

>> Yes. You have to be signed in as a person in

order to do these queries, so all these are going to be personalized.

However, basically, there's a pretty clear demarcation

in terms of people that you know, and then sort of generic results below that it.

So yes, the results are personalized.

But it can still be useful to find people outside your network.

>> Okay, a couple of people were asking, Susan one of them, can

you segment the audience from these search queries into custom audience for ads?

I didn't see that you could do that, but you would have to think Facebook would

want you to, so either they're gonna build this functionality into their ad platform,

or make that a sort of drag and drop, similar to AdWords.

>> That might be a question we follow up with Marty on-

>> [CROSSTALK] Yeah exactly, and

that is a great question for Marty.

>> Okay, and finally I had a question.

So you talked about finding influencers.

Finding people to leave reviews on Yelp, I thought that was brilliant.

What is the next step for actually reaching out to those people?

Do you message them within the Facebook app?

Or what's the- >> Sure it sort of depends on

the business I think.

Personally, I'd love to get a Facebook message from my favorite coffee shop, or

if the owner of Breakside reached out and said,

hey you want a free pint next time you come in?

I think it depends on the business.

You may be not that cool if you're in the insurance world or something like that.

I would just be discreet and act like you would in the real world.

That's kinda my best advice for anybody, so.

>> That's awesome. >> Yep, cool.

>> David thank you so much for stopping by again.

>> Thanks Cyrus. All right.

Hug? All right.

>> [APPLAUSE]