Leave a comment
Get the GH Bookmarklet

AMAs

Amelia is a storyteller, masquerading as a marketer. She currently leads the content and brand team at Mixpanel. In her two years at Mixpanel, she has built and led teams who have exponentially grown blog and website traffic, increased blog subscribers by 25x, and launched industry-leading research, like the 2017 Product Benchmarks Report.     

Before Mixpanel, she was the first marketing hire at Inkling, working in both B2C and B2B. She arrived in Silicon Valley several years ago by way of Bangkok, cutting her marketing teeth for a couple of years at a boutique branding agency. 

Amelia has a B.A. from Princeton in English Literature (which she contends is a wildly underrated degree for a career in tech). Outside of the office, if she’s not reading, running or indulging in wildly overpriced pour-over coffee, she’s likely up in the Russian River Valley with her partner, fixing up their 100-year old cabin.  

But enough about her! Check out Mixpanel’s publication The Signal. With more than 50,000 thousand subscribers from product, marketing, and analytics, The Signal is a go-to resource for anyone who wants to learn how to build great products and businesses with data. 

She will be live on Jan 18 starting at 930 AM for one and a half hours during which she will answer as many questions as possible.

  • PH

    Pradyut Hande

    3 months ago #

    Glad to have you joining us, Amelia!

    For SaaS companies in the mobile marketing sphere looking to build a global presence, brand messaging is key. However, different markets react differently to certain "industry buzz words". For instance, "marketing automation" might find resonance in one market, while "user engagement and retention" might resonate in another market. In such a scenario, do you recommend a uniform brand messaging or messaging customised to specific markets?

    • AS

      Amelia Salyers

      3 months ago #

      Can I be lame and say “both”? Ideally, you have a messaging framework that starts from a singular mission and positioning statement -- since what you actually deliver to each market is probably not that different -- and then you customize underneath that umbrella with the terms and phrases and ideas that research (customer interviews, surveys, desk research) tells you resonate with a particular audience. Ultimately, whether you call it “marketing automation” or “user engagement and retention”, you’re still selling the same thing -- and the marketer in Mumbai probably has many of the same needs as the marketer in New York or London.

      It’s not that different from adapting messaging for different personas, right? Except, instead of a persona (the marketer, the IT guy, the working mom, etc), you’re looking at a market.

      3 Share
      • PH

        Pradyut Hande

        3 months ago #

        This definitely offers me some clarity moving forward.
        Thank you for your detailed response!

  • TW

    Tom Whatley

    3 months ago #

    Thanks for joining us, Amelia!

    Could you give us a quick snapshot on your content distribution strategy? How do you "bake" promotion into the content creation process? Are there channels/approaches you've found are more effective than others?

    • AS

      Amelia Salyers

      3 months ago #

      Ha, not sure I can answer this without a massive essay, but here goes … To begin, I always try to start from first principles: what is the goal we’re trying to accomplish that content and stories can help us with? Are we trying to build awareness and buzz with a certain audience? Are we trying to own a certain term or idea out on the interwebs? Are we trying to get traffic that converts into leads? Of course, most folks will probably say “All three, please!”, but I’ve found that, at least when starting out, it can be very difficult to do more than one or two things well.

      Understanding that goal will then drive what kind of content you create -- and that, in turn, drives the promotion and distribution strategy. For example, a blog or publication is great if you’re trying to get traffic and conversions over the medium to long term. This is because blogs are great for building a specific audience and being top-of-mind when someone decides it’s time to solve that problem you’ve been telling them about. Blogs are less great when you’re pre-product market fit and/or not sure who your target audience or buyer is, exactly. In that case, maybe using a major report or controversial stance on something to get PR is better, since you can see quickly who is interested in and who isn’t.

      The other factor I like to consider is any key “homefield advantages” you might have at your company. These are generally pretty unique to you and hard to replicate from company-to-company. It could be things like a well-known, in-house expert or original data or an amazing customer willing to go on the record.

      For example, when I joined Mixpanel a couple of years ago, we had a massive customer email list, but we almost never used it, except for the occasional product announcement. In my third week, I persuaded our CEO to let me test sending out an email to this list promoting a really stellar blog article, because I believed it would really resonate with that audience. The results were conclusive: we double the blog’s monthly visits in a few hours and social sharing took off, while unsubscribes were very low (aka, folks didn’t feel spammed). That customer list became a secret weapon for awhile, as we found other distribution channels.

      Another advantage for us has been working with influencers in the space -- once a few folks took a chance on being featured on our blog, it got easier and easier to approach others, because each new name became a calling card … and we started to get inbound recommendations, too. The obvious advantage of influencers is their network and additional distribution -- but you have to find ways to give them something valuable in return (which I could write another essay on).
      All of which is to say: the channels will change based on your company, industry and goals. Ultimately, the more thought you can put into your marketing strategy and company goals upfront, the more likely you’ll be successful with your content. Seems rudimentary, of course, but I think it bears repeating. :-)

      5 Share
      • AA

        Anuj Adhiya

        3 months ago #

        +1 for more responses starting with "Not sure I can answer this without a massive essay" :D

        4 Share
      • TW

        Tom Whatley

        3 months ago #

        This is gold. Thanks so much for putting all this together.

        For anyone wondering how to give influencers something valuable return, I have some insight on how we do this at Grizzle:

        Get them involved in content you create for other people's audiences (OPA). Guest posts, for example, on industry blogs. Interview them for insights on a particular topic and turn that into a story. You get their name in front of a wider audience at the same time as yours.

        Ahem... I'll stop stealing the mic now. Amelia, thank you again!

  • NM

    Niel Malhotra

    3 months ago #

    I've got two questions all about the content creation process. I hope you don't mind if I lump them together:

    How do you determine what topics to write about?
    How do you promote your content?

    • AS

      Amelia Salyers

      3 months ago #

      I addressed this somewhat in Tom’s question, but I want to be a bit more specific. Determining topics and promoting the content go hand-in-hand. You can ask my team: whenever they pitch an idea or topic to me, my first question is usually, “So, how’s it going to get read/seen/distributed and help us hit our growth goals?” If there’s a not a clear answer to that question, then we should think very carefully about doing it.

      On the flip side, I see a lot of very crappy (can I use that word here?) content out there that’s driven purely by a promotion tactic -- I’m looking at you, SEO-keyword salads and clickbaity “I increased our blog traffic by 300x overnight with my eyes closed -- and you can, too!” articles. One of the things we learned over time was that optimizing content for the top of places like Hacker News, while fun (and terrifying, because, ooo boy, the comments) because of the dramatic spike in traffic, was ultimately bad for our conversion rate to subscribers and overall engagement.
      When it comes to our content, we generally try to have a balanced diet of content. What does that mean? It means I’m looking for a mix of “sugar spikes” or hits-driven pieces and a “fiber” or compounding, sustainable growth pieces. The sugar spikes -- the ones that go viral on Hacker News or GrowthHackers or on Twitter -- generally are emotionally-driven (ex: “Why most A/B tests give you bullsh*t results” or “Why this former director of Product thinks you shouldn’t be a manager”) or tasty bits of data (“Top email subject lines for open rates”). These help you find new audiences and get buzz, which hopefully you can harvest with CTAs on your content and site, but they have massive decay curves.

      The fiber pieces are the ones that slowly build traffic over time. These tend to be more educational, answering questions people have. One of our best-performing articles over the last year is simple: “What’s the difference between data science and statistics?” It didn’t get us loads of attention initially, but it answers an intellectual need for our audience.

      By having this diversified mix, we can build both short and long term growth.

      2 Share
  • JP

    John Phamvan

    3 months ago #

    Hey Amelia!
    If Mixpanel were just starting up their blog today, what would you do differently, if anything (and why)?

    • AS

      Amelia Salyers

      3 months ago #

      This is an interesting one, because the Mixpanel of today is actually quite different than the Mixpanel of two years ago or four years ago and so on. Plus, the landscape for content and storytelling is always shifting.

      All of that is to say, if I had to start the blog again today, I’m not entirely sure I would even begin our content efforts with a blog! Frankly, we were a little late to the content game, and if we were starting today, I’m not entirely convinced that a blog would be the best, first way to start sharing our story. Going back to the idea of goals and audience, I’d probably consider starting with a podcast or a series of guest posts on other sites or places like LinkedIn, instead, since those avenues are less crowded (though probably not for long!) and I could learn about the audience a lot more quickly than with a blog.

      If we did create a blog again, I would do one thing again (and still do to this day) for sure: develop a forecast model for growth and conversions to set weekly and monthly growth goals. I was very lucky to work with one of our analysts on an initial forecast when I started at Mixpanel, and it was hugely important to have that forecast model in hand to share with our leadership team what we could expect from the blog, given certain resources and certain experimental assumptions.

      4 Share
      • JP

        John Phamvan

        3 months ago #

        This is great!
        Would love to hear more about how to go about creating that forecast model

  • GH

    Glen Harper

    3 months ago #

    Thank you for joining us today, Amelia.

    How do you quantify the ROI of your content marketing efforts?
    In other words, can you tell what the approximate dollar impact is?
    What tools/processes do you have in place for you to be able to get such data?

    • AS

      Amelia Salyers

      3 months ago #

      Aha, the old ROI question. I’ll be honest: this is something we’re actually working on as a marketing team right now, to build something more sophisticated than just last-touch or first-touch attribution. Our stack starts with Mixpanel (duh) and then includes some of the usual suspects, like Marketo and Salesforce. I’m no expert in multi-touch attribution, though, so I’ll have to get our head of demand gen on here to go into more detail. :-)

      I can say this: different content channels have different measurements, but ultimately, as part of a B2B marketing team, I’m accountable to helping grow our sales pipeline. Sometimes, that is directly measureable, e.g. this person downloaded a report and then became an opportunity, and sometimes it’s indirect, e.g. we saw a large increased in direct traffic when this awareness campaign ran, leading to more website conversions to “Contact Sales”. One of the trickiest pieces of enterprise marketing is that you have to both find the buyer/decision maker -- aka the person who connects with Sales initially -- AND you need to influence all of the other people on her team who will be users of your software or approvers of the purchase … and getting attribution for those folks can be tricky.

      In general, though, we think about leading and lagging metrics, or paired metrics. The leading metric is one that we can generally measure pretty quickly and is a great indicator of initial traction. This might be traffic/unique readers for the blog or number of downloads for a data report or ranking for an SEO page. The lagging metric is one that ensures we’re getting the right kind of traction, but often it takes more time. Most of the time, my lagging metric is pipeline creation -- of the people who downloaded this report, how many of them are associated with an opportunity in 3 months? 6 months?

  • TN

    Tri Nguyen

    3 months ago #

    Hey Amelia

    How are you all organized around demand generation? Who focuses on what, and what do your KPIs look like?
    How does content feed into demand gen at Mixpanel?

    • AS

      Amelia Salyers

      3 months ago #

      We work very closely with demand gen, especially when it comes to the website. In general, we think of the content and brand team as focusing on organic, owned and some earned media, while demand generation is more focused on paid media. This means in practice we’re running influencer marketing campaigns and SEO campaigns and brand awareness efforts, while demand gen is buying ads, syndicating content, running email campaigns, etc.

      Of course, those lines can get blurry sometimes, since a single campaign will have multiple components, and they all influence each other -- for example, I’m sure others have seen the “halo effect” that occurs when you run a big paid media campaign and see an uptick in direct traffic to your homepage or site. At the end of the day, the customer on the other end of our marketing efforts doesn’t really care if something came from “demand gen” or “content” -- they just want their questions answered and their needs met! So, we build integrated marketing plans every year, together with product marketing, to ensure that the end customer has a seamless experience.

      2 Share
  • MD

    Mark Anthony de Jesus

    3 months ago #

    It's always a challenge balancing targeting keywords and writing actual, quality, shareable content.
    How closely do you work with an SEO (if at all) and how much does (or should that) affect your content?

    • AS

      Amelia Salyers

      3 months ago #

      So, the way I like to think about SEO when it comes to content strategy is as a very valuable input, not the end-all-be-all for our strategy. Ultimately, what people search for online is important not because of rankings, but because it gives us really valuable insight into what folks care about and what they’re trying to learn about.

      To that end, we do a few things. First, we have tactical, SEO-specific content that’s designed to rank highly for certain long-tail key terms. I am very allergic to bad writing, so we work hard to write great pieces that truly answer the question behind the search box. We know if it’s working, of course, when pieces start to rise through the ranks and drives traffic.

      Second, we look at keywords and Google Trends as one line of research for one kinds of topics people might be interested in now and over time. But, we also look at popular posts on other sites, talk to customers and our customer-facing team, and work with product marketing to understand the pain points that our audience faces. Sometimes, if we do our research well and get a bit lucky, we can anticipate what a popular term might be in several months, and therefore build content that ranks for it in advance.

      3 Share
  • JF

    Javier Feldman

    3 months ago #

    Hola, Amelia!
    What was the impetus behind doing the Product Benchmarks Report?
    How much work did it entail to put that together (if you had to approximate human hours).
    What were your biggest learnings from putting it together and distributing it?

    • AS

      Amelia Salyers

      3 months ago #

      This was one of my very favorite things we did last year. As a quick bit of context, the Product Benchmarks Report is an analysis we did on product usage from aggregated user behavior data from 1.3 billion end-users around the globe, resulting in a set of benchmarks for different industries in usage, retention, engagement, and conversion.

      The impetus behind the project was simple. We kept getting a question from customers that they (reasonably) thought we could answer, “What is a standard or average result for [xyz KPI] in my industry?” When we started looking around, we realized that no one had that kind of information out there, so we could tell there was a major opportunity for us to satisfy a customer desire with original research. Plus, our mission is to “help the world learn from its data”, so this felt extremely well-aligned.

      In terms of work, we moved relatively quickly, largely because we had an amazing internal cross-functional team. From the kickoff meeting for the project to launch, it took a little more than three months, which is now astonishing to me, given how much data we analyzed. The biggest lift was in cleaning up 50 billion events and getting them into a Mixpanel project, such that it would be easy for us to do analysis.

      We learned a TON in the process, and I could go on all day about it. Some highlights: 1) while painful at times, it was worth it to be extremely thorough in how we processed and cleaned up the data. Why? Folks took the data very seriously and were extremely thoughtful and creative in how they wanted to apply the learnings (which surprised me), so I felt really confident that we were giving them statistically significant information to work with. 2) Influencer marketing was a much better distribution mechanism for us than traditional press, which was pretty surprising, since original data is a tried-and-true press strategy. 3) A strong design and editorial POV was really important to making all of this data digestible and actionable, and I’m glad we spent lots of time on those pieces.

      We’re working on more reports like this, so we’ll be putting all of these learnings to good use -- and then “learning” more things along the way.

      3 Share
  • AA

    Anuj Adhiya

    3 months ago #

    Hey Amelia - so cool to have you on!

    Why is the content and brand team one at Mixpanel?
    I've normally seen these be separate functions?
    What are the advantages of having both report to you?
    Also interested in hearing about any challenges, if any, of having the team structured in this way.

    • AS

      Amelia Salyers

      3 months ago #

      Ooo, good question. Actually, my team includes content, the website (a major brand asset), and customer marketing. What’s the one thing that all of those have in common? Stories. From the ways we try to educate the market about analytics broadly through content and stories to a website that educates our audience on Mixpanel specifically to the stories our customers share on how Mixpanel has helped them achieve their goals, they are each part of the same journey. By having content and brand under one little roof, we’re trying to ensure a consistent thread of storytelling throughout our marketing programs.

      In terms of challenges, this is actually relatively new for us, so we’ll see! So far, the main hurdle is one of making sure that content and brand both serve the needs of the demand gen and PMM teams, as well as drive organic growth and pipeline on its own.

      Now, I’ve definitely seen content and customer marketing live under demand gen or product marketing in other worlds, as well as being part of Brand or corporate marketing. I suspect it depends on which function got developed first and the kind of talent/skills you have in your organization. Ultimately, I don’t believe there is any one answer to this question -- and folks who tell you otherwise are probably just selling you something/a book.

  • DH

    Dani Hart

    3 months ago #

    Hi Amelia - so excited to have you here!

    You have experience marketing in the East and the West.
    Are there some fundamental differences between how to approach marketing in different parts of the world?
    Are there some things that work, say in the US, that will never work in Asian countries and vice versa?

    Also, if you're lucky enough to have a global audience, where do you draw the line between being (for lack of better words) "generic" and "specific" with respect to geography?

    • AS

      Amelia Salyers

      3 months ago #

      First, I think there are, of course, often many cultural markers and taboos that folks need to be careful and respectful of. For example, when I worked in Thailand, we had to be careful of certain colors and verbiage if we wanted to build a hospitality brand, let's say, that resonated in Thailand, India, and China. I'm far from an expert on this, but suffice to say, it pays do your research to avoid decimating brand equity with a cultural faux pas that gets plastered all over your collateral and site, etc.

      However, I also believe that people are people! As marketers, we often spend a lot of time on parsing our audience into smaller and smaller segments, in order to personalize our marketing or better answer their needs, but I think we can forget that there is a large Venn diagram of overlap in what people need from our products and services. For example, for us, a PM in India or China does have some different needs from a PM in the US (oftentimes, it's about how much more data apps in India and China have to handle), but all of those PMs ultimately care about retention, engagement and growth. Especially when your company or team is small, I'd focus more on the commonalities and overlaps than on too much customization.

  • AS

    Amelia Salyers

    3 months ago #

    Thank you, everyone! This was super-fun and my fingers are now super-tired. :-)

    • AA

      Anuj Adhiya

      3 months ago #

      Mission accomplished!
      Thanks for spending this time with us Amelia - much appreciated!

  • NU

    Nitin Upparpelli

    3 months ago #

    What in your opinion are the best tools for email marketing and landing page optimization?

  • HS

    Hyderali Shaikh

    3 months ago #

    How did you increase the blog subscribers? How long did the process take & was/were there any specific blog post(s) that triggered the increase in subscribers?

  • DO

    Danielle Olivas

    3 months ago #

    1. What tools are in your box that help you generate the look and feel of your content?

    2. On a related note, what do you think the most innovative piece of content you've put out (and why do you feel that way)?

  • JD

    James Dunn

    3 months ago #

    Hey Amelia

    What channels have you found that have over delivered or under delivered on your expectations for how people find Mixpanel?
    I'd love to learn more about what worked well or didn't, and what you learned in the process.

  • SK

    S Kodial

    3 months ago #

    Hi Amelia
    You're always tempted to try lots of different tactics when starting a new blog/publications.
    So how do you balance staying patient with switching tactics that aren't working?

  • TP

    Tomasz Piotrowski

    3 months ago #

    Great to have you here Amelia and thanks for doing this!

    I would like to ask you if you have a list of "do not touch" topics like i.e. politics or similar and what are those?

Join over 70,000 growth pros from companies like Uber, Pinterest & Twitter

Get Weekly Top Posts
High five! You’re in.
SHARE
34
34