I want you to sign up for something very, very special I'm doing: Writing short stories from the intersection of marketing and analytics.
My goal is to get you promoted, you are going to love it. So. Please do sign up. But, first, as you've come to expect from this blog… Context…
Should you own or rent?
The logic we are taught from when we were babies is that it is better to own than rent. Reality is actually a bit more complicated.
In our context, let's consider two applications of own and rent.
In the past I've spoken about own vs. rent in context of platforms. Facebook, YouTube, LinkedIn are platforms where you rent your existences. Mobile and desktop websites, mobile applications are platforms you own. You own the content, you set your own creativity limits (no 140 characters or videos of only xx resolution), and you own the data on platforms you own.
I've also spoken about own vs. rent in context of audiences. You rent audiences on TV, Magazines, Search, Display, etc. You own audiences on your email lists, on forums you host etc.
Audiences and platforms, two places to bring our own and rent lens.
Before we go on, a quick word on the word own. It does feel odd to say you own anyone/thing. Own in this context is like as much ownership as you can apply to a seed you plant in your front yard. You need to protect it, you need to nurture it, you need to champion on its behalf, and you need to be unselfish for a long, long, long time, and maybe some day you get flowers. And, here's the most amazing thing, if you only do it for flowers it won't work out as well as you want it to. Unselfishness.
Clear on what own means? If not, please comment, please argue with me, please share better metaphors with me (especially, after you read what I'm up to below).
Why evolve the strategy now?
Some recent events made me wonder what the best strategy for a business is. Rented platforms look incredibly appealing. Just Facebook has 1.2 billion MAUs. That is crazy big. YouTube is 1.1 billion. Basically everyone with internet access. How can you ignore them?
Facebook in particular is very hard to ignore because it is much more of a closed system. You have no other choice. If you are not Facebook there is no chance anyone will see your content. This is less true on other platforms. For example, on YouTube if you don't totally suck, your content will show up on other sites. It is also likely people will bump into your YouTube content as they search on Bing or Google. Still. Unignorable, massive, audiences on these platforms where you rent existences.
The problem is, they also control who sees how much of your content and when (as is their right, they are for-profit entities). Organic reach on Facebook is now under 2% (and, that is ok, their call). Twitter is rumored to be switching to an algorithmic feed, they get to decide (again, this is ok, they are trying to improve a product only actively used by journalists and one man running for President).
The end result for a business is that you lose control, predictability, and, worst of all, the ability to form relationships with your audiences. If I can't predictably can't reach an ardent lover of my brand, Ms. Zenobia, on Facebook, do we really have a relationship? Or just a one time hello? And, what is the point of that hello (/like)?
Here's a great example of the risk you carry by only having rented platform strategies, so much worse than the Edgerank chokehold Facebook has on your organic reach… Google+ is a good platform, and Google continues to experiment with it. I, as you can expect, went all in on G+ and worked hard to build an audience of 496,601 followers. Yes, you read that right, almost 500k.
As I created unique content, from my unique vantage point, I would get huge engagement on G+ with a hundred comments, many more +1s etc. And then, as Google evolves its strategy from it's own point of view, one result was a tweak to it's algorithm. This caused a hit with Conversation Rate, Amplification Rate and Applause Rate. Then, as strategic evolution continued, the Google+ team decided to become a different product (as is their right, I do not begrudge them). The impact? Normal engagement for me is now 12 people.
That is less people than are sitting around you as you read this post.
While I'm pointing out my example, your brand faces the same risks. While I'm pointing out Google+, this has been replicated on every social channel (my LinkedIn influencer channel is another great example), every newsgroup, every other rented platform over time – all you need to do is look. Sadly, we don't think about this enough. We don't worry about what we are doing on the long run, and if it is a dangerous strategy to ONLY rely on rented platforms.
There are people in our world who are crazy enough to run around saying this: "I have a Facebook page, why do I need a website?" #WTH
Or a variation, "I'm posting pictures on Instagram and videos on YouTube, why do I need my own audiences?" #OMG #WTH
For myself, for the companies I'm a part of, I've decided that the risk is too high. I'm creating a strategy to diversify both platforms and audience relationship quality. I'm not going to rely on building audiences that I kindasorta own at the mercy of rented platforms.
Occam's Razor, and now TMAI!
The best platform where I own content, creativity, data, relationships, and everything else is this blog. Based on my analysis above, I'm evolving my strategy by adding a newsletter to build audiences I can nurture and own.
I'm really excited about it. It allows me to take the daily analysis I've channeled into creating content on Facebook and shift the best bits of it into a platform I own. It allows me to be more creative in my expression and I don't have to deal with their limits, just mine. It allows you to be in control, you can choose when to ignore me rather than have the rented platforms determine if you and I should engage. It allows me to write in formats and about topics that don't quite fit the blog. It allows me to experiment.
My newsletter is called The Marketing < > Analytics Intersect. It will contain short stories from where I live, at the intersection of the worlds of analytics (where I started) and marketing (my tool to drive change in customer relationships/behavior/branding). You can sign up for it here:
If you can't see a webform above, you can sign up by visiting my sign-up page.
And please remember to go to your email immediately and click on the activation link!
In a moment, I'll share three newsletters to give you a feel for what I'm writing about. But first, here's something important as this post is not only asking you for something (your attention!) but also trying to influence your marketing strategy.
Should you abandon rented platforms or audiences?
Should you ditch Facebook? Or renting audiences on TV or Instagram?
For most people, the answer is No. You should not abandon them.
There are such huge audiences on Facebook, Twitter, TV, Google+, creating content that causes serendipitous contact/conversations is of value. (Of course, measure that using the four best social media metrics!) There is no doubt that if you do something that catches fire (I refuse to use the v word), these rented platforms can really reach massively move people than you can all by yourself (often, you can't even get that reach with paid advertising).
But. My plan, and that for businesses I work with, is to invest less in shallow transient relationship with large numbers, and invest more in deeper persistent relationships with smaller numbers. More emails (and other owned platforms) and fewer Facebook/G+/Twitter posts.
For small and medium sized businesses, I think the answer is yes. Ditch. For more on that, please see this post: How To Suck At Social Media: An Indispensable Guide For Businesses!
Excited? What can you expect in the TMAI emails?
More of me. A bit more conversation, like we would have at a party. A little less of the formality that exists on this blog. Many fewer of the numerous considerations that go into something that ends up on Facebook. A lot more of, here's what goes on behind the scenes, as you'll see in TMAI #3 below.
I do not plan to publish the newsletters anywhere (no web versions, not even an archive), to allow for more openness and intimacy. I do plan to still engage actively. Ask for your ideas and contributions, do an exercise together, send topical things that might not make other places and so on and so forth.
I'll write about analytics and marketing of course, but I also plan to write about digital experiences I adore, the occasional career advice (as I have the privilege of leading a large team now), and yes a rant now and then (eBay just sent me a confirmation email where the Universal Remote picture is hugely squished!). My hope is to have a material impact on our career by exposing ideas you might otherwise not have bumped into as easily.
I hope you'll sign up.
To give you a feel for what the newsletters will be like, as a one time exception I'm sharing the first three. Even though they are rough, this is all still an experiment and I'm still figuring thing out, I hope they'll give you a feel of the value you can expect.
Here's the very first newsletter I'd sent, two weeks ago, and it touched on a confusion I find common, and frustrating…
TMAI #1: Metric or KPI, how do you decide?
People tend to use the terms metrics and KPIs interchangeably. I humbly believe this is a disservice as it results in a lack of focus on what really matters: Business impact.
In my post Web Analytics 101: Goals, Metrics, KPIs, Dimensions, Targets, I'd defined metric as "a metric is a number." Simple. :) I'd defined a KPI as " a key performance indicator (KPI) is a metric that helps you understand how you are doing against your objectives."
See the difference? Only one of them, if it changes, will have a predictable impact on your business. A KPI.
Let's consider an example.
Bounce rate, a number I'm fond of, is a metric. It can never, ever, never, never be a KPI. Why? Let's say your bounce rate is 50%, and you reduce it to 0%, is it likely that you had a business impact? Maybe, maybe not. Because all bounce rate measures is that you saw more than one page. If you see two pages, that's zero bounce. But, you might have puked on the second page and left, or third, or fifth. It is hard to tie you seeing just one more page to business success, as it takes a whole collection of pageviews/hits to deliver business success.
Bounce rate is a metric. A good one. It will help you do better, and you'll use it. It's just not a KPI.
So, what's a KPI?
It really depends on what your business objectives are!
But, some common KPIs for ecommerce websites can be CPA, Conversion Rate, Unique Pageviews, Time Lag and Path Length, Task Completion Rate etc. Any movement in these metrics, and money falls from the sky.
Some common KPIs for non-ecommerce websites can be micro-outcomes conversion rates, Unique Pageviews, Page Value, Task Completion Rate, Recency and Frequency, etc. Any movement in these metrics, and… you know… money falls from the sky!
Be careful what you call a number. Train your organization that if we call something a KPI, it is the most important thing for everyone, obsess about it, trend it, segment is, custom report it, basically just shy of making love to it.
What's in a name? The difference between winning and losing.
PS: If you love anything as a KPI for an ecommerce site or a non-ecommerce site (sorry, that's all the context you get :)), would you please reply? Thanks.
My goal is to have them be under 300 words, and pack them with actionability. There will always be something that will make you stop and think.
The second newsletter addresses something that I'm deeply passionate about, context… it is the one thing that get's in the way of understand what the data is trying to say…
TMAI #2: Context, Context, Context, Your Data's Begging for Context!
I believe that a massive contributing factor to us, or our bosses, making poor decisions based on data is that it is missing one key ingredient: Context!
You see a sparkline, you see a number, you see, and I hope you don't, a pie or, even worse, a speedometer chart. And you pause. Ok. Things are moving/happening. But. Is that good or bad?
That's it, right there. The problem.
The solution is to give your data context.
Here are five effective strategies:
1. Leverage advanced segmentation. Ex: Rather than looking at overall trend for any metric, segment it. Compare a trend of Unique Visitors with those from your largest free and paid traffic source.
2. Use time to your advantage. Ex: Compare this month to last month, this week to same week last year, this year to last year, etc.
3. Industry Benchmarks FTW! Ex: The average mobile traffic for newspapers in Croatia is 88%, but our Croatian newspaper only gets 18%. #wth
4. Competitive Intelligence. Ex: Our referral traffic is up 40% YOY, but our despised competitor's is up 500% from these top five sources.
5. Company Targets. Ex: Our average Cost Per Acquisition is down to $38, much better than the target of $45 we had set at the beginning of the quarter.
Bonus: This post has three more strategies I've not covered above: In Web Analytics Context Is King Baby!
You don't need to use every strategy above all the time, but you have to use at least one, and usually two, in order to get the context you need to allow data to give your mind the good kind of headache that causes you to dig just a little deeper and make smart recommendations.
No context? Lonely data? No soup for you!
I hope you can see a continuation of the tone and texture you are used on on this blog.
My third newsletter was an experimentation in engagement possible with a smaller focused audience, the topic was something I've expressed a great deal of affection for on this blog…
TMAI #3: Visualizing Lies: Two Answers.
This is an experiment, I know it will be hard to read this email on your phone. Would you be so kind as to reply if this still works or if I should avoid sharing data visualization examples in my newsletter? Thank you.
Someone sent me this graph about how many lies and how much truth the US presidential candidates are sprouting…
Line graphs are not great at this, they are best at trends and hence we can't help but "follow the trend from left to right."
The obvious answer is to just make it a bar chart. The challenge is that there are too many data points and it will look messy.
I felt a table would work best, with good old conditional formatting in Excel.
But as I created the table, the challenge was that there were two stories I could tell. Given that I'm the Occam's Razor person, :), I decided not to mush them together.
Story one I created was to show which candidate "wins" each category.
You can easily see that Mr. Trump tends to be pants on fire (surprise!) most of the time and Ms. Clinton wins the coveted true category. As you look across each category, I think you'll get really interesting pauses.
Did you notice I'm using the borders to "guide" your eye to look vertically? I'll try horizontally below. Let me know if it works.
Of course like you, I'm tough on myself. I did not think the above visualization does a good enough job of showing how each candidate performs across all the categories.
So, here's that story, framed as simply as possible…
The one self-critique I have to make here is that green is not necessarily the best color to use as Mr. Trump's performance is not really green (as in green is good). But, hopefully that is not too big. You can see through that little thing and see where each candidate resides.
I worry a lot about visualization because most people will jump to conclusions, not just understand or not understand what you are saying. I've opened the kimono here and shared with you what I worry about in this specific case, and shared more with you than I would in a public setting. Benefits or a newsletter. :)
If you want to play with the data, it comes from Politifact. I can also send you the table. I would love your feedback on the above two, or even see a version you create.
This was such a fun exercise, even with the limitations of the medium. If you want to see a larger size of the above three images, click here to see it.
You'll see more of the above type of here's what's going through my head as I play with this data in the newsletter.
What was awesome was that I got over one hundred submissions in return! Way more than I'd expected. I replied to every single person with my feedback (though as the list grows I'm worried about how to find time to reply to everyone).
Let me share two examples. This one's from Kate Driver…
Very cool, right? It is a useful stacked bar chart, makes the case pretty well as to who is where. Simple, and effective.
She's rightly followed the order that we had in the table originally. I was wondering how it would look if I moved Trump to anchor the other end of the spectrum. Here's that view…
In this case, the lying seems to stack-up very nicely (see what I did there?).
I learned something from Kate, and that is what's amazing about these conversations.
The second example from the submissions comes from my BFF Thomas Baekdal. [If you are not subscribed to Baekdal Plus, you can't be smart about digital anything! Example: Free report: Do You Need Real Loyalty as a Publisher?]
Here's Thomas' data visualization…
Thomas did something I'd tried to avoid, add opinions to the data. He's added weights to each rating, including penalizing quite a bit for pants on fire (-10) vs. rewarding for true (+5).
I think his visual is better. It gets to the point much, much faster. My worry, rightly or wrongly, was that the partisans will come out with pitchforks and soon data will fade away while we argue about the two parties. Sad.
You should read Thomas' post in greater detail on Google+, it is insightful.
Was it not a lot of fun?
You build a hand-picked, relevant, focused audience and the impact is larger than one might anticipate.
The subject of TMAI #4 was "Purpose first, data second. Yes, second!", I'd shared one magical golden question that brings purpose to our data analysis better than any other. It is amazing, because it takes the focus to the one thing companies really care about.
TMAI #5 was a lot of fun, the subject was "What's your brand's most salient consistency?". I feel so few brands know the answer to this, and hence they can only compete on performance. I used the example of OKGO (plus Sony, Patagonia and Bill Watterston!) to share a long-term point of view you can apply to your business.
Upcoming newsletters cover subjects, from that valuable intersection of marketing and analytics, such as over-the-counter data, diamonds are not a girl's best friend, and my all time favourite data cartoon.
If this sounds interesting, please sign up for the newsletter now:
Except for this one exception, the newsletter content will be non-public and only accessible through your subscription..
As always, it is your turn now.
Does your business have an own AND a rent strategy when it comes to platforms? How about audiences, just renting or also a dedicated effort to own? Email continues to be an amazing medium for connections (as long as you don't suck), has it worked for you? If not, why not? What do you think of my retro-strategy of having a twice a week newsletter? Sound use of time, or a waste of time?
I invite your feedback, insights, critique, examples and challenging ideas!