Posted in Uncategorized

Online Student Again, Part 9: Web Analytics & ROI for Better Decision Making–Decisions, Decisions

I guess the Doctor isn't that great in figuring out KPIs or reading Google Analytics.
I guess the Doctor isn’t that great in figuring out KPIs or reading Google Analytics.

Anytime I think of any kind of tech comm analytics, I don’t think of Google Analytics or Web Trends, but my mind races back to the first time I heard Mark Lewis speak and how my mind was blown at the idea that tech writing was measurable in any form.  Now, a few years later, I’m looking at this latest module in my online Digital Marketing course at Rutgers about Web Analytics and ROI, as taught by Rob Peterson of the marketing firm, BarnRaisersLLC.

The idea of analytics and ROI (return on investment) sends a shiver down my spine. While I understand the function of analytics and some basics, it makes me think of complicated mathematical formulas, and that in itself makes me anxious. (I was a decent math student in school, but it was not my forté.)

Peterson started his course with a quote that I could swear I’ve heard either Mark Lewis or Joe Gollner repeat (I think it was Joe) by Peter Drucker to frame the objective, in which Drucker said, “If you can’t measure it, you can’t manage it.”  So, the first question was, what is “it”? “It” could be almost anything.  But Peterson feels that “it” is success–that your success (or lack thereof) is what needs to be measured. He also felt that the verbs “measure” and “manage” were the keys to understanding the material for this module.

Peterson broke down these by 6 steps to demystify web analytics:

  1. Know why your site exists
  2. Identify who you want to attract
  3. Find out how they find your site
  4. Know the action(s) you want them to take
  5. Create an actionable scorecard aka figuring out your KPIs (Key Performance Indicators)
  6. Listen to your digital ecosystem

You want to do these steps because the buying cycle has changed due to the internet. The buying “funnel” has been replaced by a flatter, more circular course of evaluation and re-evaluation before buying, but there is a subsequent loyalty to a brand after the initial buy that shortens the cycle the second time around.

Petersen then talked about how to make your digital ecosystem thrive, describing it as the intersection of earned (sharing), owned (web properties), and paid (advertising) media.   People come before software in analytics, he insisted, stating that there’s a 10/90 rule for analytics, namely 10% of the marketing budget should go towards the analytics software, and 90% towards getting good people to analyze the results. We use measurement to understand our role, which is understanding the customer journey with the goal of optimizing the customer experience and find more customers.

To do this, there are several (he said 12, but I could figure out which were the specific 12) measurements to know.

First, you have to understand how consumers find you. This is primarily done through keywords and links.  It was emphasized that page rank (aka “SERP”- Search Engine Rank Page) mattered because the first listed item in the ranking outweighs subsequent 5 ranks combined, usually.  The metric that mattered most with paid media? He gave these two equations to show how this worked:

CPC (cost per click) = CTR (Click thru rate–this makes the click relevant)

CTR= clicks/impressions

He noted that a good conversion rate is actually low–2% is considered good!

So, once you’re at the website–now what? To analyze this, you need to look at sessions and users (which were recently renamed by Google, and formerly known as visitors and unique visits, which I understand better). Usually the period of time usually measured is looking at the last 30 days.  An important metric in this is looking at the bounce rate, which is a percentage that means the number of people who only viewed one page then left. It’s important because it shows if the site is relevant or not. A good bounce rate depends on your goal or objective. If you wanted people to come to a specific page to sign up for something only on that page, then you achieved your goal, but if not, then you have to figure out why they didn’t stay. Petersen noticed that if you notice that your bounce rate is high for the wrong reasons, it’s not easy to change a bounce rate overnight.

One also has to understand where users people come from and what they do. Traffic resources are organic, direct, or paid searches,  but can also be referral and social media sources.

The focus then turns on the key content (aka the webpages) by determining how many take the actions that you want. Conversion, here, is an important metric. There is the macro conversion, which is revenue generating, such as person to person, B2B, B2C, WebEx to WebEx, and the like, versus micro conversion, in which users would subscribe to the blog or a newsletter or participate by making a comment on a blog, and so on. This can also be done through word of mouth, more specifically “likes”, comments, shares, people talking about the site, followers, retweets, reviews and rating, sentiment, text analytics, and email open rate. The conversion rate is determine with this equation:

Conversion rate=(Desired outcomes/total visits) X 100

There are several online tools to help you listen that are generally free tools and work better on bigger sites, but easily available to use, according to Petersen:

  • Google Trends – This tool uses keywords. It can compare two different topics to see where they’ve been and where they are predicted to go. The example given was the buzz between Justin Bieber and the anticipated Samsung Galaxy 4 among teenagers as potential buyers. Fortunately for Samsung, the Galaxy 4 was trending more than Bieber!
  • Compete.com – This is a paid tool focused on doing competitive intelligence. It lets you see how you compare to your competitors so you can figure out your strategy.
  • Alexa.com – This is another tool that looks at the competition, powered by Amazon. It can use to compare your competition by providing ranking and metric information.
  • Marketing.Grader.com – This is a HubSpot tool that grades your website on several levels from social media marketing, blog posts, SEO, lead generation and mobile. I tried this tool and this blog got an 82/100, mostly because I’m not really trying to generate leads, and the mobile aspect of using WP needs help, so I considered that pretty good. I like this tool because it was really easy to use and understand. At the time of the recording of the lesson, Petersen’s business website had an 81! And I had an 82? Hah! I must be doing something right!

The course continued by using a case study using Google Analytics. The objective of this part of the lesson was to learn what to look for and how to use it. It was pointed out that it only will look up your own company due to admin rights, and to make it work optimally, you need to get specific code and embed it in all of your page (perhaps in the header, for example) so that Google Analytics can track it correctly by the correct administrators.

The main focus in using Google Analytics is looking at the audience figures. Look at the Engagement section to see who is really spending time on your pages. It helps you to understand the bounce rate. Petersen pointed out that on average, 95% of the audience never views more than 5 pages, and 95% don’t spend anymore than 5 minutes on a website. You can also look at location for geographic data, drilling down from national to town level. The section labeled “Mobile” can let you see how the site is being accessed.

This information helps to frame the marketing “funnel” that is often talked about, which is where marketers start with the action of acquisition, narrowing the focus to engagement, which further narrows to conversion. If it circles back to the engagement, this signals that there is loyalty to the brand, and this cycle begins again.

Google Analytics can also identify how your audience finds you, which is mainly through acquisition, behavior, and e-commerce. Behavior metrics can show what pages they are going to. Channels section knows keywords even if analytics don’t tell, as it shows keywords that people are using to access the site. You can find out about keywords from the Site Search section. All this information helps us understand the conversion rate by allowing us to see what’s been bought and look at average order value if you offer goods on your site.

For better information on keywords, Google Webmaster Tools is a better choice. It can tell you if your website is set up correctly so that the webcrawlers can find you, and helps correct errors on your site, providing rich depth of information on keywords.

So, you have all this information–what do you do with it? You do a lot of testing, because it’s an activity and a philosophy in which you build and test and repeat the cycle, using lots of small steps continually. Surveys are one way to do this, as they are essentially a tool that acts as the “voice of the customer”. Surveys use simple questions regularly executed such as, “Why did you come to this website?”, “Did you find what you were looking for?”, “Would you come back?”, and “What would make your next visit better?” There is also what is known as A/B Testing, in which you show two different versions of an ad or webpage to customer groups, and then seeing which people went to the goal page based on the two different models. A third way to find out some of this data directly from customers is looking at reviews, as they help searches and sales for other customers.

This is when the talk steered towards the ROI part of the discussion. Petersen started by letting us know how much a “like” on a site was worth. Research showed that among 150 brands researched, it was worth $71.84. Fans of a brand are 28% more likely than non-fans to continue using the brand, while fans are 41% more likely than non-fans to recommend a fanned product to their friends.

What does it tell us about “Like”-ability? It tells us how we can learn from social media audiences, which can be achieved through Facebook surveys and provides sentiment analysis.

KPIs (Key Performance Indicators)are crucially important as they are the “scorecard” needed to keep a strategy on track. Petersen defined KPIs using a quote by Avinash Khaushik of  Google, saying that KPIs are “Measurements that help you understand how you are doing against your objectives.” Petersen also used the quote by Laurence Peter, saying, “If you don’t know where you’re going, you’ll probably end up somewhere else.” This, Petersen decreed, is where you define the “success” using metrics to measure.

There is a difference between KPIs and metrics, namely that a KPI is a metric, but a metric is not a KPI. More specifically, KPIs:

  1. Relate to a business objective
  2. Are chosen by the people accountable
  3. Provide context by being tracked over time
  4. Are based on legitimate data
  5. Are easy to understand
  6. Create meaning that gives control
  7. Require action

The role of KPIs, per Petersen, is that “KPIs are an actionable scorecard that keeps your strategy on track. They enable you to manage, control, and achieve desired business results.”

How do you choose the best digital marketing measurements? Start with a KPI scorecard that compares raw numbers against progress against the percentage of change.
For example:

Numbers Progress % Change
Example # of new customers 95% complete 22% increase in sales
source CRM project plan P&L
Frequency Monthly quarterly monthly

From there, you create a dashboard for the scorecard that should include both web and offline metrics that looks something like this:

KPI 1st Qtr 2nd Qtr 3rd Qtr
Sales Metric #1
Sales Metric #2
Website Metric #1
Website Metric #2
Social Media Metric #1
Social Media Metric #1

The KPI dashboard shows key areas and results from metrics, and can help you to figure out what the key points are to create a KPI report for management.

Some Metrics for KPI consideration:

  • Before: Keywords, keyword trends, indexed, links, authoritative links
  • During: visitors, segmentation, bounce rate, traffic sources, key pages, conversion rate, average value
  • After: reviews, surveys, A/B tests, social promotions

Starting from KPIs to deriving ROI is about the “Show me the money!” You need to look at the results from latent effects to direct effects. Direct online effects usually make up 16%, latent online effects makes up 21%, while latent offline effects makes up 63% of the results.

What is ROI (Return on Investment)? The equation given for this was:

(gain from investment-cost of investment)/cost of investment ($)=ROI
ex. ($500,000-$100,000)/$100,000=400% or 4-to-1 ROI

ROI is important because it reflects the idea of good management of money. Examples of gains and investments include:

  • Gains: sales increase, shorter purchase cycle, more leads, higher close rate, lower internal operating costs
  • Investments: marketing, advertising, promotion, PR, customer service, staff, overhead, time and energy

Metrics we use to create both annual and lifetime customer value and help us be better marketers include:

  • Mass Marketing (Mass media, like TV, radio, magazines, newspapers, billboards — create awareness, interest, trial, sales)
  • Direct marketing (leads, conversions, retention, sales)
  • Digital marketing (unique visitors, downloads, register, redeem, convert, buy)

What ROI calculation can and can’t do:

  • Can: identify direct effect, provide relationships between direct and latent effects, give insights how to drive ROI higher, define consumer value
  • Can’t: define ROI goals and expectations, establish a baseline, identify a timeframe

What’s in store for the future? More data (aka BIG data) is growing exponentially coming in. “Big Data” is large amounts of data from web-browser trails, social network communications, sensor and surveillance data that form unstructured data stored in computer clouds, not servers.

The course was concluded with the statement, “It’s not the data, it’s what you do with it.”

Overall, this module was pretty good. I feel less anxious about web analytics and how to analyze the information provided, and I now have some more robust tools to use as well. The only thing I didn’t like about this module was that while Mr. Petersen was obviously knowledgeable about the topic, the structure of the module wasn’t completely clear. At the top of the course, he said from the beginning that there would be a top 12 things, but they really weren’t defined as “this is #1, this is #2, this is #3,” and so on. While there was a good progression, part of this came off as scattered because I felt like I didn’t understand how he was going from Point A to Point B and why, and no clear outline on how he planned to cover the lesson. Call it the content strategist in me, but that structure was something I just needed. Other than that, it was a module that I definitely needed for a better understanding of the topic, and the information was sound.

The last module is coming up, which looks like it will tie all the previous modules together! Until then…

Posted in Uncategorized

Adobe Day @ Lavacon 2013: Scott Abel’s 5 Technologies Tech Comm Can’t Ignore

voodoodonutsignI realized as I was writing this post that this would be my 500th post on TechCommGeekMom. Who knew that so much information and thought could accumulate through original posts and curated content?  I’m also very close to my all-time 15,000 hits mark (only a few hits away at this writing). I wouldn’t have believed you if you told me that I’d hit these benchmarks when I started this blog, but of course, I’m going to keep going! I debated about what I should write for my 500th post–whether to finish my Adobe Day coverage or do something else, and in the end, it seems fitting to finish my Adobe Day coverage, because in many respects, knowing and writing about the presentation of Scott Abel, aka “The Content Wrangler”, shows how far I’ve come already in my tech comm journey from beginner to covering internationally known presenters.

Scott is one of the most prolific and vocal speakers out there on the conference circuit speaking about content–whether it be content management or other technical communication topics.  It also seems like he has written the forewords of many of the best tech comm books out there. He’s everywhere! To boot, he’s an accomplished DJ, and I found myself “bonding” with him over dance remixes and mash-ups while at Lavacon, because I always enjoy when he posts either his mash-ups or his favorite mash-ups on Facebook. (I’ll be writing a post about the relationship between tech comm and dance mash-ups in the near future.)  He is a person who is full of so much kinetic energy that you wonder when he’s going to explode, but he doesn’t. Even the time I saw him at the STC Summit last spring with a bad cold, he was still more on top of his game than a lot of people would be on a good day.  Much like Val Swisher, my love for all things Scott Abel also knows no bounds.  He knows how to stir things up at times, but there is no denying that in his frenetic pace of delivering a presentation, you learn SO much. I’m lucky that he’s so kind to be one of my cheerleaders!

ScottAbel
Scott Abel checking his files before his presentation

So when it came to thinking of a garden in Portland to use as an analogy to Scott, I had to deviate. In my mind, he’s the Voodoo Doughnuts shop located about four or five blocks away from the Chinese Garden. Scott’s talks always have lines going out the door, and like many of the Voodoo Doughnuts themselves, the unique flavors dispensed open your mind up to new and delicious possibilities and ideas, and you come back wanting more (hence, more long lines!).  They are both crazy and sweet at the same time. You can’t beat that combination.

Scott was the keynote speaker for Adobe Day as well as the moderator of the discussion panel later in the event. Scott’s topic for his talk was called, “Five Revolutionary Technologies Technical Communicators Can’t Afford To Ignore.”  If Joe Gollner went fast during his presentation, then Scott went at lightning speed, so my notes below are the highlights.

Scott started by telling us that translation is going to be an important part of automated content going forward. It’s important to understand that for the web, the World Wide Web (WWW) is equal to the “land of opportunity.” The WWW can reach a global market reaching new consumers. As American users, we forget that 96% of web users are not in the US. We don’t all speak English globally. In fact, less than 6% of the global population speaks English well, but don’t necessarily read or write it well.

Scott’s list of the five technologies the Tech Comm can’t ignore were as follows:

1) Automated Translation
Why would be need automated translation? We write for the *worldwide* web.  There are over 6000 languages in the world, so translation is a big deal for a global reach and global connection. We need to recognize that content is written for both machines and humans. Even though we write for both machines and humans, we need to write for machines first, as they are the “gatekeepers” of content, such as for searches. Everything goes through the machine first. We need to recognize that writing rules learned in elementary school are no longer sufficient for a world in which language science is needed.  We need to examine our content from the vantage point of a rules-processing engine and ensure it’s optimized for machine translation.

2) Automated Transcription
Automated transcription involves software that translates speech to text for machine use. Without transcription, content is locked and hidden from view. Transcription allows for better searchability of content.  Scott recommended Koemei as a good transcription software tool for video and general transcription, as it can help transform editable content into other languages.

3) Terminology Management
Terminology management controls words in a central place, namely the words used the most and used consistently for branding, products, etc. Terminology management is important for consistency as well as for regulatory reasons. This is an instance where seeking a global content strategist is needed to help standardize processes.  It’s best to adopt a terminology management system, such as Adobe partner and Scott’s suggestion, Acrolinx.

4) Adaptive content
Adaptive content is content that is structured and designed to adapt to the needs of your customer; it’s about substance of the content. Adaptive content adapts to the devices, e.g. laptops, GPS, and smartphones.  Customers are demanding exceptional experiences, so we need to meet their expectations, so it’s up to responsive designers to meet that challenge. Adaptive content makes it possible to publish to multiple platforms and devices.  It is content separated from formatting information. By allowing authors to focus on what they do best, adaptive content makes content findable and reuseable by others who need it. We need to rethink content, as the move to adaptive content involves work, but the ROI (return on investment) can be realized in months instead of years.

5) Component Content Management
Component content management systems are needed. They focus on the storing of content components that are used to assemble documents. Components can be in all sizes, and can be photos, video, and text. It’s about managing CONTENT not FILES.

Scott provided these slides as his example to show this:

ScottAbel_ExampleA ScottAbel_ExampleB

Structured content, combined with a component content management system, supports personalized content and  targeted marketing, which in turn increases response rates. In this end, this process can save money! The key is to remember that all customers are not the same! Reusing content without the “copy and paste” methods produce the best results. You can ensure that content is consistent by seeking a content strategist who understands content and is a technologist. Implement a component management system. Scott suggested checking out Astoria Software for a good component content management system. 

At this point, Scott’s talk had pretty much finished, but in answering audience questions, he pointed out that there’s a lot more than just these five technologies to watch. He suggested that we should look out for wireless electricity, flexible surfaces, more wireless devices, wearable computing, and augmented reality as well. He also said that in order to mature as a discipline, we need to be content craftspeople, content designers and content engineers. We need to leverage using content and code. We need to think more like engineers, and less like writers and editors. Even websites that are very localized still need to be written for global purposes to improve the English used for the native speakers as well. Controlled vocabulary helps all end users!

Scott covered a LOT of information in a short amount of time, and he set the tone for the rest of the session, as the presentations that followed repeated much of the same information. (This is a good thing, because then we know that the information is valid, coming from several experienced technical communicators!)

Scott posted on Twitter than his presentation was available on SlideShare, but I have it below.

And as always–Scott, if I misinterpreted or misquoted any of the information I summarized above, please let us know in the comments!