AMA: Kevin Indig

AMA: Chad Mizee
August 10, 2018
AMA: Hugh Forrest
September 24, 2018
The Kevin Indig AMA is happening on 13th February 2019 at 10:00AM PST (1:00PM EST)
 
Kevin Indig has helped companies acquire +100M users over the last 10 years. He currently runs technical SEO @ Atlassian and mentors startups in and outside the German Accelerator in scaling growth. Companies Kevin worked with include brands like eBay, Bosch, Samsung, Dailymotion, Pinterest, Columbia, UBS and many others.
If you’re not an OG yet feel free to join us.

 

Kevin Indig is Head of SEO at Atlassian, Former Director of SEO at Dailymotion.

The AMA with Kevin was held on February 13th, 2019. This transcript has been edited for punctuation, grammar, etc.

 

Hey y’all! Thanks for tuning in. I’m a bit under the weather but that doesn’t stop me from answering your questions. Feel free to ask me anything.

Welcome! What is one common mistake that you think smaller firms/startups make in the SEO space?

I actually wrote an article about that! ->

https://www.kevin-indig.com/seo-mistakes-young-startups-make-every-time/

But to summarize it: it’s often strategic decisions like putting a blog on a subdomain, or focus on vanity metrics like traffic without taking bottom line into account.

What are some suggestions you have for someone trying to further their skill sets in SEO today?

Be a practitioner as much as you can. Create projects that you can try things out on. Ask experts in different areas on Twitter or per email about advice. Aaaaand: write your lessons down. Question them.

Wondering if you have any insight in how to influence sitelinks for your site – like how do you get Google to show them to begin with (the nice big ones I mean), and how can you impact which pages show?

After my experience, they are very much derived from internal linking. Check what the most internally linked pages on search console are. Often, you find a big overlap between sitelinks and internal links. To be fair, Google often messes ’emup.

How do you find ideas for new SEO based projects, and prioritize the projects you and your team will work on?

Love that question! I cluster projects into three categories: expansion (how to generate net new?), solidify (where to optimize to reach top3 positions?), maintain (where to rework/edit content and do tech SEO hygiene). We try to take care of one project for each of those categories per quarter.

What’s the first thing you look at to gauge the level of SEO effort on an existing site you begin working on?

There is a maturity model I create for every site depending on their content, links, and technical structure. It also very much depends on the competitive landscape, meaning where competitorsare strong.

Wondering what SEO tools you like best?

I like AHREFs, Screaming Frog, SEMrush, Chrome dev console, and a couple of smaller SEO tools.

Have any experience with SEO agencies? If so, what’s the best way to leverage that relationship?

Yes, I started on the agency side! I learned a ton during that time. In terms of steering/controlling agencies, that’s a huuuuge topic. I like it when the agency doesn’t need a full-time role to be controlled, but that very much depends on how long you work with the agency and what they handle. I can recognize a good agency quickly by how they plan and report their work. Does that answer your question?

Did you do metadata snippet split testing at Atlassian? Which hypotheses were most often proved right?

We never tested the meta-snippet, but title, h1, and such drivers. The honest answer is that we didn’t see meta-descriptions move the needle much. Only in edge cases.

Was alluding to testing `<title>`, etc. any metadata not just descriptions – sorry if not clear.

Yes, we did some of that stuff, and in most cases the common assumptions were proven right. If you read AJ Kohn’s new piece on “query syntax,” we did some of that stuff and found it to be true! Really interesting, mind-blowing stuff, man.

Do you think Google TIPR classifies links based on whether they’re template links vs main content links?

Yes, I think there’s a difference, but it’s hard to say what _exactly_ the difference is. For tipper, it makes the model unnecessarily complicated, as of my experience. But happy to be convinced!

Is there such a thing as real, “low hanging fruit” for SEO? Obviously—real results take time. But are there some baseline, semi-universal tactics you can take to get *some* proof of concept for less-enthused stakeholders—quickish?

Yes, optimizing internal linking is something I’ve seen to yield huge results, and then the basics: xml sitemaps, title tag optimization, backlinks.

The usefulness of schema is still questionable and we don’t see much documentation of Google about where exactly it adds an extra edge. Do you suggest still marking up?

I very much agree that the verdict is still out. I like the idea that schema is like a dictionary for entities on your site, but that takes the question a bit too far. I think for known rich snippets it makes a ton of sense, but do you need to markup everything? I think that’s to be decided case by case.

Do you think it’s better to launch projects on different domains or test them out on a sub-domain or sub-directory?

I think it depends on what you want to do with the project. To test things out, I suggest a new, clean domain. What’s the purpose of the project?

What tools do you recommend to look at page speed in terms of how Google will evaluate your site? I’ve seen people advocating Google Page Speed Insights, but results seem to vary quite a bit from other tools.

I love Lighthouse, webpagetest.org, and the Google page speed tool. But honestly, I find Lighthouse to be the best.

What’s the best SEO strategy for some startup that begin into the market?

Identify the most relevant pain points or jobs-to-be-done of your target group and create very targeted content around that. Don’t spread yourself too thin by addressing too many topics. Stay focused and become an expert in one topic. That’s what works for most saas, enterprise, and app startups.

For ecommerce, social networks, and publishers, I recommend to focus on scaling and optimizing page templates.

Newer Schema like HowTo or QAPage – how much do you focus on implementing new markup as tactic for getting featured snip. or other search features?

Depends on how easy it is to implement. If there are enough resources and short shipping cycles, I’m always for trying it out.

How often do you edit the on-page content of pages that are ranking #1-3? Seeing as everyone below you is gunning for your spot, daily.

It depends on how much volatility the SERPs have for the main queries the page is ranking for. If there’s high volatility, it often means Google hasn’t found the right fit. In that case, I make adjustments on a monthly basis.

Question around back-linking –what approaches to securing high quality backlinks have you found most effective?

My top 3: building personal relationships, link reclamation, and creating statistics and reports.

Do you have any advice for reducing flux when in position 1 but keep dropping and then returning?

No, it often signals that Google hasn’t found the right mix or fit for that SERP. It’s hard to avoid, but going back to user intent and thinking how to satisfy it better is often a good thing to do.

How would you rank a page like “Employee Management Software for {business type},” where the business type can be anything from nursing home, to restaurant, to gym? If the page is going to look like a variation of the home page (obviously with text and images changed up to target the specific business), there won’t be much content on that page (text wise). It would look weird to have an article on that page, so how would you go about ranking a page like that?

I’d create topic clusters around that topic and link back to the landing page(s). I’d especially focus on targeting long-tail queries.

What’s the SEO “myth” that you hope would die?

Many! But my favorite is that all you need is backlinks. It’s simply not true. Look at SERPs for highly competitive queries and look at the backlink profile of well ranking pages.

Question about AI and internal links:If you had to layout the architecture of a new (WordPress) blog, a very big one, that’s going to cover a lot of ground (countries, states, places and tourism-related stuff), do you have any general guidelines on how you would approach this task? Regarding things like the use of mega menus, siloing, specific sidebars for countries, the use of tags, interlinking, etc…

A sound and well-planned taxonomy is worth its weight in gold (if it weighed anything). I would structure a new taxonomy around the most important entities and reflect that in content planning, internal linking, and URL structure. I would also try to structure the taxonomy in terms of logic and research that goes beyond sheer search volume.

How do you see people optimizing more for “Voice Search” + what do you think of the new w3 standards being developed around voice site maps?

I think it’s a thing that’s coming in the future but only for certain use-cases. I think focusing on featured snippets is enough for right now. I wouldn’t stress about it any further (for now) and focus on doing the basics better instead. What do you think?

What do you use to implement Schema markup at scale? Working on a site with 1400 URLs with no schema and need to implement for all current pages and need a process for future pages.

Have you considered tag manager?

Tools you use on a daily basis? Not only commercial SEO tools but, custom tools or scripts, project management tools, etc.I’m especially interested in tools to help me visualize a site’s structure and its internal linking / hub pages, etc. For small to very big sites. Screaming frog? Sitebulb? some DIY stuff with KNIME or something else?

TBH, I don’t use much custom stuff. I love to use frameworks and mental models for managing SEO and thinking about it in certain ways but tool-wise, I think I’m pretty boring. Looker and Tableau are awesome for visualization. I also really enjoy Gephi, to mention a free tool.

Do you think cadence of content changing on a page is a consideration…e.g. if your page compares deals, Google may look for both expected brands and deals to change frequently?

Yes! I very much think that and saw that myself. Freshness goes beyond just queries that “deserve freshness.” Content that’s up to date is super important!

About ”optimizing internal linking” any effective strategy for doing that? Would you just use main KW as anchor text, mix it up (50% exact, 20% partial etc.) or do you feel you can be as aggressive as you want here? Not like external links.

I wrote an article about the “TIPR model” that captures most of what you ask for. As for internal anchor text, I’d try to use the most common query variations of the page I’m linking to. I think there’s no internal link penalty, unless to try to link certain letters to certain pages. Someone really did that, no joke!

You’ve had lots of experience working with remote-based employees through the likes of Atlassian. Has it been harder to get buy-in when dealing with remote staff, when compared to other in-house gigs?And any advice for other SEOs working with other remote staff?

I think it’s a person by person situation and hard to generalize. I grew up in Germany, and Germans focus on input, i.e. how many hours a day do you work? I don’t believe in that. I believe in output. As long as the results are coming in and the work gets done, I don’t care what people are doing. I think using rituals, like video lunches or meeting for a casual chat on slack at x PM, creates a digital culture and helps to build rapport (as good as possible). It’s super important to have and share very clear expectations. Plus, I like creating a manual (written or presentation) that explains to others how you like to work.

Is there any point where you’d advise that SEO just doesn’t make sense for a very niche/obscure subject and search terms? I’m currently doing a project for a machine learning platform, but it’s for a very, very narrow field. I’m not getting any traffic (obvs.), but should we keep plugging away? Upside is a ~50k product purchase, but it just seems like blood from a stone.

Yes! I think there are lots of cases in which SEO isn’t the right channel. For such an expensive product it makes sense to invest in paid search and other paid channels. You want to keep your CAC < LTV (customer acquisition cost; lifetime value) but as long as your CAC is below 1/3 of LTV, it’s fine. Does that answer your question?

What sort of tools and/or processes do you use for SEO project management?

I like Confluence a lot but also enjoy working with Trello, Asana, and G Docs. To me, I think the system you follow is more important than the tools.

Do you have any workflows/tools for blocking erroneous code from deploying or for monitoring breakages so they alert when something is wrong before it effects results?

ContentKing helps a ton with that! Really love the tool and company. Good folks.

Hey everyone, thanks a ton for your questions. That was great, and I had tons of fun! If anything’s unanswered, please let me know. I’ll answer each question!

To join these AMAs, enter the #AMA discussion channel in the Slack group at the specified date and time.