I’m going to be attempting my very first live blog today. Please be gentle. :-)
#ACCELERATE13 is a conference about Big Data, analytics, and everything else they can fit in about measurement and customer engagement. ACCELERATE originally started in 2011, founded by Web Analytics Demystified. This year it looks like it’s around 300+ in attendance.
I’ve been attending Web Analytics Demystified in Columbus, Ohio for a few years, now, and it’s simply a great group. If you’re interested in checking it out, they meet once a month on a Wednesday (natch). Free food, excellent presos on analytics, and a whole lotta fun.
Just shared this excellent article: What Keeps The Chief Marketing Officer Awake At Night – Part 2. Really great article.
A coffee break, and then at 9:00 AM, we’ll be off to the races…
8:58 AM
Nationwide’s CMO Matt Jauchius is getting ready for his Q&A with John Lovett, Senior Partner, Web Analytics Demystified. I have a feeling we’ll be hearing some of the same thoughts that were mentioned in Mitch Joel’s article from Six Pixels of Separation.
9:08 AM
We’re starting “fashionably late” (ha!). Found out it’s actually 400 folks here. Excellent.
Eric T. Peterson, Senior Partner and Founder giving the intro. He’s very excited by the turnout. :-) He’s talking about the history of Web Analytics Demystified. They’re the only organization in the world to new analysts who are breaking into the analytics field.
Eric just had everyone stand up from the West Coast, East Coast, Midwest, and overseas. Someone actually traveled here from Dubai. Whoa.
9:19 AM
Introduction of Matt Jauchius. He started off with a degree in accounting. Then he became enamored with marketing and went to Michigan to start out. Hooked up with McKinsey. This was a certain type of marketing. Very analytical, different from P&G and most companies. From McKinsey, he went to Nationwide and has been the CMO for three years.
JL: Where does digital analytics sit?
MJ: Digital analytics sits within marketing. I create a segmentation of duties vs. the analytics folks who focus on results.
Okay… not a court reporter here. I’ll try to summarize thoughts a little better…
Making people aware and bringing people to Nationwide’s website is the function of marketing. You have to have good partners and work well with them to getting online applications through.
Nationwide has over 2000 marketing professionals. Again… whoa.
According to MJ, it’s important to know the topical matter that you bring to the CMO. The marketing involves tens of millions of dollars. Don’t just focus on the numbers. Ask yourself “So what?” Don’t lead with the numbers but the “So what?”
The numbers often make the C-Suite feel clueless. Interpret it into English for the C-level folks.
The analyst who can think of that second step, what this information means to the rest of the organization, will bring true value. MJ says this is very hard to do.
What MJ’s team is working on: credibility, creating “pull,” a forward motion toward more engagement. They invested tens of millions into digital marketing. The demand for Nationwide comes from customers wanting more information, which 80% starts on the Internet, but then continues through to agent involvement.
Nationwide got involved with NASCAR. 70 million NASCAR fans. They wanted to have regional strength and NASCAR fans are brand-loyal. Digitally, the fans are savvy. They use several platforms to access the NASCAR developments. Nationwide sought brand awareness. There was no difference between those who knew about Nationwide and those who don’t. Six years later, there is a huge difference and most fans now know about Nationwide and have become customers. Everything is measured both online and offline.
The rise of data scientists capturing this data will become increasingly a competitive advantage. You have to be careful with how much the consumer knows… they don’t want to be stalked. The companies who are able to do this better than their competitors will win out in the end.
Talking strategy… what’s the roadmap? It’s important to have a strategy but you have to be careful. The role of the strategist is to prepare, sort of like the Marine Corps. Have a strategy but don’t think of it as set in stone. View it as a creative process.
Making analytics results available to the rest of the organization… it’s hard. Even if it’s clean and well-organized, it often creates more confusion. There are pitfalls that can happen. When those who aren’t trained try to analyze the data, they often don’t have the assessment skills needed to interpret the data.
There are tools and methodologies. How do you know the data is secure? (and other questions need to be asked) Data has value. Insights are worth gold. Mining insights is a profession. It will continue to evolve. Analytics is a profession.
Nationwide has a Chief Analytics Officer (CAO). Hmm. “Cow?”
John had a meeting with a CMO last week. The CMO wanted data that would help make the right decisions to move forward. MJ isn’t that dogmatic. He thinks it’s important to view the data in non-traditional ways. It’s both a science and an art. Digital marketing is one of the more scientific areas for an organization. The art is there, too, though.
How can you impress your CMO? If you get in front of the CMO, but don’t be afraid to let your personality shine through. Be yourself. Speak with as great of an economy of words as you can. Then stop. Let them pull on you. If you show up with your points synthesized, deliver it, but allow the senior person to ask the questions. Get to the point and then stop talking.
If the CMO asks questions, they’ll follow up with more questions. But if they look at you with silence, go ahead with more information.
9:51 AM
Next speaker:
Ten Tips for Presenting Analytical Insights to Senior Leadership
Jeff Leonard, Vice President, Digital Marketing Strategy, Wells Fargo Advisors
We definitely want to make senior leaders happy… :-)
#1: Know what drives your business. You have to be thinking about what drives your business. Either you’re trying to drive more business, create loyalty (clients who stay with you for a long time), which means more money.
#2 Understand your audience. How do they want to consume data? (flashing a slide of Tom Cruise shouting “Show me the money!” Love that movie…) Jeff says that Nationwide is very methodical. Know what kind of information your audience likes to consume. Think about who you’re presenting to. At Wells Fargo Advisors, it’s about storytelling. Your numbers have to fit into a story that will resonate. Wells Fargo is trying to get new clients (as everyone else is) but their competitors are growing.
#3 Start with a POV. Where could they change or revise to allow a customer to access them more easily? The original locator tool was not delivering. So they changed to more of a “match.com” approach. They realized they needed to be more of a configurator of the space rather than a “plug in and we’ll give you an adviser.” The customer now vets the early decision process online.
#4 Be open to the unexpected. They realized that they had a Facebook potential powder keg. An advisor was found doing a keg stand (and he’s 34 years old). How do you participate in social media but be effective? At first Wells Fargo prevented social media participation.
#5 Build a story around key insight. Instead of watching, dancing. This is what the consumer wants. The advisors need to learn how to dance with the customers.
#6 The appendix is your friend. Use it. Simple = good, complex = bad. The C-suite likes knowing you thought it through but don’t drag them through every stage of how you got there. Economy of words (hmm… where have I heard that before?). If your message is going to resonate, you have to realize you need to package it well.
#7 Make the data real. They checked out adviser profiles and realized not many were on LinkedIn. Most investors expect to be able to easily find information about the advisers online. LinkedIn is the easiest way to do this. Only 27% of the advisors have a profile and just 35% use best practices. Many didn’t even have a photo posted for their LinkedIn profile.
#8 Highlight success stories. Wells Fargo interviewed the advisers who were successful using LinkedIn. One reached out to an old classmate, they connected. His mother needed an adviser and within two months, had $2.5M of new business. (awesome!)
#9 Probe for questions. You need to pause when presenting to senior leadership. They have about a hundred people a day vying for their attention. They’re trying to process the information you’re giving to them. Sometimes their silence means they’re processing. This is the time to prompt them, “Is this making sense to you?” It’s a great opportunity.
#10 Ask for what you want. Don’t be afraid to ask for the business. Think of yourself as a salesperson. You’re a professional. At the end of your presentation, say, “This is what I need from you. This is what I need to act upon these insights.” Don’t walk out without asking for what you need to take it to the next stage.
Ending thoughts:
- Focus on business drivers.
- Know youre audeince (and their hot button)
- Use stories
- SIMPLIFY, SIMPLIFY, SIMPLIFY
- Be clear about what you’re asking for
10:14 AM
Next speaker:
Ten Tips for Presenting Results to GET Results
Lea Synefakis-Pica, Manager, Digital Insights & Analytics Projects, Prudential
Working hard… you want to present some great insights. You may realize that you don’t have everyone’s undivided attention. Your insights may have just flown into a “black hole.”
Here’s how to get their attention…
She’s going to be sharing lots of resources. I already love her.
@leapica
Think: what is it they really want? (Your senior leadership) They may realize development is important but not sure where to allocate funds. Ask, “who else is attending this meeting?” The higher level the answer, the more you have to tailor your tone. (Don’t get too technical too fast)
#2 Know your stage. Visualize delivering your performance. Understand your venue.
#3 Use your presentation tools wisely. You can go wrong with PowerPoint. (bo-ring… yep.) She just showed a slide that had a bunny and a hot dog on it as a bad example. But the slides aren’t the problem. We’re the problem. We’re not using the tool right. It ends up hurting us than helping.
#4 Unload the bullet points. The only person who wants to see all the bullet points at the same time is you, the presenter. The points usually show way too much information at once. Why do we love our bullet points so much? It helps us but it becomes a crutch so we don’t have to prepare. Remember: your slides are for your audience, not you. They’re not a script. Your audience wants you to talk to them.
Slide:ology (good book)
#5 Create ideas, not slides. The author’s firm believe is that each slide should only communicate one idea at a time.
Take one bullet and blow it up to a collection of ideas.
Once you stop being trapped by a collection of slides, you’ll be able to explore more your ideas.
#6 Resist the fluff. The “fluff” lives on your slides. Logos, watermarks, doodads that don’t add any value to your slides. The bottom company logo really doesn’t add value to the slide and prevents absorption of the idea.
Presentation Zen by Garr Reynolds (another good resource)
Think about designing your slides like a designer would. Only 7% of presenting professionals study design (have a background in graphic design). But any analyst who presents would serve themselves well by studying a little design principles.
Get rid of the clipart!
#7 Harness the power of real imagery. Full color, high def, high res imagery.
Brain Rules by John Medina (another book resource)
Vision is the strongest human sense. Imagery increases recall. Images can stir emotion. How to stir emotion about web analytics? Use an emotionally evocative image to demonstrate a customer’s struggle. Use dramatic text against imagery.
Data IS imagery.
#8 Choose the right visualization. Data can be sliced and diced in many different ways. This is a judgement call for you as an analyst. Consider using charts. The Pie Chart, though, always seems to go wrong. Execs see them all the time. But pie charts are like pie and done wrong, can cause indigestion. (slide of a pie segmented in really tiny slices).
Your brain has trouble deciphering the data in the pie. If you cannot ignore the pie, learn to do it right.
#9 Doing the pie chart right. Learn data visualization best practices. Kill 3D. Sort segments so it makes more sense. Use simple labeling. Also use strategic coloring to make sense. (showing slide of both light blue to show Apple mobile devices compared to the rest of the manufacturers, which are in gray.)
WSJ Guide to Information Graphics by Dona M. Wong (book resource)
#10 Test and refine your approach.
5 things to do next week:
- Beef up your reading library
- Watch a TED talk – you’ll learn something new about presentations and how the speaker engages their audience
- Make changes slowly. Don’t overwhelm yourself.
- Ask yourself, “would I like this slide?”
- When panicking, try to remember: YOU are the presentation. Not your PowerPoint. It’s just there to send a message home.
Final thought: don’t let bad things happen to your good data! It’s time to be heard. :-)
10:40 next speaker
Ten Tips for Building a “Big Enterprise” Data Analytics Organization
Pradip Patel, Manager, Digital Marketing Analytics, FedEx Services
FedEx connects 99% of the world’s GDP. It’s all about connecting the world. 300,000 team members and 9M shipments daily. That’s a whole lotta packages…
#1 Perform a health check. Take stock of whether you’ve inherited a practice or developing a new one. Step back and do an inspection of where you stand. It’s important to have a balanced approach. It’s money well spent if you can hire an expert to help you.
#2 Know your sponsor. It’s why your team has been created. Knowing their expectations comes in time. FedEx set up a weekly meeting with their internal sponsor to understand what their needs were and how they could move forward. It’s up to the manager to provide the recognition for what your team is doing.
#3 Establish a center of excellence. There’s a hub with FedEx’s involvement with NASCAR and things like a business/functional unit is a “spoke” in the wheel. Sometimes there is resistance to letting go of control.
#4 Assign everyone a role. Just like NASCAR has a pit crew, every person should have a particular assignment of responsibility. The pit crew won’t be successful if no one knows cold what they’re supposed to be doing. As a manager, you have to balance the workload but yet share learning across the team.
#6 Create a clear vision. (Sorry… bio break. Too much coffee…) Show a roadmap – where you’ve been, where you are, and where you want to go.
#7 Pace yourself. You can’t “gun it” to win. The fastest car won’t necessarily get you a victory. Campaigns aren’t going to stop. You need to be flexible to work with them in parallel to gathering the data.
#8 Win as a team. It is the team that brings you a win and yes, some luck is involved. Recognition and celebrating as a team is really important.
Browser crashed… grrr.
Pradip compared building your digital analytics team to a pit crew. Good stuff.
11:05 next presentation
Ten Tips for Leveraging Big Data in the Enterprise
Thuy Kim, Senior Director, Enterprise Analytics, Experian
Pronounced “twee.” Sharing her hobbies, very visual. Loves Pictionary. Wow… long time since I played that one.
Started as a media buyer on the agency side. Then moved to Experian.
#1 Purpose. Why do you connect data? Only 6% of marketing decisions are made using company data per Google. What are the key business decisions that are not being addressed properly? What are your current data challenges?
Experian wasn’t 100% confident that they were making the best decisions. They embarked on a journey to connect all their data.
#2 Pieces of the puzzle. Search Bid Mgt., Tag Mgt. etc. Thuy sees all the pieces as a puzzle. (left-brain and right-brain!) It’s important to understand they wanted to integrate the data and how to do that.
#3 Position and connect all the pieces. (Pre-site, off-site, and on-site) All those channels have different tools.
#4 Prioritize how you’re going to connect the data pieces. Mapping out your plan will help. Partner with key business leaders. It helps connect what you’re doing and how you’re doing it.
#5 Partner with stakeholders. Provide an overview and share the added value and insights for their teams (marketing, customer loyalty, product teams, finance, technology)
#6 Professionals. Use different consultants. About half the room said their analytic teams were short-staffed. They use Web Analytics Demystified as one of their consultants. Different consultants come with different approaches to solving a problem. They offer creative solutions for surface limitations.
#7 Prove value in increments. ROI capabilities at a granular level. marketing attribution across display campaigns. Tagging efficiency allows for flexible web analytics data collection, thus better data.
#8 Plan to iterate. New tools come into the marketplace and new needs develop. Onboarding or subsetting “B” or upgrading “C” will happen. Nothing stays constant.
#9 Patience. (I am so bad at this…) Each piece of the puzzle is a major initiative. Web analytics tools are usually passed on and few of the original team members are still involved. But it’s important to revisit the original analytics tool. Cleaning up data is important. Be patient. It’s just one piece of the puzzle that you have to connect to other pieces. Yes, things may take longer than what you want but patience will help you plan better in the long run and keep the bigger picture in mind.
#10 Perseverance. What you’re building in terms of connected data is cool and the organization will reap the rewards. It just won’t happen overnight. Infuse that passion into your team. Be sure to celebrate the small wins… you will have highs and lows. Appreciation goes a long way.
5 things to do
- Gather your team
- Identify all the tools in your toolbox
- Map out your ecosytem
- Identify challenges and limitations
- Evaluate your implementations
There are so many years ahead of you, think about scaleability.
11:24 AM next presentation…
Ten Tips for Testing in the Enterprise
Jinzhou Huang, Manager, Web Analytics, The Home Depot
#1 Testing as diplomacy. It’s about understanding and using testing to solve OTHER people’s problems and make them winners. (This is exactly what I tell clients regarding copywriting. It’s always about them… not you.)
#2 Teach basic stats with games. Most people won’t understand you. www.testandtargeting.com/simulator. This data sounds like a lot of randomness to many people. Teach stats through games.
#3 Ideation – discover and leverage. It’s important not to jump in data, but understand the context. Understand what type of problem people want to solve. Read your chat transcripts to find what your customer is experiencing with your site.
#4 Prioritization; MINE! Money, less “I” (more data), Nearness to productions, level of Effort. Tie this data to the bottom line.
#5 Anticipate pitfalls.
#6 Testing is NOT optimization. When you do testing, it’s optimizing for the masses. You’re just trying to piss off the least amount of people. Give people a choice and go for the majority. Push test results by like groups. It matters who likes a green button vs. a blue button. Continuously refine the results.
#7 Work starts AFTER the test. Report, communicate, act. This is where all the hard work starts. After you scale, you have tons of knowledge. A lot of this data is re-usable for different projects. This is where you gain visibility for your program. Make a big splash about your tests. People will remember you and then the funding will come your way.
#8 Solve the resource issue. Reduce, recycle, and reuse. There are a lot of times you can get people to evaluate what you’ve already tested and measured. Find image assets that can be recycled.
#9 Consider tag management. Home Depot is at the early journey of tag management. It gives you the freedom to insert code at your discretion and offers a “get out of jail free” card from IT.
#10 Think freely, act responsibly. Testing is the most responsible way to taking risk and innovating.
5 things to do next week:
- Find out how testing can help other people
- Teach statistics in a fun way
- Add segment reporting
- Try one target test…
And that is that! Slide was moved off fast..
Adam Greco is next.
11:48 AM
Ten Tips to Revive Your Analytics Program
Adam Greco, Senior Partner, Web Analytics Demystified
Found that most data analytics in companies sucked.
Employee #165 at Omniture. Associated with 700+ SiteCatalyst projects. Co-founder of Beyond Web Analytics podcast. Senior Director of Web Analytics at Salesforce.com. Cool.
#1 How bad is it? Ask the following: how accurate do you feel web analytics data is? How often do you use the data? How much value do you derive from the web analytics program?
#2 Re-establish purpose of analtyics. It’s important to change the website to make it better (it’s not just about reporting).
#3 There is “no spoon.” (Love “Matrix” references..) What would you tell a CEO who says “the Internet is a fad,” why they should spend money on building a site. How would you prove that the website would get leads, for instance. A lot of things have changed since you may have inherited the website.
#4 Play requirements Jeopardy. Turn the website into the form of a question. Such as looking at the analytics and observing what is already being collected. This exercise is cathartic because it helps them refocus on what’s important to them.
#5 Less is more. Knock out the highest priority requirements. An implementation with a small # is useful, accurate data points is better than a large implementation that is unfocused or untrustworthy. He actually stopped collecting analytics at Salesforce.com because it was being done so badly… They re-focused and re-started.
#6 Focus on data quality. If you’re going to re-launch, make sure that you QA the dashboard. They looked for standard deviations. All your work is for nothing if people don’t believe the data.
#7 Put the pedal to the metal. This is helping the client do the most with the tools. Tools do matter. Most companies use only about 20% of the tools they already have. At Salesforce, they were trying to reach everyone but were seen as “spammers.” They started to score the leads according to the data received (segmentation).
#8 Re-organize your team. The analysts may not understand how the data is gathered and the tech people don’t understand how the data is being interpreted. Bring the two together. You need both sides to make it work.
#9 Keep your team happy. A team becomes miserable if they’re too stressed. They’re in a thermometer environment all the time. Give your team time to be creative and innovate. Salesforce for instance, implement lightboxes for all their forms and it did well. (Worked for one environment and was expanded for more.) Web analytics people leave their job on an average of once every two years. Invest in their training.
#10 Create ambassadors. There are other departments, regions, countries. Create ambassadors. You can’t be everywhere. They need to be checking the quality. Train people to be ambassadors and know as much as you do.
5 things to do next week:
- Survey your analytics constituents
- Document tangible ROI
- Check data quality of all data being collected
- Identify what questions you can answer today
- Identify your current business requirements
Make the list of ‘Jeopardy” questions… it will help greatly to prioritize your goals.
Lunchtime! (yes, my head is exploding…)
Lunch over
1:05 PM next presentation
Tim Munsell
#1 Remember the golden rule: do unto others as you would have them do until you. This is about putting yourself in their shoes. If you’re given the keys to your business unit or your entire organization, what do you want your analysts to tell you? It’s easy to get caught up in side projects. All those “points” of data are real people who have a purpose.
#2 Trust your data. Your rep is riding on it. If it’s not trustworthy, you’re not trustworthy.
#3 Test your implementation. What you measured years ago may not be so important now. Know what you can and cannot change. Look at your data daily.
#4 Know what you measure. It’s not a good idea to sit in a meeting and say “yes, we can get that for you” and then find out that no, you can’t.
#5 Ask silly questions. It may be intimidating to ask but many times, it’s just an assumption analysts may have. Ask what’s important. Don’t assume you know. Without that context, it doesn’t matter how wonderful the analysis is, it ends up being irrelevant.
#6 Be curious. If you want to be interesting, be interested. Be aware of what’s going on around you and how others are using the information you’re providing.
#7 Facilitate understanding. Sometimes you need better presentation skills, but it’s important to know what kind of information you’re communicating and to whom.
#8 Your team. Surround yourself with those who will be able to help you. Bounce ideas off of others. Allow yourself to be challenged.
#9 Make developers your friends. Communicate what the end point is and they’ll often be able to help you reach your goals. Often, they have better ideas how to accomplish them. Share information with them.
#10 Know what you do well and what you don’t. Be true to yourself. Tim isn’t great with details but have learned how to manage them. He’s much more visual. He’s surrounded himself with people who can help him with those tasks.
5 things to do next week
- Get to know your business owners
- Use your site as a cutomer
- Take a developer to lunch
Ack… gone again.. apologies. Some speakers are quick to move the slide off the screen.
Next!
1:23 PM
Ten Tips for Getting In Front of Your Mobile Analytics
Josh West and Tim Wilson, Partners, Web Analytics Demystified
Intros… from marketing (Tim) and web dev (Josh)
#1 Mobile is not a strategy. Creating an iPhone app is not a strategy. Ask: what are we trying to achieve and what mobile tactics will help us achieve. And then… how will we know we’re successful? To measure success, know what it looks like.
#2 Know your audience. Mobile is evolving so quickly, that it’s tough to keep up with who your audience is. (Goodbye “brick” phones! My fave was Mulder on The X Files using one. Back in the day. Yeah..)
#3 Take advantage of the medium. Checking out the Salesforce.com site on my iPhone. The phone number changes with landscape mode. (in the top of the screen)
#4 Have a roadmap – even if it changes. Change the way you look at things.
#5 Get organized. With mobile, decide what will work best for you. One or multiple data sets? Does the mobile site serve the same purpose as the desktop? Some companies decide to release a mobile app and/or develop a mobile-only site.
#6 Pick the right tools. There are traditional tools but also tools like Flurry. Specialized mobile analytics and optimization tools. Sometimes sessions are very short. Your visitors may do something quickly on your site and then they’re gone.
#7 Asking trumps inference. Behavioral… what users DO. You’re inferring a lot of things when you don’t ask your customers questions. With mobile, you can do this. Ask users why they are visiting. If you develop a mobile site, make sure you’re developing the questions that need to be asked. Purpose, purchase intent, satisfaction with the experience.
#8 Privacy considerations matter. Buzzkill! Yes, it matters. Users want to know their information is secure but also that their information won’t be misused. Don’t creep out your customers. If your user knew exactly what you were capturing, would they be pissed? (there will always be the paranoid fringe… just be sensitive and smart)
#9 Look for ways to connect your dots. Omnichannel is the next frontier. Find patterns across the channels. Start identifying what data points exist for say, a search for a new computer. Desktop, mobile, optimization, etc.
#10 Something trumps nothing. Don’t get paralyzed.
It took a long time to refresh this screen. Already into the next presentation…
1:45 PM
Ten Tips for Buying and Deploying a Tag Management Platform
Kyle Sheppard, Digital Metrics Consultant, Humana
This presentation should be interesting for a few vendors in the room… :-)
Human uses IBM Digital Analytics and ensighten
#1 Ask: do you need a tag management system? (TMS) This would be wanting to focus on implementation agility, whether you manage various vendor tags or have multiple teams trying to implement tags. Also, are you trying to improve site performance? Trying to migrate from one platform to another? If so, a TMS may be a good idea.
#2 Ask: Analytics platform’s tool? Weigh the pros and cons of an analytics platform tool vs. a 3rd party tool.
- Platform integration
- Tag depth and breadth
- Pricing (or lack thereof)
#3 Align technical aptitude. What philosophy is the TMS geared towards? What is your philosophy? Technical or more business focused?
#4 Assess TMS bench strength. Does the vendor have a lot of tag types already build? Not just templates, but bench strength within their organization. How quickly can TMS vendor add a new tag?
#5 Ready to blaze a new trail? Don’t underestimate the time needed to get up and running. There will be trial and error. Educate others in your organization about how the mindset and process will change. Is IT ready to lose that responsibility?
#6 Adopt a data exposure mindset. Instead of implementing a web analytics tag, but move into exposing data.
#7 Ask: what data to expose? What is the measurement strategy? How will the data be exposed? What are the use cases for the data? This is a constant struggle, but it’s something to drive for.
#8 Create data classification. use a common syntax and naming convention for all data elements. Position data in tags consistently across sites.
#9 Get smart about data layer. use building blocks like parsing the page’s URL, using query parameters, etc., to dynamically create tag content. Use REGEX and wildcards to tag broader areas of your site.
#10 Be mindful of tag timing. Understand when tags fire. When is exposed data available? Digital metrics platform requirements and dependencies. Different libraries need to execute consistently… it’s important to get a feel of when timing happens.
Next speaker…
2:05 PM
Ten Tips for Fully Leveraging the Voice of the Customer
Chauncy Cay Ford, eCommerce Manager, Dell
#1 Be your own customer. Use your own site and notice things. Is it easy? Confusing? Does it make sense? She found out 50% of their traffic is coming to the site for support.
#2 Use multi-type VoC (Voice of Customer) solutions. Direct mail, surveys, comment cards, useability studies, social listening… You’re getting someone’s general overall impression of the site (they get a random survey to fill). Drill down to a customer just wanting to leave a comment. Lots of data windows… too many to explain here.
#3 Frame your feedback. Perception > customer chosen category. Actual > analytics. Layer them for the full scene. Dell VoC Analytics (Tealeaf + OpinionLab)
#4 Keep it in perspective, genius. Push > survey. Pull [+] feedback.
#5 See it. It’s nice to layer data and have different data points, but you also need to visualize it. Understand how your filters work.
#6 Create a data ecosystem. Listening, analysis, monitoring… Text alerts go to action replay go to analytics go to output. (slide shows an action wheel graphic)
#7 Accessibility matters. Make it easy and fun! Your tool users shouldn’t have to learn new tricks.
#8 Meet the expectation. Some who fill out a survey, don’t expect a return. A customer comment card is almost like someone is calling you. They’re expecting a return.
#9 Know your resources. Everyone in the company cares about your customer. She gives out capes to their “data ambassadors.” I want a cape!
#10 Fix it! It may look okay today but if your customer is complaining, you need to fix the useability features.
Impressive presentation! She drilled down, even though a lot of it was over my head. Yow.
Next presentation
2:32 PM
Ten Tips to Practically Apply Social Media Analytics
Amy VanHaren, Senior Social Media Manager, Stonyfield (John Lovett presenting since Amy couldn’t be here)
#1 Set goals ahead of time. Stonyfield did a campaign and John emphasized the importance of setting goals. They wanted more engagement. They wanted more photos uploaded with the campaign. Secondary goals: true to their unique message. Set the goals ahead of time so you know what you’re aiming for.
#2 Prioritize what matters most. Stick to the goals of the campaign.
#3 Build off previous learnings. Know what to expect and build off what was done before. Not just what you’ve done but what your competitors are doing. One of Stonyfield’s competitors was exposed for using GMOs in their yogurts, Stonyfield emphasized their own pure brand.
#4 Think holistically. Think about your digital assets.
#5 Select a hashtag you can own & measure. The conference changed from #accelerate to #accelerate13 because there was so much noise with the original hash. #WakeUpStonyfield was unique.
#6 Monitor to inform the work you do. Every day of this campaign, they had a theme and creative things to do. Stonyfield measured what worked and what didn’t. If one received more attention, they used it later in the campaign.
#7 Measure the big and small. Because this campaign encouraged uploads of pics, there were details that started to come out, like where customers bought their products. It allowed Stonyfield to pick up on trends and helped them build a larger perspective.
#8 Keep everything in one place. Use tools like Basecamp.
#9 Tell a great story. Over 2008 unique users were involved in the campaign in June by tweeting 2x or more, which is +220% over March. 70K unique users created or clicked on Facebook content in May, which is a 14% over 2013 average.
#10 Repeat! Their next campaign is #fightpesticides and having fun with “Cow Fu.” (ha!)
Break time! And it may be time for me to head home and focus on some client work. This has been an amazing conference and has challenged me in many ways. Even though I’m not an analyst, knowing how companies are measuring their digital marketing is important. I hope to be an even more helpful asset to them in the future.
