By Dan Leibson, Relevant Ads
Google’s recent Penguin 2.0 update affected 2.3% of English language queries. Despite significant foreshadowing during the weeks leading up to the rollout, a large number of affected websites obviously were not able clean up their link profile in time and were hit with an algorithmic link spam penalty. This is what happens when you can’t act rapidly.
Web marketing is a rapidly changing field. It’s important to have a process that allows for rapid adjustments and focuses on rapid iteration. Basically, systems are needed where experiments can be implemented and measured at the speed at which the research methodology and environment changes. As web marketing professionals in a constantly changing landscape, we can’t have processes that need to be re-done constantly while we adapt to a rapidly changing environment. For this reason, agile/lean startup methodologies are super hot topics in marketing communities right now. For those of you who are unfamiliar, I will give a quick overview, but I highly recommend watching this Whiteboard Friday video by Jonathan Coleman that gives a quick rundown of agile marketing.
Agile started out as an offshoot of extreme programming that put the developer at the forefront of development work (shocking right?). If you are interested in the origins of agile you should check out the “Manifesto for Agile Software Development.” Meanwhile, The Lean Startup, a book written by Eric Ries, combines some of the methodologies of agile with customer development as well as some of the tenets of lean manufacturing. I highly recommend reading The Lean Startup as well as checking out Jim Ewel’s website.
One quick concept before we start:
Do you persevere or do you pivot? This is a core concept to know when it comes to lean startup methodologies. The basic point of agile/lean is to ensure you are in a feedback loop where you are constantly working to iterate on your existing projects or develop new ones. Ries calls this the ‘Build-Measure-Learn’ cycle. After the learning component, you have to decide if you are going to continue to iterate down your existing path, aka persevere, or if you are going to alter your tactics and strategy because you aren’t seeing the results you would like: a pivot.
Prerequisite: It’s Okay to Fail
In fact, this is probably the most important part of implementing agile/lean and may require a major cultural shift within your organization (or for your clients). Sometimes we end up on “Failure Road,” and that’s okay, as long as there is knowledge gained from the failure that propels you towards success. However, very few managers want to tell their people that it is okay to fail. If you are in this position, then you must advocate for this change. Luckily, there is some correlation between telling people it’s okay to fail and subsequent success that you can share with people. In my experience, the resistance to this concept is based around the misconception that you are encouraging failure. This is not true. What SHOULD BE encouraged is people putting their best ideas forward, to think experimentally, and to try risky things. If people are worried about failure, they will only put forward sure things, which MAY succeed more often, but generally won’t be huge successes. When people believe they won’t be punished for failure, they will contribute more. Remember, you don’t have to succeed at first, as long as your efforts help to iterate or pivot toward the success you want.
Tip # 1: Establishing Metrics
You may think this is obvious, but I am always shocked when I hear people talk about the metrics they put in place before launching/changing a marketing campaign or altering something on their website. In today’s online marketing climate, data is pervasive and should be used to determine not only relevant success metrics (KPIs) but also metrics that may help determine if it’s time to pivot. This is particularly important for the current wave of search marketing, where data are already being used to drive decisions. Marketo has a useful guide on marketing metrics, and Joanna Lord has a great post about marketing analytics that I highly recommend. If you need more help, the amazing Avinash Kaushik lays out his best web metrics for you in this blog post. The best piece of advice I can give here is both tried and true: stay away from vanity metrics.
Tip # 2: Bias towards Action
Don’t let the focus on metrics and data bog you down with analysis paralysis. A key component of agile/lean is continual development, so it’s important to constantly try new ideas out, measuring them and pivoting/persevering based on the success of the strategy. Keep in mind, having a bias toward action is not the same as regularly throwing stuff at the wall to see what sticks. The latter is not a strategy; it is a tactic that results from a lack of a strategy. A tactic I have used to help create a bias toward action is “Unless I Heard Differently.” This technique is based on putting your best hypothesis forward and explaining the reasoning behind it, while at the same time preparing to implement it in accordance with your internal agile/lean processes if you don’t hear back from key principles. I understand that all teams are different, so your mileage may vary, but I would urge you to check out the website linked above and give the technique a test drive.
Tip # 3 Use a Canvas
Canvases stemmed from some of the problems surrounding business plans, project plans, and product requirement documents. They are one-pagers that generally take less than an hour to create and are designed for constant updating and manipulation. There are tons of lean and agile canvases out there. Seriously, there are lots. My personal favorite is the Lean Startup Machine’s Validation Board. It can be used for search marketing with very little adaptation. Another good choice is Jim Ewel’s marketing model canvas. Whatever canvas you chose, the important thing is to treat your canvas like a living document. This is important because the decision to pivot can be a tough one, and when you have an easy-to-consume document that contains all the relevant data, getting buy-in for that decision will be much easier.
Tip # 4 The 5 Why’s
A tactic Eric Ries talks about in The Lean Startup I have found to be incredibly useful is asking “why” five times to get to the root cause of an issue. Here is a great article about the tactic. Not only does this process help to get to the root cause of problems, but it helps break down transparency barriers and teach people that failure is okay as long as you work rapidly to correct the problem and drive toward success. This is one of the most successful tactics I have implemented to get to the bottom of issues surrounding poor performing campaigns, or a lack of proper training.
Tip #5 Using an Agile Board
Don’t try to manage projects in a spreadsheet or worse, Microsoft Project. An agile board is the preferred project management system for agile/lean and for good reason. It is designed to work best in an environment where there is constant iteration. It is different than a canvas, at least in the way I use them, because the canvas is very high level while the board is for project management. You can use a wall and a bunch of sticky notes or you can use a software program. Trello is a free agile project management web application that is pretty amazing. At Relevant Ads we use LeanKit, but there are a lot more available for choosing if you find either of those options lacking. Personally, I would recommend Trello. Speaking of which, Miranda Rensch gave a fantastic presentation on agile at MozCon 2013 and created this work sheet for creating your own agile board using Trello.
What it Looks Like When You Are Doing it Right
A great example of having a bias toward action is Oreo’s Super Bowl tweet. In case you haven’t heard about this, there was a blackout during Super Bowl XLVII that delayed the biggest American sporting event of the year. While the stadium was still dark Oreo put out this tweet:
They developed messaging to a real-time event, had creative designed, copy written, and went live. The result of this rapid development process for their marketing campaign speaks for itself.
Look at all the top tier sites that have embedded this tweet! This tweet was retweeted by 11% of Oreo’s followers. Granted they are not all Oreo followers but we are going to use that number, retweets divided by followers, to normalize disparate follow counts and act as a proxy for reach. For example, it was retweeted 101 more times than this post by Rand Fishkin announcing the rebrand of SEOmoz to Moz.
Those 157 retweets only represent .09% of Rand’s “reach.” On the other end of the spectrum, the following tweet by Barack Obama announcing victory in the 2012 presidential election is the most retweeted post in the history of Twitter. It was retweeted 50 more times than Oreo’s post, but it accounts for only 2% of his “reach.”
Bringing it back to agile/lean, do you think the copy for the Oreo tweet was the first copy that came to mind? Or that the graphic design of the image was the version that got done the fastest? Absolutely not! They have an internal process that had been perfected during the 18 months prior. They had some early success like the infamous gay pride post that they iterated on to build a process so finely tuned that in the matter of minutes they were able to put out one of the most memorable ads of the Super Bowl. For free, on Twitter.
A Couple of Final Takeaways
It takes time to get your agile/lean processes to the point at which you want them. Post-mortem reviews are critical because it takes a while before hitting your stride. This allows applying the same methods applied to campaigns and tactics to be applied to your agile/lean process and allows you to get to where you want to be quicker.
Do you implement agile/lean in your company? I would love to hear from you regarding successes and/or horror stories in the comments below.