Showing posts with label analytics. Show all posts
Showing posts with label analytics. Show all posts

Monday, 3 February 2014

Why a Facebook Like Has Very Little Effect

Facebook recently made an earnings announcement showing that they are making significant progress in generating revenue.  As reported here by Forbes, Facebook said fourth-quarter net earnings jumped more than eightfold from a year ago, to $523 million, and importantly mobile ad revenues constituted 53% of overall ad sales.  An impressive set of numbers.

I find it unusual now to find a brand without a Facebook page.

The discipline of social media marketing is still underdeveloped however, partly due to the fact that the success metrics are not clear and as obvious.  What makes a success?  Someone seeing a post, reading it, liking it, sharing it?  What ROI are advertisers looking for when they run a Facebook campaign?

In an interesting article just published entitled, "Brand Comparison Study: What's Really Happening on Facebook?", Thomas Baekdal looked into the engagement patterns of 8 brands over the past two years to try and unpick what trends if any could be found.

It really is a fascinating article, albeit someone inconclusive.  I'd recommend any digital marketing specialist to read it as it will probably challenge many assumptions.

One key takeaway is that the more brands and publishers use Facebook, the more people like each brand, the less chance there is of any given brand getting a share of visibility.  It becomes more difficult to get the attention of users the more users there are and the more brands there are.

The graphs below show views per unique dropping over two years.

Analysis by Thomas Baekdal

And - remarkably (well done Facebook) - page likes per unique reach are up during the same period.

Baekdal goes on to go into some detail and helps us to realise that a "like" is not enough.  What really matters in social media marketing is if people talk about your brand in a positive way, for example, recommending your products or talking about their experience with their friends.

He says, "You need to get people to talk about you, because just liking your posts on Facebook has very little effect."

Those "likes" are just the tip of the iceberg and actually don't mean that much. They might make you feel good and make your corporate ego shine.  Just like in the real world, what matters is not how many people know you or say they like you but how many say good things about you behind your back.

That's tricky to measure and it's why social media marketing measurement has some way to go.


Thursday, 30 January 2014

Measure Early and Measure Deep - Extend Your Conversion Funnels


It's a couple of month's old but today I came across an interesting infographic from Infectious Media which highlights the importance of "viewability" of an online advertisement.  

The headline: 54% of all digital advertising is not seen.

The key message is that optimising the placement of your ads will deliver a better result.  Much like focusing on the deliverability of emails will help optimise an email campaign.

In a conversion funnel, you need to address all of the main points.

1. How many people arrive at the entry point (search, navigation, subscriber list)
2.  How many people see the message (deliverability, viewability)
3. How many people click on the call to action (click through rate)

Except it's not that simple.  Ultimately what matters is if a customer pays.or transacts.  Either in this web session or a future one.  Either online, on the phone or in store.

A conversion funnel is not complete until the transaction of money in can be traced back to the spend of money to acquire the customer.

This is never easy to do because the customer may have seen many different messages along the way and might be transacting on the 3rd or 4th visit.  This is where attribution modelling comes in, a geeky sport in it's own right.

What you can focus on with some simplicity is putting tracking in as early and as deep into the funnel as you can.

Early: viewability is a good example

Deep:  Instead of measuring clicks from the ad, measure purchases as well.  Or sign ups.  Or downloads.  If you are taking the user offline to a phone call, measure phone calls.  Even better, measure phone calls answered.  Even better, measure phone calls answered where the call time is more than a time threshold. If you want to know how to do that, let me know.

Your conversion funnel is more accurate the earlier and deeper you measure.


Wednesday, 29 January 2014

The Telephone - It's Still Good To Talk

British Telecom (BT) famously told us, "it's good to talk".

There are some things that are just better if done over the phone. 

Transactions and information transmission suit online and text, synchronous conversations work well on the phone or face to face.  Asynchronous conversations suit chat or text.

So far this year, I found the phone (voice) to be my channel of choice when; enquiring about holiday options, discussing sensitive work topics, ordering a taxi, speaking with my family arranging for a decorator to work on our house.

There are many high value transactions that are initiated over the phone.  High value, offline leads.  Buying a car, buying a house, booking a holiday, finding a tradesman, purchasing financial products.  Conversations convert.

Apart from talking, my phone has been my preferred device for the following activities; emailing friends, reading updates from family on their travels (Facebook / Blog), ordering groceries, renting DVDs, making bank payments, checking the weather, reading the news, tuning my guitar, finding out how to get somewhere, selling second hand goods, buying movie tickets, reading a book on the train, following content of interest (Twitter, Quora), taking photos, checking train schedules and using a calculator.

I may be stating the obvious but we are all becoming heavier and heavier users of phones.  We all have them, 62% of UK mobile phones are now "smartphones" and we generally have them with us 24/7.  17.4% of global web traffic comes through mobile devices.  49% of UK adults access the Internet on their mobile and the average household now owns more than three types of Internet enabled device, with one in five owning six or more.

Call volumes have been declining slightly but there are still over 233 billion minutes a year made (2012 data: OFCOM).

If my personal experience is anything to go by, I have moved transactions online where I find it easier to do so, however I will always use the phone for more complicated transactions or where a conversation is involved.

It's not surprising that investors have been piling money into mobile.  Latest figures show that in the US, in Q3 2013, venture capital investors poured $1.12 billion across 150 deals into U.S. mobile & telecoms companies, marking the largest financing quarter to the mobile sector ever.

As we look to the future, here's 12 2014 predictions  for mobile by app development company Goldengekko.

My 13th predication is that measuring all of the phone activity (on voice and in apps) will become increasingly used and understood by businesses  as they calculate the return on investment they get on their marketing where the response (call to action is a phone call).

In the same way that web analytics is now commonplace and understood by almost all businesses with a web channel, phone analytics will become known and used by all businesses that rely on the voice calls for their inbound leads.

It's very good to talk.