Skip to content

Tag: Data Analysis

Community Management Tidbits – Analytics

Community Management in the Context of Analytics

So, the truth is, analytics are a term that scares a lot of people. But don’t panic.

Let’s back up for just a second, all right?

You’ve got a community. And you’re working hard on it. It’s growing. But you have no idea whether what you’re doing is having any sort of an impact whatsoever. This is where analytics comes in.

Now, don’t panic if you don’t have a data analysis background. It’s not strictly necessary. What you do need, though, are (a) a means of measurement (preferably you should have a few of these) and (b) the willingness to measure. Really, it’s that easy. You do not need a degree in Advanced Statistics.

Google Analytics 4 (replacing Universal Analytics)

First of all, the primary measurement stick you want is Google Analytics. And it is free and very easy to use. It’s also a rather robust measurement system, capable of showing trends in Visitors, Absolute Unique Visitors, and more.

In addition, it shows, among other things, where your traffic is coming from, where your users land, and where they departed your site from. It also shows Bounce Rate, which is defined by Measurement Guru Avinash Kaushik as, “I came, I saw, I puked.” In other words, the visitor only visited one page of the site.

Keep in mind, though, that it’s entirely possible your visitor loved your site but got everything they needed in just one page. So, while they may have bounced right out of there, it might not have been due to any fault or failing on your part.

So, try not to take it personally, okay?

AHRefs

Thank God for AHRefs. While free website measurement tools have come and gone (apart from Google Analytics), AHRefs will review whatever is out there.

So, one thing to keep in mind is that as this post is updated, I keep finding new yardsticks. And then they go away after a while. At least AHRefs is still hanging in there. Whew.

Also, consider SEMRush and Ubersuggest.

Analytics From More Yardsticks

Furthermore, there are also measuring websites specifically designed to help you comprehend how you’re doing on Twitter and elsewhere, namely:

  • HootSuite – count the number of clicks you receive on shortened URLs, to supplement your Google Analytics click counts
  • HubSpot – measure how influential you are (with a hugely helpful diagnostic) and
  • Tweepsmap – analytics and info on who unfollowed or followed you on Twitter.

Facebook also has its own metrics, which you can see if you have a page.

Using Your Findings

So what do you do with all of this information once you’ve amassed it? Why, you act upon it! Does one page on your site have a far higher Bounce Rate than the others? Check it and see if the links on it are all leading users away from your site. If that’s not the culprit, perhaps its content just plain isn’t compelling enough.

Got a series of links you’ve tweeted that have consistently gotten you the most clicks? Then check to see what they all have in common, and offer similar links in the future. And maybe even build some onsite content around those subjects.

Has your HubSpot grade tanked in the past week? That might be due to external factors beyond your control, but check to see if any of it is within your purview. Perhaps your server was down.

Finally, small fluctuations over short time periods are perfectly normal and are no cause for concern. However, much larger hikes and drops, or trends over longer time periods, are more of an issue. But you’ll never know about any of these things unless you start to take measurements, and read and use them.

4 Comments

The Numerati by Stephen Baker, a Book Review

Time to Look at The Numerati by Stephen Baker

Who are the Numerati?

The Numerati by Stephen Baker remains a fascinating work about sensors, technology, data mining and where it’s all going when it comes to our privacy.

And it ends up all about data, about collecting, refining and interpreting it. People are, well, a bunch of fish in a bowl. Or, if you prefer, hamsters on a wheel. We are lab rats, we are subjects, we are collections of bit streams. We are … information.

And the kicker is that, put together enough things about us, and conclusions can suddenly be drawn.

Conclusions from The Numerati

Let’s say I go to the same grocery store every week (not a stretch – I really do). And I buy fish every single week. What if I buy, say, tuna steaks 70% of the time, and swordfish the other 30%? Am I automatically a tuna lover? Or am I simply scared to try something new?

Or am I getting to the fishmonger when everything else is sold out?

And what happens if a coupon is introduced into the mix? Does my tuna consumption go up to 80% if you give me $1 off per pound? However, that’s not too much of a victory, seeing as I normally buy it anyway. Will a $1 off coupon entice me to buy the more pricey salmon instead?

Ideas But Not Gospel

The data gives its interpreters (Baker refers to them as the Numerati, which sounds a tad like Illuminati and perhaps he means that) ideas. However, it’s not really a slam-dunk. Or, at least, not yet. Hence essentially the Numerati bucket you.

So I am a tuna buyer. And I am a sometime swordfish buyer. And I am also a Caucasian woman, in her (ahem) sixties, married, no children, living in Boston.

So far, so good. And when we herd all the data together, when we aggregate the bits and bytes of our lives, this may very well have a lot to say about us. Because it might be a predictor of how I’ll vote in the next election.

Or perhaps it will show how I’d use a dating site if I should ever need one in the future. Or it may even tell whether I’m likely to become a terrorist.

Border Collies and Data Goats

The data matters, but, to my mind (and to Baker’s as well, it seems), there are not only herds of data but there are also nagging outliers. And these constitute the Border Collies amidst all the data goats.

Perhaps I am buying tuna to feed to a cat. Or maybe I buy it with the intention of eating it to improve my health but, alas, never get to it and it goes to waste every single week.

So consider this case: a health professional places a sensor into a senior citizen’s bed, to determine whether that person is getting up in the morning. And, let’s say we also collect weight data.

Because a sudden dramatic rise in weight would indicate the possible onset of congestive heart failure. And let’s say the senior in question is a woman who weighs 150 pounds. Your own mother, maybe.

Day one: 150 pounds. Day two: 158 pounds. And then day three: 346 pounds. Day four: 410 pounds. Golly, is Mom really that sick?

Maybe Mom’s dog is 8 pounds. Okay, that explains day two. But what about days three and four? Maybe Mom’s got a boyfriend.

Or maybe she’s got two.

Messy Feelings

When I had the occasion to meet Stephen Baker at the 2010 Enzee Conference, we had the opportunity to talk a bit about these squishy, messy feelings.

Sure, our hearts are in the right place. And we want Mom to be safe and healthy, and we can’t be there. She might live in a warmer climate, and we cannot (or won’t) leave our cooler climes.

Or the job opportunities may be no good there for us. For whatever reason, we are here and she is there. So we want to be aware, and caring and all, but in our desire to gather information and protect her, what else are we learning?

If Mom is competent, and single, and protecting herself from STDs, we truly have no business knowing who she spends her evening hours with. Yet this technology makes this possible.

And if we have any sense of the future at all, we have to think to ourselves: what happens when I become Mom’s age? Will my bedroom and toileting habits potentially become a part of this huge bit/byte hamster wheel lab rat canary in a coal mine data stream? You betcha.

Are the Numerati Worrisome?

So, it is often said that only people who have something to worry about in their private lives are the ones who have worries. Everyone else should be fine, blithely giving up their warts and preferences, their virtues and secrets, to all who ask.

I say bull. I like my secrets. And I like my hidden life. And I’ll be damned if I give it up, even in the name of health, diet, voting, national security or even love.

A terrific read. I highly recommend this book.

Rating

Review: 5/5 stars.

Leave a Comment

Avinash Kaushik’s Web Analytics 2.0, a Book Review

Avinash Kaushik’s Web Analytics 2.0 – Yeah, I’m a Fan

Web Analytics Matter!

We Go Way Back

First of all, the first book that truly caught my eye and made a huge impression on me was Avinash Kaushik’s Web Analytics 2.0: The Art of Online Accountability and Science of Customer Centricity.

As a (hopefully) former data person, I can relate to the idea of needing web analytics. E. g., the measurements of how your website does. Why do you want to measure with web analytics? Why, you need to see whether your message is actually going anywhere.

For e-commerce sites, the ultimate test is, naturally, whether you’re getting sales. But it’s hard to tell – particularly in a complex organization – whether the website drives sales or offline marketing efforts.

And even measuring orders via these channels may not tell the entire story, as customers may see offline advertising and then come online to buy, or they may do the reverse and buy in-store after researching a product online. Or they could just be coming online to think about it and compare and mull it over and could convert to a paying customer days or weeks or months later. Or never.

What if You’re Not in e-Commerce?

And what about sites (such as my own) where nothing is offered for sale? My ultimate customer becomes, of course, someone to hire me, either permanently or temporarily. And this would mean as a consultant or a partner or a founder or a director or whatever, but that might be months away. What happens in the meantime?

I might be able to dope some of that out with SEO and seeing where I am in search engine rankings, but just because people can find my site doesn’t mean they’re going to convert into hiring me or are even in a position to do so. My mother (hi, Mom!) can find my site and read it, but she won’t hire me any time soon. Unless I want to come and clean the gutters or something.

How do you or I know what’s happening?

Enter analytics.

It is, admittedly, still an imperfect science. But Mr. Kaushik breaks it down and describes the reports that you need to understand what’s happening with your site. He talks about what is essentially a Trinity strategy: experience, behavior and outcomes.

User Experience

It’s not enough to just track sales (outcomes). It’s also about user experience and behavior. This is much like in the offline world, if you think about it. Going to a restaurant is an experience and many of them are packaged as such. But it is a far different experience going to a McDonald’s or a Chik-Fil-A versus a Bertucci’s.
And that experience differs from going to Legal Seafood’s which in turn is different from Blue Ginger (celebrity chef Ming Tsai’s restaurant). You can intake the same amount of calories. You might even be able to get in the same quality and types of nutrition. And you might enjoy a Big Mac as much as you enjoy one of Chef Tsai’s specialties. Aside from price, what are the differences?

Web Analytics for What Sort of User Experience?

When you go to a McDonald’s, a part of the price is wrapped up in the experience. For chain entities in particular, it’s about sameness and predictability. If you find yourself in rural Oshkosh and have never been there before, you see the golden arches and you realize what to expect.

For Bertucci’s, even though it costs more and there’s table service, there’s a similar vibe. You go there because you can depend upon it to be a certain way.

And Blue Ginger is also dependable in the sense that it’s very upscale so you know you are going to be treated a certain way and it will look a particular way and presumably the food will taste in a way that reflects that kind of investment, both by you and by Mr. Tsai and his team.

Enhanced User Experience

Mr. Kaushik shows how understanding analytics can help you to enhance user experience.  And this, ultimately, drives user behavior. While conversions (sales) are the ultimate in user behaviors, he doesn’t forget about other valid behaviors.

Hence for the e-commerce site, product research is a valid and valuable behavior. So is printing a map to a brick and mortar store. Or comparing prices.

And for a non-e-commerce venture (again, I’ll use myself as an example), valid user (reader) behaviors are things like reading my writings and getting to know me. I put myself out there in order to be known, because that’s a piece of the hiring puzzle (why are there interviews — it’s not to know about skills, which should already be known.

It’s to see if there’s a personality and a culture fit). Plus it enhances networking. Know me, think I’m worthwhile (at least, I hope you do) and you might think of a place where they might need me, or someone I should meet. And I do the same, in turn, for you. And cosmic karma gets us both into better places.

Back to the Book and More Web Analytics

But I digress. Let’s get back to the book.

The book has a lively, engaging style. It’s long but I sailed through it. And Mr. Kaushik (who is very gracious and seems to be very approachable, by the way) is clearly having fun and loves what he does. It’s a refreshing joy to read a book where the author is constantly delighted.

Read his book. Learn about analytics. Make the web a better place.

May your bounce rate be low, and your conversion rate high!

Rating

5/5

1 Comment

Quinnipiac Assignment 11 – ICM 527 – Continuing Program Evaluation

A Look at Quinnipiac Assignment 11 – ICM 527 – Continuing Program Evaluation

This week, we continued studying the evaluation of public relations campaigns as a continuing program evaluation.

Ethical Issues Regarding Evaluation

As is true for any presentation of numbers, there are ways to spin findings which can lead a reader to believe one thing or another. You can use numbers to make a case. And some numbers, if suppressed or deemphasized or just plain omitted, could alter organizational decision-making. This only gets into telling the truth with numbers.

All bets are off if a strategic planner or any sort of analyst out and out alters the figures they have to present,. Or if they didn’t get accurate or truthful numbers to begin with.

Cans Get You Cooking

But even if the analyst is completely honest about results and figures, there are still issues with emphasis and language. For the Cans Get You Cooking campaign, the initial purpose had to have been to increase the sale of canned goods. Instead, they labeled the campaign as a success for leading to an increase in awareness of canned foods.

Awareness is a perfectly legitimate (and objective) goal for a campaign. But they seem to have swept the goal of increased sales under the rug. This was in favor of the one, demonstrable, favorable outcome – a boost in awareness.

On page 125, Place notes

“The role of ethics in public relations evaluation was described by participants as inherently associated with truth and fairness. For some professionals, this meant conveying evaluation data accurately and truthfully to organizational leadership or clients. For other professionals, this meant measuring whether the most accurate story or brand image reached an organization’s publics.”

Upshot

Professionals, fortunately, realize that others can misinterpret their words, even if they are reporting accurately on the numbers. If a campaign increases, say, signups for a class by five over an initial figure of five, then how do they report that?

Is it a report of a new five signups, or does the professional state that signups have doubled? Both are mathematically correct, but there is an exciting spin to the latter which may be making it look more significant than it truly is.

The Real Warriors and Okay 2 Talk Campaigns

A review of both campaigns revealed good attention to detail. Both campaigns seemed to be rather carefully planned.

The Real Warriors Campaign was designed to encourage active armed services personnel and veterans of recent American military campaigns (since 9/11) to seek psychological counseling and other help for post-traumatic stress disorder, e. g. ‘invisible wounds’. Primary research included focus groups and key informant interviews. All of the campaign’s goals were awareness-based.

The goal was to decrease stigma felt by veterans seeking mental health assistance.

Measurements

The measurement of the effectiveness of the campaign included the distribution of campaign materials, website visitors, and social media interactions, plus news stories. This is good for an awareness campaign, but where are the actions? Where are the increased numbers of veterans seeking help?

A far more germane measurement would be to show an increase in personnel hours for armed forces mental health professionals.

Or perhaps there could be a measurement of the hiring of more counselors, or agreements with more civilian counselors. Without naming names or otherwise violating privacy, the number of patients in treatment is easy to tally. So can the number of appointments made, even if some of the appointments were never kept. Another objective measurement of success would be a decrease in suicides and fewer calls by veterans to suicide prevention hotlines. The campaign shows none of that.

OK 2 Talk

As for the OK 2 Talk Campaign, that campaign’s goals were to create awareness and also to launch a safe social media space. Tumblr was their chosen platform as it allowed for anonymity. It seems to have also been chosen for a demographic match although that is not spelled out.

Metrics

The measurement of the effectiveness of that campaign was a lot more closely aligned with its initial goals than the Real Warriors report showed. For example, the OK 2 Talk report gave objective figures regarding engagement on OK2Talk.org. The page views are not necessarily indicative of much. It is the content submissions which seem to better reflect engagement.

On the Tumblr blog, they encourage visitors to anonymously post about how they are feeling. The blog makes it clear that they will not post everyone’s writing.

However, there are several well-written or illustrated posts showcasing various viewpoints. OK 2 Talk intelligently shows all kinds of posts. This is even those where the writers clearly need help or are just reblogging messages put together by creative professionals.

The Continuing Program Evaluation Campaign

The campaign report shows the number of content submissions and the number of clickthroughs to a ‘get help’ screen. There is also a statement regarding ‘thousands’ of comments but no specifics. They could have shown this more clearly. But that does not truly matter.

Showing the number of clickthroughs to the ‘get help’ screen was an objective and direct measurement of how the campaign is going. It answers the question, ‘did it work, or was it just a colorful and fancy waste of time?’ with ‘yes, it did’, and far more effectively than the distribution of materials ever could.

Smith Says…

As Smith notes on page 335

“Guesses aren’t good enough; Hard work and cost aren’t measures of effectiveness; Creativity isn’t, either; Dissemination doesn’t equal communication; Knowledge doesn’t always lead to acceptance; and Behavior is the ultimate measure.”

In particular, Real Warriors should have remembered that dissemination does not equal communication. After all, the distributed campaign materials could have gone right into the trash. Yes, the campaign’s stated goal was awareness. But the campaign can only really measure it with some form of observable action. Without some demonstrated actions, Real Warriors seems more like a lot of paper redistribution.

The two campaigns have similar goals, and both have the valiant ideal of helping the mentally ill. But it’s only OK 2 Talk which is showing objective and relevant results.

Relating it all back to the ILSC

For the Institute for Life Sciences Collaboration, deciding what to measure, and to make sure it is being accurately measured, are important steps to take. It is pretty easy to count website visitors using Google Analytics or the like. But a better measurement is actual engagement like blog comments, Facebook comments and shares, and LinkedIn comments. This will tie directly to awareness objectives.

For objectives on adding high schools to the Small World Initiative, good measurements include the number of times that educators click through to a ‘get information’ page. The ILSC should add one to a revamped website. They can also expect such inquiries in the comments and messaging sections of a possible future ILSC Facebook group.

A similar vehicle for obtaining such inquiries could be a possible future LinkedIn group for the ILSC, and its topics.

Measurements of the campaign reaching donors could be a look at the number of visits to a donations page. It would also be the percentages of site visitors who went all the way through the online donations funnel. Knowing where they stop (if a visit does not lead to a donation) would be extremely helpful information to have.

More About the Continuing Program Evaluation

For the website, Google Analytics should be used to tie back to visitor acquisition. If Facebook turns out to be the most popular place for visitors to come from, then the ILSC should concentrate there. A surprisingly small amount of money (e. g. $20.00 or so) can boost a post and reach even more people.

This measurement is useful for all types of objectives, as it helps to define where to best concentrate the ILSC’s social media time. There is little use in devoting substantial time to LinkedIn if the publics don’t come to the website and don’t donate any funds.

Awareness needs to be related to action, for it is action that will get the SWI out of its funding gap and help keep the ILSC going for years to come.

Leave a Comment

Quinnipiac Assignment 10 – ICM 527 – Program Evaluation

Quinnipiac Assignment 10 – ICM 527 – Program Evaluation

Let’s do a program evaluation. This week’s readings were about evaluating a strategic plan and program.

Key Concepts

As Smith said, on (Page 331), “Program evaluation is the systematic measurement of the outcomes of a project, program or campaign based on the extent to which stated objectives are achieved.”

With a plan in place and measurable, clear objectives included in it, the next question is whether anything is working. This comes from figuring out how to measure results and what’s ‘good’ or at least adequate.

In Module 8, we studied Cans Get You Cooking, where the idea was to increase awareness of cans’ use in cooking via cooking shows and blogs. However, another objective was increased sales (after all, why bother with such a campaign if sales don’t increase?), and in that respect the plan was unsuccessful. According to Companies and Markets, the purchase of canned goods declines because of improvements in the economy.

When consumers have more discretionary income to spend on foodstuffs, they purchase fewer canned goods – no matter how well-crafted a campaign is. There was increased awareness, yes, and under that criterion, the campaign worked. But under the criterion of increased sales, it did not.

The Case of the Traveling Goalposts

It seemed a little as if the goalposts were moved in that campaign, that increased sales became a less attainable goal. Awareness was a far more readily attainable goal, and so they presented awareness as the premise behind the campaign.

These moved goalposts are the difference between what Smith refers to as awareness and action objectives, on pages 332 – 335, with the third type of objective, acceptance, straddling a line between both of the others. For the Cans Get You Cooking campaign, it seems as if the attainment of the awareness objective was the only cause for celebration.

Smith makes a compelling case on page 334, that creativity, effort, and cost don’t count as measures of effectiveness. All of those facets of a campaign are on the side of the organization. But measures of awareness, acceptance, and action are all effects felt (and acted upon) by publics. By definition, creativity, etc. should not have anything to do with the effectiveness of a campaign.

The Eight-Step AMEC Social Media Measurement Process

Jeffrey (Page 4) outlines, “The Eight-Step Social Media Measurement Process

  1. Identify organizational and departmental goals.
  2. Research stakeholders for each and prioritize.
  3. Set specific objectives for each prioritized stakeholder group.
  4. Set social media Key Performance Indicators (KPIs) against each stakeholder objective.
  5. Choose tools and benchmark (using the AMEC Matrix).
    • Public Relations Activity
    • Intermediary Effects
    • Target Audience Effects
  6. Analyze the results and compare to costs.
  7. Present to management.
  8. Measure continuously and improve performance.”
Quinnipiac Assignment 10 - ICM 527 - Program Evaluation
Avinash Kaushik, author of Web Analytics 2.0

What’s it Like?

This process compares favorably to methodologies learned in ICM 524 – Social Media Analytics. In that class, we read Web Analytics 2.0 by Avinash Kaushik. On pages 29 – 32, Kaushik outlined his Step 3 – Identifying Your Web Analytics Soul Mate (How to Run an Effective Tool Pilot) (average time: 2 years) Evaluate the following –

  • Usability
  • Functionality
  • Technical
  • Response
  • Total cost of ownership

Also –

  • Get enough time
  • Be fair
  • Ask about data sampling
  • Segment like crazy
  • Ask about search analytics
  • Test site content grouping
  • Bring on the interns (or the VPs!)
  • Test support quality
  • Reconcile the numbers (they won’t add up, but it’s fun!)
  • Check the daily/normal stuff
  • Sweat the TCO (total cost of ownership)

What Kaushik said, and what Jeffrey said, are similar. Because measurement is an objective activity. This is why objectives need to be clear and measurable. Five percent is measurable; better exposure (in general) is not.

For both authors, the idea is to have specific objectives and then act on them, whether those objectives are to launch a strategic campaign or select a web analytics vendor. Then, once they choose the vendor, get the yardstick in place, and use it.

Kaushik further reminds us that, while our intention may be to select a vendor and essentially ‘marry’ it, we still need to be evaluating the evaluator. If it’s not performing up to our reasonable specifications, then it’s time for vendor divorce court.

Key Performance Indicators (KPIs)

On page 7 of Jeffrey, it says, “Shel Holtz, principal of Holtz Communication + Technology (www.holtz.com) defined a KPI as a ‘quantifiable measurement, agreed to beforehand, that reflect the critical success factors of one’s effort.’”

This puts KPIs on a par with what we have been calling objectives. Wanting to ‘get better’ is one thing. But it’s vague and subject to weaseling.

Wanting to improve recognition of the Institute for Life Sciences Collaboration  (ILSC) and its missions by 5% as is measured by surveys taken during the second quarter of 2016 is a measurable key performance indicator. Therefore, anyone who can read numbers will be able to determine whether the KPI has been met.

Program Evaluation Re: Applicability to the ILSC

Beyond just recognition measurements, there are any numbers of KPIs to measured. These include the number of schools served by the Small World Initiative by a certain date, or increasing donations by a particular amount, subject to a clear deadline.

The ILSC

Currently, the ILSC website in particular seems to be just something people threw together. But it’s without any sense of how to deal with technological and design changes, or scalability. Keeping measurements out of the mix means the ILSC website can be started and then forgotten about. And it seems a lot like that’s exactly what happened.

However, a website cannot be a flash in the pan, as that can cause the publics to feel the organization behind it is also fly by night. Particularly when asking for money, an organization needs to give forth the impression of trustworthiness and solidity.

Adding Key Performance Indicators and measurements means there needs to be a sea change in how the ILSC views the website. It isn’t just something you throw together in an afternoon, for some temp hired for a few weeks to handle and then never see again. Instead, it needs to be an integral part of the organization.

Program Evaluation: Takeaways

While the organization’s work is (generally) offline, there still needs to be room for the website in the minds of the organization’s board members. One facet of their thinking has to include how to best use the website and social media, to better communication the ILSC’s mission and goals, and to communicate with its publics.

The website has got to have a place in those conversations, and it currently does not. That has to change.

Leave a Comment