Category Archives: Marketing & Advertising

The Only 3 Times You Should Speak In A Meeting

One of my mentors, who I still call on for advice, shared with me her approach to meetings. I was relaying that info to someone today and it dawned on me, I’d never shared this great advice more broadly.

I think many of us take 1 of 2 paths in a meeting. We speak all the time or we say nothing. Those who say nothing, of course, are given the horrible label of “wallflower.” [sarcasm font] Wallflowers of course aren’t leaders [/sarcasm font]. The other end of that spectrum are people who talk all the time; they yearn to hear their voice heard. Well, at least that seems to be the prevailing thought.

Microphone

So with that said, as I was entering a point in my career where I was routinely finding myself in meetings with the C-suite, my mentor shared with me her philosophy for when to talk in a meeting:

Speak First: If you’re the first to speak in a meeting, you set the tone, frame up the context of the meeting, outline what’s to be discussed and the decisions that need to be made. As the first person to speak, all other commentary will have to pivot off of your opening remarks. This is a great role when you’re the sponsor of an initiative, but not responsible for overall completion of the initiative or the tasks required to complete it.

Speak When You’ll Own: When you’re responsible for key parts of an initiatives, you need to fight for your fair share of the resources, budget, scope, timing, support and more. You need to both pick your spots and speak in a tone and octave that enables you to explain risks clearly and the reasons you need what you need. Also, since you’re responsible for the “task” it’s critical to make sure people understand you have that responsibility. Equally as important is making sure you understand what’s being asked of you and what scope changes are becoming your responsibility.

Speak Last: The last word is undervalued. The recency effect is very powerful in large organizations. As the person who speaks last you can sum up the key points (which are almost always captured flawlessly by scribes), outline the next steps, assign responsibilities, express satisfaction/dissatisfaction and more. The tone you set when you speak last, often sticks and it’s remembered. The “speak last” approach is something reserved for the person who owns the responsibility for the overall initiative.

It should go without saying, but obviously speak when you’re asked a question too.

I didn’t get this advice til a bit later on in my career. It’s something I’ve been working on for the past year and have tried to lean in to since I returned to Walgreens. Try it out.

What It Takes To Be A Truly Enterprise Social Platform

It’s 2014. I can tweet from 36,000 miles up in the sky. I can turn the lights in my house off and on from my phone. I can adjust the temperature in my house from 1000s of miles away, with just a few taps on my phone. Taking selfies on stage, during a live event is common place. We have public political discourse on platforms like Facebook and Reddit. Edward Snowden can live stream in for an interview at SXSW. With platforms like IFTT I can have have every photo I take on Instagram be automatically backed up to my Google Drive account and then have a text sent to my wife, letting her know they’re backed up. Yeah, technology is amazing. It really is. Tech has evolved to a point where anything is truly possible. We don’t think in terms of “can we do that” – we think in terms of “how we’ll do it.”

With such maturity in technology, business adoption of digital/social and the marketplace as a whole, it’s perplexing that we still don’t have a truly enterprise social platform. Oh, we have platforms. We have no shortage of platforms. These platforms are like tools that go into a toolbox. Ask a social marketer what’s in their social toolbox and you’ll get a wide variety of answers. The toolbox will run the range of small and niche platforms that do 1 thing exceptionally well, to platforms that do nothing exceptionally well, but do everything. You’ll find platforms that you pay for monthly on a credit card and you’ll find platforms that are hundreds of thousands of dollars and billed every quarter.

There are platforms for publishing (Hootsuite). There are platforms for sourcing (Percolate). There are platforms for monitoring (Sysomos). There are platforms for analyzing (Crimson Hexagon). There are platforms for reporting (Simply Measured). There are platforms just to make data look better in the form of dashboards (Geckoboard). We have platforms for everything. What we don’t have is 1 truly enterprise social platform that can do it all, and do it all exceptionally well.

We had the promise of such a platform. Remember the marketing behind…SalesForce’s Marketing Cloud? It was to be the end all, be all answer for a social business. But, it fell flat. Actually, it imploded. Don’t take my word for it though. Marc Benioff, SalesForce’s CEO, at DreamForce 2013, all but admitted, Marketing Cloud never realized its vision. The revenue numbers reported at the end of fiscal for SalesForce back that up as well. I’m not picking on SalesFore, more so, it’s important to note that if a company like SalesForce can’t get it right, you can’t expect others to as well. It’s hard. Social, at scale, is hard.

At any marketing conference, you can bet, leaders of social are discussing the tools/platforms they’re using, with one another. If you were to listen in to these conversations, it would sound a lot like something connected to something via tape, bundled together with rubber brands, with Google docs in there somewhere, filling a gap. It’s a mess. And, honestly, as we close Q1 of 2014, I can’t believe it’s still a mess.

I always say, don’t shake a stick at something, if you aren’t willing to offer a recommendation for how to fix what’s wrong. With that in mind, here’s what I think a truly enterprise social platform needs to have…actually, before I get into that, first let me outline what companies building platforms need to understand:

The “Plumbing” Isn’t Easily Changed: If you’re already working with a suite of platforms and partners, and chances are, you are, changing platforms and partners can be painful. While there are some platforms you can change out with very little pain, the fact is, often times, changing platforms brings upon legacy challenges. There’s also the likelihood that there are some platforms you’d like to retain. With all that in mind, what makes a buyer’s life easier is if your platform works more like a connector than a pipe. If you’re coming into a situation where there’s already existing plumbing in place, your platforms ability to connect and play nice with all the existing platforms that are in place, makes your platform enterprise ready.

The Seat License Model Is Antiquated: There’s simply no reason you should be offering a seat license model, if you’re claiming your platform is designed for the enterprise. A truly enterprise approach would mean that anyone, at any time, from any team, any partner, across the globe, should be able to access and use your platform. What makes your platform sticky in an organization is having it used by a significant number of people. The more people who use, enjoy and rely on your platform, the better. When you charge per seat, what you’re conveying is that you don’t want your platform used at the enterprise level. Let’s use some simple math to illustrate this point. Let’s assume your platform uses a $100 per person per seat per month approach. At a company like Walgreens, where we have 200,000+ employees, you’d in essence be charging $240,000,000 a year. Yes, I said, $240,000,000. Even if I go with a 90% discount, and it’s only $10 a person per month, we’re talking about $24,000,000. That’s insane. Beyond the fact the dollar amount is ridiculous, it also positions your company as someone who nickels and dimes. If you have a great platform, price it like one. Don’t try to compensate for how much your platform lacks, by charging on a per seat basis.

Sell The Platform or Services, Not Both: Building on my nickel and dime comment, don’t adopt a car dealer mentality where after I’ve bought the car you keep selling everything from warranties to undercoating. If you make a killer platform, make a killer platform and charge a fair price for it. If you’re great at enterprise services and strategy, start a consultancy. When you try to sell both, it makes buyers wonder if the reason you’re pushing your enterprise services so hard, is to compensate for an inferior platform.

Ok, now on to what a true enterprise social platform should look like. It’s actually quite simple. A social enterprise platform needs 7 core features:

Light Listening: A light listening feature needs to answer one simple question, “what’s going on, right now.” That’s it, it’s that simple. I’m serious. This feature needs to address the business scenario that often arises, where someone asks, “what do consumers think/feel about X.” This feature takes the pulse of a situation. It’s directional. It doesn’t have to be 100% accurate. It needs to be accurate enough to offer some directional context.

End To End Publishing: Your platform should allow for the sourcing of content, review of content, editing of content, distributing of content and the evaluation/measurement of content. Many platforms do 1 or 2 of these things really well. But, no one does them all at an A-level.

Rich Analytics: Where light listening answers the question, what’s going on right now, rich analytics answer the question, “what happened.” Hindsight is supposed to be 20/20. Rich Analytics need to be 20/15. If we’ll accept 70% accuracy for sentiment analysis with the light listening portion of a platform, the rich analytics must be 90%+ accurate. You also can’t lock this data behind a wall. It needs to exportable into a wide variety of formats. It should also have the ability to mashup other data sets. For example, a conversation volume chart, is a nice 1st step. If everyone is talking about a specific link on your website, wouldn’t it be great to be able to see your Omniture data related to that link, in the same enterprise platform?

Customer Care Management: If you want to be a serious enterprise player, you need a customer care module. The key here is that module must work as a stand alone feature, because in many orgs, care is handled by a separate team. It also needs to fit into the other modules for situations where an org has 1 team working across all aspects of social, including care. Most platforms can only do 1 of these flows well.

Mobile At The Core: We don’t want to hear it’s mobile web friendly because it’s built in HTML 5. That’s a nice first step. But, your platform better have an app there iOS and Android ready. Simply put, 75% of everything I can do from a desktop experience needs to supported in your app.

Flexible Integration: This is the hardest part. I get it. I want you to integrate with Simply Measured, but you see them as a competitor. I get it. But, I also need you to figure it out. Again, if you want to be enterprise and you want to become the irreplaceable plumbing, you’d better figure out how to play well with a whole host of other platforms and partners.

Multiple User Roles: Probably the easiest feature to nail. Some people need to be administrators, capable of changing, creating, deleting, etc. Others require just read only access. There’s also a host of other roles in between those two end points.

I don’t pretend to know how to build software. It’s not easy. I can appreciate that. I can also appreciate that in the Wild Wild West that has been the last 5 years, in social, it was much easier to build a half baked platform that sorta did 1 thing really well. Clients were lining up to buy your version of Windows ME and there was little reason to think bigger and establish a higher bar for quality of experience.

It’s 2014. Social is growing up faster than we all thought. The bar is high now. It’s no longer amateur hour. I’m sure I missed a few things in this post. But, I’m also confident I hit on 90% of the pain points and requirements, expected by Sr. leaders in social.

You can do better. You need to do better. If there’s one thing history has shown us, the first one to really nail it, wins. In this space you don’t want to be 2nd place.

The First World Problems Of Living With Second World Internet Speed

Cut the cord they say. Just stream everything. Join Netflix. Don’t buy music, just listen to it from iTunes Radio, Pandora of Spotify. Move on to the new revolution of gaming, where platforms like the XBOX One require an always on connection so you can download content…on demand.

In theory this sounds amazing. And the geek in me relishes that promise. As a digital marketer and dad, I’m always connected. It’s part of the lifestyle. This morning, I was checking emails, while both kids were streaming Netflix. There were no hiccups. No congestion. No buffering. Everything just worked. Then again, I was in Minnesota, not at home.

We recently moved back to Chicago. It’s a big city, right? We bought a condo. Our building is awesome. It’s in a fantastic location. It has indoor parking (a rarity in Chicago). The combination of hardwood floors, high ceilings, big windows and awesome views make for the perfect new home. Well, not perfect. Almost perfect. There’s 1 big problem. Get your small violin and kleenex box out. The problem…the bones of the building are a bit old and we’re stuck with DSL. At present, there’s a “Cable Committee” (GIANT eye roll) investigating alternatives, because I’m not the only one feeling stuck in 1997.

Internet Speed By State

So you might say, well, what’s so bad about DSL? Great question, glad you asked. Before I answer, let me take a step back and provide some context. My first connection to the web was via AOL 2.0 on a 28.8 modem. We graduated to a 36.6 and we were blazing! When I left home and went to college I had my first taste of a T1/T3 setup. The entire University of Minnesota was wired for speed! Of course, these were the days when infrastructure outpaced demand. Well, until Napster, Grokster and Limewire appeared. As sharing networks like that gained traction we had finally had a situation where demand/usage matched the speed of the infrastructure. I’ve always been pretty fortunate when it comes to internet speed. Be it Omaha, Minneapolis or most recently, Camden, NJ, the pipe has been wide. At our last place in Camden, we were operating on a 100mbps connection. That’s just ridiculous. More ridiculous, we paid less than $60 a month for that. Today, we get about 12mbps at a cost of nearly $75 a month. Think about that…1/9 the speed for 25% more cost. How does that make sense? It doesn’t, but that’s a topic for another day.

For the past month, we’ve been living the DSL life. Is 12mbps slow? Is it really that bad? Well, based on the current global speed data, that puts us in the company of Aruba, Mexico, Kazakstan and Vietnam. For even more context, that’s just above Guam, Brazil and Qatar, but well below Uruguay, Ukraine and Slovenia. The United States speed average is 21mbps.

Ok, so now I’m off the soapbox. The 460 words above were equal parts rant and context. Similar to my experiment a few years back, when I deleted by Facebook account, at about day 3 of living with DSL, I started to ask myself, how can the experience make me a better and more informed marketer. There are 4 things I’ve taken away from the experience:

  1. We’re years away from a truly connected home that relies on the cloud. It’s not desire. It’s not interest. It’s not consumer spend. It’s something basic: infrastructure. I wrote on the subject of infrastructure strangling mobile’s growth in 2009. It did. It has. And, the recent changes to net neutrality are keeping mobile from becoming dominant. Companies should be “designing” heir experiences to a lower common denominator. Microsoft is the biggest culprit here. The XBOX One was designed as an always on experience. It’s constantly checking for updates and downloads. This was annoying on the 100mbps pipe in NJ. It was damn near catastrophic in the new place. I purchased a new game. Inserted the disc. Instead of playing immediately, it needed to download an update. That update took almost an hour. By that point, I didn’t even want to play. I don’t think I’m alone in that feeling. As we think about new experiences to create for consumers, a major filter for me will now be, are we designing it to the right lowest common user?
  2. I’ve become less connected and more selective in what I do on the web. If you’re crippled by speed you have to adjust and change your behavior. For example, I’m back to a behavior I had in the mid 90s. I now setup all my updates before I go to bed and let them do their thing while I sleep. Before, I would have multi-tasked and done these types of things while I was also streaming a Netflix movie, after dinner. Speaking of Netflix, I haven’t watched 1 single thing. The experience is simply far too compromised. We need to create experiences that inspire a consumer to choose interacting with us, than all the other things they could do on the web. That’s a high bar.
  3. I’m thankful I never got rid of, nor did I stop investing in “analog” content formats. The 50,000 songs I have on my computer enable me to not rely on streaming Spotify. The 100 or so DVDs I own mean I’m not stuck waiting and waiting and waiting for Netflix, HBO GO or Hulu+ content. We think the world is digital and that digital provides a benefit over analog. Often times this is true. But, we need to consider that sometimes analog is faster, easier and better. My favorite example is list creating. No matter how cool a list making app is, it’s rarely faster or easier to use than a pen and a piece of paper. That’s what we’re up against, every day.
  4. The future is clearly mobile. I’ve used my cell phone or iPad, both on Verizon, as hot spots, more in the last 4 weeks than I ever have…combined. Seriously. As I looked into options to move off of DSL, the only consistent and real option was to use a LTE whole home device. I think it’s likely, I’m going to head in that direction. With that in mind, the consumption of content will happen more and more, away from home. In my role, it reinforces the need to make sure content is liquid, linked and loved.

Method Cards

This post wasn’t written to make you feel sorry for me. If you do, great. I certainly appreciate it. If you can set me up with something faster, I’d really appreciate it :) – no, this post was more about forgotten technique of walking a mile in your consumer’s shoes. IDEO’s famed Method Cards are designed to get consumers to look, listen, ask or try. As marketers we can get so caught up in the data, spreadsheets, decks and white papers that we forget to simply act like our consumers so we can understand them better. While I can appreciate that, I’d still really love to have my fast internet speed back…

First world problems, indeed.

The Human API Is The Future Of Big Data

You are the next app. Yes, you. And you. And you over there in the back. You are all the next app. Ok, technically, not you. But, your data. No, I don’t mean your name, your email address or your birth date. Those are old school pieces of data that will continue to depreciate in value, over time. When I say you and when I say your data, I literally mean all the data you generate.

On some level, the near future is mobility, big data and personalization. The irony of course is that’s not really a new trend. Companies as far back as the catalogue business models of the 1930′s relied on our data. Where we lived. What we ordered. How much we spent. How often we ordered. Since that boom of the 30′s companies have always valued our data and tried to collect it. Remember those warranty cards you filled out? They were really just an easy way for the manufacturer to learn more about the person who just bought their item. Smart, right? Today, when you login to a site with Facebook Connect, you’re trading your data for simplicity.

Digital DNA

We trade data all the time. But, something is changing. As companies look to advance their product pipeline, they’re more reliant than ever on us to power those products. Let me offer some examples:

  1. For Nest to continue its meteoric growth, it’s going to need more data about you and your home. Devices that are part of “connected home” will take off, provide value and offer a wow factor that leads to broad adoption when it’s powered by more and more data about us. Imagine Nest changing the temperature of your house when it knows you’re within 5 miles of it. How does it know? Because you allowed Nest to track your location so that this happens automatically.
  2. What about Samsung’s entire foray into wearable devices, including the biometric tracking that’s part of the Samsung Galaxy 5? Those “wow” features are only of value and help Samsung provide better phones and better experiences if YOU allow them to have access to your fingerprints, your pulse and more.

Those are just two very simple examples that underscore a simple truth: companies will need our “DNA” to make their products work.

It’s been long said, if you’re not paying for it, you’re the product. That’s helped us accept the pervasive tracking by platforms like Google and Facebook. We get access to their great platforms at no financial cost, because we’re providing them with data that they are able to resell to advertisers for a significant upside.

But, what happens, when you’re not only the product, you’re also paying for the product? Shouldn’t my data lead to some type of financial benefit? After All, without my data, they’re products are underpowered, which slows their roadmap, which leads to poorer financial performance.

We are entering an age of mutual exchange. Never before have we been on such a equal playing field as companies. They need us as much as we need them. Let’s take a relatively small example that proves this theory out. Today, Progressive Insurance gives you a discount for letting them track your car driving habits with their Snapshot product. You’re getting a financial value for your data. That’s mutual exchange. That’s the future.

Now, let’s take a bigger example. Today, Google takes your data and in essence sells it to companies to market you; you don’t see a dime. Yes, you get to access things like Gmail for free, but even with those products, you’re at the mercy of Google’s roadmap. Remember, Google Reader and how Google killed it, despite people not wanting it to be killed? If you’re not paying for it, you’re the product.

Well, what would happen if Microsoft’s Bing, said, you know what Joe Consumer, you’re in control of your own data and in doing so wrote you a check every month for how they’ve used it, based on how you’ve opted in. The more data you shared, the more you’re worth and the more you’re paid. Would you use Bing more?

You AND your personal social network are now the new API.  Don’t turn your most valuable asset, your data, into a commodity. You’re worth more than that. Companies should be paying us to access our API, not the other way around.

13 Reasons Tim Cook Just Bought Twitter

Ok, I lied. Well, technically, I didn’t lie per se. Sure, there’s nothing in this blog about Tim Cook, Apple or twitter. But, I didn’t lie. I just played by the wide open and loose rules of today’s publishers. See, what I did, was I link-baited you. You saw that salacious headline, “13 reasons Tim Cook Just Bought Twitter” and you clicked. If I had been selling ad-impressions on my site, I’d have just made a fortune.

Admittedly, you’re irritated. You expected to find an article outlining why Apple decided to buy twitter, instead, 1.5 paragraphs later, you’re still reading my lecture. You should be irritated.

Tonight, I was a bit irritated too, so, I got a bit cheeky on twitter and started generating semi on topic/semi off topic headlines that were completely made up. For example:

 

The number of people who tweeted me back asking for the link or thinking I’d forgotten the link was staggering. We have been conditioned to look for headlines/tweets like this…so we can click on them.

WhatsApp Link Bait

When the news first broke about Facebook’s acquisition of WhatsApp, I rolled my eyes and I debated avoiding social media for the next few days. But, I didn’t Being plugged in to social is part of the job and the responsibility that comes from leading an organization’s social marketing efforts. Why did I want avoid? Simple, I’ve seen this news cycle before. The announcement comes out and we end up with hundreds of posts that seem an inch away from the Tyson Zone. They all follow the same formula:

The + X (a number) + Y (a noun) + Z (the actual news) + A (preposition) + (simple phrase)

For example The 18 Ways Facebook’s Acquisition of WhatsApp Is a game changer. At this point, I’m half sure publishers have simply written a script that generates these headlines. After all if Len Kendall can do it as a side project, it stands to reason a large publisher could do it too.

So, yes, I got a bit cheeky, had some fun, but also learned a lot. For example, I’m not the only marketer who’s self-aware enough to realize that:

  1. We have become conditioned to expect headlines like this
  2. We know it’s a problem

This approach to “reporting” the news could very well be called link baiting. An interesting headline rarely is paid off by the actual content contained in the article. The headline is salacious, which of course gets you to click. This bothers me. It’s always bothered me. But, now that I also have the responsibility of our Walgreens enterprise content strategy, it don’t just irritate me, it really concerns me. Let me break this down…at the end of the day branded content can only live in 3 places:

  1. Our owned real-estate: For example our website or opted-in eMails. In this case, we need to think about how we use a variety of paid and organic approaches to drive people to those locations.
  2. Distributed on another platform (e.g. twitter) organically: In this situation, we’d be recognizing that you might not want to leave the experience you’re currently in, but you still want content from us.
  3. On another publisher’s site: Because buzzwords are king, let’s call this content “native.” If it’s native content, in essence we’re paying to have our content embedded on another publisher’s site. The upside here is rather than trying to drive someone from where they already are to my site, I can “engage” them where they already are.

Bucket 1 has been around since the early 90s. Be it web-rings (yes I said web-rings) or the earliest form of display ads (remember the 120×90?) companies have been “buying” ads across the web to drive people to their sites.

Bucket 2 isn’t quite new, but, it’s not quite a mature space. Brands are still figuring out how to balance the value of building a base of followers on someone else’s platform, for the purposes of marketing to them. Yes, I said marketing. I didn’t say engaging, which, let’s be honest, is simply a more polite way of saying, marketing.

Bucket 3, though, well that’s an interesting one. You can call it “native” or any other name, but it’s still an ad. I won’t get into the merits of native ads vs. traditional display ads, here. It’s a subject I’ll tackle at a later date. With native ads the publisher is selling traffic. They’re ultimately claiming, hey, we get X millions of eyeballs to our site, thus your reach is some % of X. Simple enough, right? After all, that’s really not too different than bucket 1. We’ve been buying ad inventory on CPM models for years. In those CPM models, an advertiser chooses to advertise on that publisher’s site, because they reach X millions of eyeballs.

The big inherent difference though between bucket 1 and bucket 3 is that bucket 1 created and built during a time when portals (e.g. Yahoo, MSN) were the starting point and people browsed for content. There was a certain assumed intent. In bucket 3, when you’re essentially advertising inside the stream, the intent is debatable. Publishers are selling reach in the form of impressions, which come from clicks. Well, if I were a publisher, I’d publish outrageous headlines, just like the one that brought you here. It’s smart economics after all. As the publisher, I craft the slightly misleading, slightly on topic headline, you click, I claim your traffic, I then aggregate all the people who clicked on the link and I tell advertisers, see look how much traffic we have.

But, doesn’t it beg the question, is it really quality traffic? And that’s the rub. I applaud Facebook for taking steps to change the newsfeed algorithm so that link-baiting sites, like Upworthy were de-prioritized. Shouldn’t the headline match the actual content on the page? Jack Marshall at DigiDay recently covered this topic, in superb fashion.

We have become conditioned to look for links that fit the: The + X (a number) + Y (a noun) + Z (the actual news) + A (preposition) + (simple phrase) formula. We can’t help but click. And doing that, allows the problem to continue.

As someone focusing on an enterprise content strategy for a beloved, large and progressive organization, I’m concerned and I’m pausing. I’m tending to scrutinize the numbers publishers are providing. I have to ask myself, how much of that traffic is actually legit and how much of it was manufactured through link-baiting headlines. The difference for some marketers could millions of dollars wasted on empty clicks and impressions.

But, see, that’s something a brand cares about. That’s something the advertiser would care about. There’s little incentive for publishers to change and the associations that should be providing leadership, like the IAB, don’t even have brand-side representation. That’s quite a conundrum and I have a feeling it’s going to change. As content strategies become ever more important for organizations, there will be many others who are asking the same questions I am. I hope that has a ripple effect and we see other platforms like Twitter start to de-prioritize content that’s clearly link-bait.

How can we expect our leadership to us seriously, when we, as an industry, perpetuate such debatably unscrupulous behavior? That’s not a sexy headline, but it’s something you should think about.

The 15% Rule Has No Place In Today’s Digital World

Read “Where The Suckers Moon.” Seriously. Before you continue with this post, go to Amazon, add it your cart and check out. Get the Kindle version and the softcover version. Why? Simple; it’s the greatest book ever written about the advertising industry. That’s broad. For the purposes of this post, Randall Rothenberg in a few hundred pages, gives all the background you could ever want or need, to help understand why the 15% model was created and why it’s been so challenging to move off of that outdated model.

In the earliest days of advertising, an agency took a straight commission off of placing and buying media. That commission eventually settled around 15%. There were companies still using this model 10 years ago. Old habits die hard. That 15% commission became a defacto standard for how much an agency should be compensated relative to the media spend. If you spend $10M in media, you shouldn’t spend more than $1.5M to produce the creative needed to satisfy that media buy. Depending on who you talk to or what association you belong to, that 15% is as low as 10% for some categories or as high as 20% in others. But, the general average is STILL 15%.

The model was simple. It was clean. It enabled planning to be easier and faster, because everyone knew the compensation model. It was also a model that was born when we were only planning against a limited number of communication channels: TV (3 channels at most), print, radio and outdoor.

The model worked well for legacy media channels, because the distribution was expensive, but production was relatively cheap. Let’s take a real world example: the cost to create a commercial is roughly 600K. The media cost for a 30-second Super Bowl spot is roughly $4M. This places the ratio of agency spend to create the ad, right at 15%, which is right in line with historical averages used for the last 50 years. It’s also quite consistent with reports from major industry associations and reports.

Today, where we have such high media fragmentation this model falls apart. Marketing in a digital world, requires a completely different set of models and requires us to rethink how we’re spending our money. With digital, distribution is relatively cheap. It’s the creative that’s expensive. Those who understand and embrace this have settled between 30% and 40% for the dollars needed to support a digital driven campaign. It makes sense. The rough cost to create the famous Oreo Super Bowl tweet was $2,000. That figure is based on the 15 minutes it took to create and publish the image, multiplied by the list of people who were attached to the Cannes submission for the ad, multiplied by a simple conservative blended rate of $200/hour.

Before we continue, let me clear, I’m not suggesting that Oreo actually paid 360i $2,000 for that tweet. I’m sure the cost to create that tweet, from an already once used image, was accounted for as part of a broader client/agency fee agreement. The $2,000 is a real number however. It’s the real dollars needed to create that tweet as a one-off piece of creative…just like the $600K is the relative cost to create a one-off Super Bowl commercial.

With that said, think about that…$2,000 for 1 tweet. What if you need 8 great tweets a day, every day for a year? Well then you’d be spending $4.38M in just twitter creative. Even with a volume discount of 50%, you’re talking $2.19M per year. AND, that’s just twitter? We know the digital patch quilt world we operate in is much larger than just twitter. We need creative and content for Facebook, Pinterest, your emails, Instagram, your website, Tumblr and so on.

But, for the sake of simplification, let’s just focus on twitter. And more specifically, let’s just focus on a 1 day twitter campaign. The most recent publicly documented cost for a promoted trend is $200K. A promoted trend can generate upwards of 90M impressions as seen my Coca-Cola, more is more generally in the 30M impression range. Let’s split the difference at 50M impressions for an average trend campaign.

Creative wear-out is a reality. If you show someone the same ad enough times they either take action or start tuning it out. With digital display, the rule of thumb is you need 1 creative unit per every 1.5M – 3M impressions. The variance is tied directly to the reach/frequency model you need for your category (eg auto vs. CPG). Twitter, of course, isn’t display. We check twitter several times a day. If anything you’d need more messages/creative units because the wear out would happen faster. That said, since there’s no publicly available data to substantiate that, we’ll roll with the following campaign specifics:

  • 1 Promoted Trend Campaign
  • $200,000 for the cost of the trend campaign
  • The campaign would yield 50M impressions with creative wear-out happening at 1.5M impressions
  • The cost to create an award winning tweet is $2,000

So with the above, we would need 33 creative units/messages at a cost of $2,000 per piece of creative for a total cost of $66,000. At that cost and the cost of the promoted trend, we’ve clearly exceeded the 15% “rule.” We’re at 33% (I did the math for you). If we get our bulk discount of 50%, we’re at 16.5%. Based on my experience, when you consider the cost of the account coordinator to open the job # to the Chief Creative Officer to sign off on the job, the cost is going to be more than $750.

Keep in mind, our example is limited to twitter. We haven’t even started looking at the costs to then produce creative unique to Facebook, Instagram, Pinterest and other large social platforms, across the very fragmented digital landscape.

Screens are getting smaller and more varied.  The number of “media” channels consumers are flocking to is exponentially increasing. Consider that Vine, Jelly, SnapChat, Medium and so many others didn’t even exist a year ago. We’re adding more and more places to visit every day…yet the time we’re spending at those places is becoming more and more fleeting. To me, this means we, as marketers, get even less time to make an impact with our consumers. And that’s why you need award winning level creative every single time. You can’t deliver C-level creative experiences. They all need to be A-level. Creating A-level creative, means making your creative unique to each publisher, placement, consumer segment and person.

For years, traditional marketers have taken issue with a “shotgun” approach to marketing. The argument is that it’s too broad and not focused. Some I’ve worked with, favor a “champagne pyramid” approach to marketing, where you fill up the top glass (usually TV) and only delve into another marketing channel, after the first glass is full. This trickle down approach to marketing spend simply isn’t consistent with today’s digital world; it’s antiquated, but easy to manage.

Is it any wonder that people who cling to a champagne pyramid approach to marketing, still cling to the 15% rule; something created over 50 years ago?

I can’t tell you if the right percentage is 20%, 30%, 50%. But, I can tell you the 15% rule has no place in today’s digital world.

Now, go buy Where The Suckers Moon…now!

At iMedia Summit Old Challenges Become New For Marketers

I love iMedia Summit. It’s on my must attend list, every year. Great locations, great content and great people, make for a valuable experience.

At this year’s summit, it was clear we’re getting closer and closer to dropping “digital” from titles and org structures. We are on the precipice of people across all industries accepting, it’s less about digital marketing and more about marketing in a digital world.

As I connected with marketers across a wide range of industries, there were 3 familiar themes that could not be ignored.

Talent: The conversation about digital talent has evolved. At one of my 1st summits, nearly 10 years ago, the conversation was about getting funding to hire someone…anyone…who could be that digital subject matter expert. While we’re definitely past those days, talent remains a thorn. Today though, it’s a thorn because we need new recruitment models to find the right talent, we need a better talent investment plan to retain talent and we need a better plan for creating leaders in organizations who have a deep and wide grasp of digital.

Content: It’s king, right? Every marketer I talked with identified different challenges in dealing with content. The most consistent pain points were how to produce enough content in a financially viable way, how to safely source and share content (legal and Pinterest apparently are still not good friends) and how to distribute content the right way. With respect to distribution, this is a battle waiting of happen in a very epic way. The old model that classical marketers still adopt where your cost to create content should not be greater than 15% of the media but, is dead and doesn’t apply to digital and social content. You will spend more than $100k to create enough quality content to support a $1M ad but across twitter and facebook. In digital, unlike TV, distribution is cheap, but the content is expensive.

New Operating Models: What should you be doing internally? What should your agency’s role be? When do you bring social in-house…and do you bring it all in-house? We need new models and approaches to building internal capabilities and for setting our partners up for success. This will require our partners to pivot quicker than they ever have before. They will need new offerings, new types of talent and different pricing approaches. We are in a sea of disruption that’s not going to calm down any time soon.

This year’s iMedia summit reaffirmed some thoughts I had and offered new perspective to think about as I lead our Social Media and Content efforts for Walgreens. It’s also fair to say, iMedia once again reminds me of why I’ve stayed in digital for 16 years…the pace of change isn’t for the weak and it’s bloody good fun to try and keep up.

The Job Hopper Fallacy

“She’d be great for the role, but I’m concerned that she seems to job hop every few years.”

“He’s the perfect candidate. The only red flag for me is he seems to change jobs every couple of years.”

“I love her experience, but she doesn’t seem to stay at one place very long.”

I’ve heard those comments far too often over the last 16 years. I myself, have even uttered a similar statement, when assessing a candidate. So, yes, I’m a hypocrite, when I say, the marketing and advertising industry needs to realize, switching jobs and roles is part of today’s workforce and it’s here to stay.

My father worked for 1 company, through a few mergers, for 20+ years…till he was laid off. My mom has been with the same company since she was in her early 20s. For 30+ years she’s given 1 organization her very best. They are the exception today, especially when it comes to the marketing and advertising industry. Today, the data shows you’re fortunate if you can keep someone for 3 – 4 years in the private sector. Keep in mind that 3 – 4 average, looks at all jobs in the private sector; the average agency tenure is even shorter.

The phrase “job hopping” has with it an indelible black mark. It connotes, and we tend to perceive it, as a bad thing. People who job hop, after all, must be hopping for only bad reasons. Yet, if you were interviewing someone who told you first they went to Harvard for their undergrad, then they went to Stanford to Master in something else and then went to MIT to earn a Ph.D it something else., we’d talk about how diverse their background was and how well rounded they were. But, have 4 jobs in 10 years and you’d be potentially unstable and a flight risk.

I also believe, most talent evaluators tend to over-value a person who’s had many roles in the same company. But, is there really a difference between someone who’s had 4 roles at the same company in 10 years and someone who’s had 4 roles, at 4 companies in 10 years?

In my opinion, based on what I’ve seen and experiences, I think job hopping happens for 4 major reasons:

  1. They believe they deserve more, whatever that more is, and they know they aren’t going to get it. It could be the promotion they deserve, but won’t receive, because there isn’t a piece of business to sustain the increased level and compensation. It could be the mentorship they crave, but aren’t receiving. The more, can be any number of things, but there’s always something.
  2. In an industry that is still very much time and materials, the account funding the majority of your salary is no longer with the agency or scales back the budget significantly. In both of those situations the agency, can’t float your overhead so you’re laid off. It happens. Often.
  3. Wrong place for the wrong person at the wrong time. Sometimes it really is just the wrong place, with the wrong culture, the wrong role and the wrong account.
  4. The role outpaces your talent. The organization wants more from you, you’re not able to keep delivering the A-game needed. In this situation you have 3 options. 1, you leave on your own terms. 2, you stay, but you’re stuck at the same level and compensation. 3, you’re let go, because the organization needs better talent.

There are a bunch of other reasons that come up, but I feel like we’ve covered 80% of situations with the above.

With the average tenure of a CMO being roughly 3.5 years and every new CMO seemingly choosing to “pitch the business”, is it really any wonder people need to job hop? Then, add in the fact, employees want more training and work life balance; both of which, it’s been well documented, they aren’t getting, and the desire to hit the reset button and start fresh is understandable.

All that aside, I want to focus on two elements of evaluating why people may job hop: Title/Seniority and Compensation. I believe there’s so much more than those two, but the other elements aren’t easily quantifiable. Thanks to the mountains of big data out there, Title and Compensation are easily measured, evaluated and compared.

Only once in my career did I take a job because of the money. It was one of the biggest career mistakes I’ve ever made. I was young. It was the dot.com boom/bust days. The company imploded in 6 months. Since that mistake I’ve always used my TPRP model for evaluating opportunities. For me the role is always the most important. That said, “role” is arbitrary and subjective. Two people looking at the same role, will value it differently. That’s why I want to focus on Title and Compensation.

The following chart represents a 10 year look at the career of someone in account management in the advertising industry. Before you pour over the chart a few details first:

  1. For the starting salary of an Account Coordinator, I pulled the Chicago market advertising agency average from Glass Door.
  2. You’ll notice I also included a column showing the average salary for positions. That data also comes directly from Glass Door. To be specific it comes directly from the Leo Burnett Glass Door page. Given Leo’s long history, their size, market location and holding company status, they make for a great data source.
  3. The national average for raises in the same title is 3%. That comes from a multitude of sources. For the sake of argument and this post, it comes from Forbes and SHRM.
  4. I connected with several folks on the talent management side at different agencies to understand the average pay increase in between account roles. It’s 7%. So if you go from an Account Executive to a Sr. Account Executive, you can expect a 7% increase on average. This tracks with my experience working in leadership roles at agencies.
  5. The same people I spoke with validated the age old belief that when you leave one agency in a more junior role for another agency in a more senior role, you get about a 20% pay bump. Those changes are highlighted in green.
  6. No one stays at one agency for 10 years and just gets the average. I’m aware of that fact. A good manager at some point looks at that person’s compensation relative to their contribution and is able to get a “market correction” put in place. This essentially raises the person’s salary so that it’s still within the range of the role, while still allowing for margin to be made.
  7. This analysis is purely to look at tittle and salary. It doesn’t take into account things like work life balance, retained knowledge or other factors.
  8. An often unmentioned fact is that your first salary does dictate, on some level, future salaries. Every future increase is based on the last one, so if you negotiate 45K as an Account Coordinator, you end up better off financially, than if you negotiated and accepted 35K. That’s just basic economics.
  9. You’ll see a column that outlines the importance of #9. The “Straight 10% Increase” column, simply applies a 10% increase every year. That never happens, but it’s a good comparison point to look at.
  10. Every situation is different and this not meant to be final law.

Job Hopping Data

What does this data tell us? A lot and nothing. It tells us that there is certainly a financial and status gain to be had by job hopping in the agency world. You climb the ladder faster and earn a fairer compensation faster when you job hop. I say fairer, because as has been well-documented, when you consider the number of hours the average agency person works, their per/hour rate doesn’t seem like a fair wage. Anyone who’s worked in agency life for 8+ years, I think would agree.

What the data doesn’t tell us and what I believe is the most important aspect of the value that comes from job hopping, is the experience you gain. Today’s world moves quick. Technology changes fast. Consumer behavior changes fast. To keep up with that change, organizations continually create and try new operating and organizational models. If all the people in your organization have only worked at 1 company before they joined you, the field of view becomes limited. Take if you will the person who’s only worked at P&G his their entire career. They know 1 company, 1 culture, 1 process and 1 approach for addressing challenges. But, what if you have a person who’s worked at P&G, Red Bull, Nike and Ford? Now, you have diversity in thought. And that diversity becomes invaluable.

I believe that a patchwork tapestry approach to building “YOU” is where the future is now and it’s not going to change. Don’t attach a stigma to someone who’s sought career diversity. Instead, value the immense diverse experience they can bring to your organization.

*****UPDATE*****

On Friday, February 7th I received an email from a recruiter that stated:

Adam–

I left you a vm about 2 social media roles that I would love to have your help with. I’m helping recruit for [company withheld] (Chicago area) and [company withheld] (DC Area). I offer generous finders’ fees and from your LI profile, you look like someone who would know qualified folks. If you hadn’t just taken a new role, I would have tried to directly recruit you.

Do you want me to send you job descriptions?

thanks!
[name withheld]

I responded and offered to help. Even provided a few recommendations. The response I received back was:

Thanks for the referrals. I’ve just started working on these roles via the internal recruiters and don’t yet have the names of who the roles report to–so you know more than I do.

I checked out the profiles of both of the folks you referred, and though I trust your judgment, I’m concerned that they may look too much like job hoppers to be desirable –but with social media being a pretty new field, that may be common and not a red flag for hiring managers.

Perfect timing and one that clearly supports we haven’t evolved our perceptions about “job-hopping.”

Why Your Super Bowl Real-Time Marketing Will Fail

The Super Bowl is a peculiar American tradition. Over 100 million people will tune in, with the advertisements as big a draw as the game itself.

Last year, with the success of the tweet from Oreo heard around the marketing world, a new tradition was born: big-event real-time marketing through the second screen. Since the Oreo tweet, brands big and small have attempted to recreate the perceived success of that one tweet. No doubt, judging from what we saw at the Grammys, we’ll see many, many more attempts to replicate the “Oreo moment” during this year’s Super Bowl. And they’ll all fail.

Having been on the inside of large complex organizations and having worked at agencies challenged with leading large complex organizations, I have seen up close and personal how brands grapple with being successful in social media.

From our couches, it’s always easy to second-guess a brand’s decisions and motives for engaging in real-time marketing. It’s even easier to critique those efforts. It’s become a bit of an event within the event, for marketers to make fun of a brand’s efforts and tag their efforts with the dreaded “#fail” hashtag.

That said, here’s why real-time marketing will come up short at the Super Bowl.

Immature client organizations.
Can you believe it’s 2014 and organizations still debate the value of digital and social marketing? It happens more often than you think. When you’re part of a team trying to change that fact, you’re looking for anything to demonstrate the merits and value of your team. You’re hoping that one tweet, that one blog post or that one Vine video catches lightning in a bottle and starts creating some type of buzz internally. Buzz eventually leads to interest, which leads to funding. It’s tough to fault an organization that’s so early in their digital maturity, for delivering something not at a Cannes Lion-worthy level. That said, you probably shouldn’t step onto the Super Bowl field, if you’re playing at a junior varsity high school level. Rarely, does something good come from that.

Social media by committee.
Social media is a team sport. Since it’s the Super Bowl, let’s look at a football analogy. The Broncos have a game plan that they’ve created for the game. The coach created that game plan. Members of the team know their role in executing that game plan. Some are meant to block and tackle. Some are meant to score. When Peyton gets into the huddle, it’s his huddle. The plays are chosen by a single person, not a committee, and only one person (Peyton) can audible the play. The person you’ve empowered with the responsibility and accountability for hitting the “send” button to deliver something on brand, on message and in context is in effect, Peyton Manning. He or she needs to have their finger on the pulse of the game, what people are talking about online during the game and how their brand fits in, if at all. For them to be effective, they need to be fast. Fast doesn’t come from committee. It doesn’t come from running every single tweet by the brand team, legal team, corporate communications team, media team and so on. Fast comes from trust.

Not enough preparation.
Measure twice, cut once. When it comes to social, 90 percent of the time should be spent on preparing and 10 percent on executing. If your planning is poor, if you haven’t thought through scenarios, if you haven’t identified your goals and if you haven’t pre-created a base foundation of content to pull from quickly, you’re not preparing enough. You can’t prep on the fly. You can make adjustments to the game plan in real time, but you shouldn’t be creating your game plan as the game is unfolding.

The wrong objectives.
What’s measured matters. It’s a tried and true maxim of corporate America. If your objectives are to make sure you drive “frequency of message” or drive “reach,” you’re more apt to post often; even when you don’t have great content to deliver at the right time. If your objectives are generic (e.g., engagement) or if they aren’t aligned with the nuances of each social platform and how people consume, share and interact with branded content, they won’t be successful.

Mistakes happen.
If you’re not making mistakes, you’re not trying hard enough. Peyton Manning has thrown 219 interceptions in his career. But without forcing it sometimes, he wouldn’t have 491 TD passes. The best thing you can do after you make a mistake is learn from it and apply the knowledge to the next effort. Being able to do that is made so much easier if your organization understands that mistakes happen. And sometimes being lucky is better than being good. For example, Miley Cyrus could incorporate your product into one of her songs and all you need to be doing is paying attention to hit social media gold.

There is no formula for real-time marketing success, even if the endless debates over it seem to expect one to emerge. On Super Bowl Sunday, there are only two guarantees: One team will win, and real-time marketing won’t.

Please note, this post was also published on DigiDay.

What CEO’s Need To Know About Digital

A colleague of mine works with CEOs across a wide variety of categories. His company provides coaching and consultation to CEOs with a focus on helping them navigate an increasingly digital and disruptive world.

Earlier this week he reached out with a very inspiring ask.

A client recently asked us, as the CEO:

What are the 2 to 3 most important things I should know about digital?
What are the 2 to 3 things I should avoid (mistakes) in making digital investments in the company?

What an incredible ask from the CEO and it certainly had my wheels turning. Here’s the advice I sent back.

2 to 3 To Know

  1. Digital isn’t a vertical channel or offering. An organization that ascends to a truly high level of digital fitness is one that looks at digital as a horizontal capability that can impact everything from supply chain to marketing.
  2. In true fundamental economics principles, there is a limited supply of exceptional digital talent. With limited supply and high demand, you need to rethink your entire recruiting, compensation and retention model. For example, at Twitter, the highest paid non-exec makes over $1M a year; he’s one of the smartest technical engineers in the world. The battleground for exceptional digital talent is serious business and one that most orgs aren’t equipped or prepared for.
  3. The space moves so fast. Your 5 year business plan approach doesn’t hold water in a space, like digital, that changes completely every 18 months. You need 5 year horizons, 3 year strategic plans and 18 month action plans.

2 To 3 To Avoid

  1. You can’t fix your digital capability gap by spending more. Spending more in digital, without the right internal capability and talent is an instant recipe for disaster.
  2. Building on my #1 of What To Know, you have to avoid making digital seem like a “thing”, “experiment” or the responsibility of 1 team or department. It needs to be something that every Sr. Exec is thinking about and investing in.
  3. Picking the wrong partners. Seems simple enough, but again, without the right internal champion vetting partners, who’s really capable of separating pretenders from contenders? If you pick the wrong partner (s) it can set you back and have a catastrophic ripple effect on your organization.

I’d love your feedback on my list and certainly welcome your thoughts on what CEO’s should know about digital. I’ll aggregate the feedback and pass it on to my colleague.

About
Digital dad to Cora and John. Love ironing, bourbon and BBQ; no necessarily in that order. Living life, like I stole it. I'm always up for a

spirited conversation. These are my thoughts and ramblings, not those of my employer.
Learn More »