Mozcon 2017: Cindy Krum

Cindy Krum at Mozcon 2017

DISCLAIMER: This post is written as a live blog from Mozcon. There may be typos and grammar to make my high school English teachers weep. Please excuse those … it’s a fast-paced conference with back-to-back sessions and no time for proofing or even proper writing.

Opening up day three of Mozcon 2017 is Cindy Krum to chat mobile.  She starts the chat with mobile-first indexing and mentions it launching in 2018 though mentioning that individual sites are already being switched over. She points out that since it was pre-announced it’s a big deal.

She points out that we’re all talking about mobile-friendliness but that already happened and is #FakeNews now.  They are not the same thing.  What we can expect in the new update which is … let’s break it down:

Mobile – mobile means anything that’s not a desktop or laptop. To Cindy we need to sub the word portable for mobile. That is – content that is built for portable devices.

First – well, it’s first.

Indexing – indexed are ways of organizing similar types of data for easy finding. So when wetalk about indexing we’re not talking about ranking.

She points out – an index is a list of sites that can rank.

“Portable, prioritized order of information for ranking.” – this is how we need to look at it.

Tech adoption

Technologies grow traditionally following the same path.  During the adoption phase others are working on incumbent technologies and innovations so technology jumps can seem quick but it because builds n somethign that exists.  For example hummingbird ties to rankbrain and voice search.  She notes, “Things like voice search, hummingbird & rankbrain make a larger story about semantic understanding of queries & responses”

Next she’s talking about the growth in voice search. It’s experiencing HUGE growth.

Chatbots are growing faster than apps.  She mentioned that chatbots help get content into indexes.

She next brings up Chromecast as a game changer.  I never thought about it but the bride it builds between devices changes the landscape. It ties to conclusions I’d made in a story I wrote on the death of organic search and her isolation of Chromecast is very interesting.

We know it’s important to Google as it’s being built into the base Android technologies connecting devices and leading to the connected home.

Next we’re talking about Google I/O and the talk about machine learning, voice search and how it’s front-and-center.

Cindy point out that we don’t have a name for the world.  I’s mobile-first, it’s AI first – in the end Google wants you to put your stuff on the cloud to access from anywhere in any form.  In the cloud it’s predictable to serve to different devices and convenient for users.

Takeaway – as webmasters we need to separate content from design. We need to make content easy for Google to serve anywhere on any device. Content that’s easily extractable. Google believes that it need o get everything but crawling has lost it’s appeal as there’s too much. Google want’s to index the worlds information but even things like a Fitbit are creating it too fast for crawling. Thus – Google needs s to store it on their servers, via XML feeds, via SON-LD, Plus-ins or APIS. The content is easier to extract in this format.

Cindy predicts that crawlers will soon be reduced to content verification but not discovery as it’s too inefficient.

Cindy’s Speculation

Mobile-first indexing is REALLY about cloud-first indexing (URL’s are not required). With all the new devices, many without browsers or with tiny screens URLs are irrelevant.

She points out featured images and going through te pages the URL isn’t changing, just the data being presented. She’s also pointing out videos, audio, etc. all being pulled from the knowledge graph that’s available but not shareable because it doesn’t have a URL … it just lives in Google. She uses a search for “cow sound” as the example. It just produces a sound as a result. Same for “play c-sharp” or translation.  There is only one, non-URL result and second place gets nothing in voice search.

How to win …

Feeds and XML get position zero.  Google can’t crawl fast enough to get them sports scores in realtime … they need feeds. Back to firebase which she talked about last year.

She notes it’s basically hosting app content and that helps them not have to crawl as they just have it.  Hosting the content also gives them engagement depth metrics – they know what of your content is more interesting to users.

Now she’s chatting AMP … we give content to Google to host. Even answers they crawl but to speed things up they host.

Google is pushing us to JOSN. It gets them info without having to crawl content.

PWAs (Progressive Web Apps) – they are like plugins that open their own browser window and restyle it.  They lean on the browser code so they’re very fast.  It’s like an app within the browser.  They cache locally anything you’ve done before so you can work offline.  A service worker (index.db) is an API for Google’s crawler so all they’re getting is the content … they don’t care about the design.

We know Google cares about it because they’re turning their Google Now content into PWAs. You can now add web content to your home screen serves by Google like top restaurants, sports, etc.  Basically spinning off knowledge graph into in PWAs.

Now she’s talking about Google connecting app data based on information it’s collecting and making it accessible across devices.

2 weeks ago Google announced that you’ll be able to download apps directly from the search pages cutting down on friction making app store optimization more important.  We’re cutting out websites and the apps don’t need URLs.

Apple iMessage is doing something similar.

The big new thing that illustrates the point is Google Actions.  You need to understand Google Assistant.  It lives in the Home, phone, etc. and they want to get it everywhere.   What they give out as replies are often retrieved from feeds for a lot of query types.   They’re using questions below the reply to allow you to learn more information and helping teach them what your next query would be.

Opentable is an example.  If you Google “make a reservation at rootdown” it’ll take you through the process in position zero in the app or cloud.  You just need to be logged in you don’t need to open the app.  This is mobile-first indexing.  This is portable-first indexing.

She talking about capturing information about your day across devices mapping out your day. It’s a creepy tech we all encounter on android but only because it’s reminding us how much they know about us.

This is all via the assistant so it can help make your data findable from anywhere.

Now she’s talking about cloud-first or feed-first programming. Now you need to be more clever.

Now she’s talking about small businesses using chatbots and that we have automated responces where we can define the query vs response getting human understanding about their data.

You can download her slide deck here.