Using Google Analytics to measure outcomes against your PR objectives

Is this relevant to you:

·       Are you interested in knowing more about what people do when they see your proactive PR outreach?

·       Do you have a specific goal in mind when you plan an outreach campaign?

·       Does that goal involve some sort of interaction on your web properties?

If the answer is ‘yes’ to even one of these questions you probably use or have someone using Google Analytics to track your users. Alternatively you might be using Adobe Analytics to track responses, which is good as it is more challenging to use than Google Analytics (GA), and so you probably a very competent PR web data analyst.

Regrettably that does not mean GA is easy. At first glance it yields some results but feels like there lots of duplication and unnecessary complication. Please persevere. If you are the right organisation asking the right questions it will help you, a lot.

I am not going to tell you how to open an account, configure settings and all that kind of thing. Get over to GA tutorial area (https://analyticsacademy.withgoogle.com/course01). Though to go from a standing start to competent will involve quite a few hours.

My intention is to offer some ideas of what GA can do to help a PR person. In particular around objective setting, and goal and conversion measurement.

A PR manager might be wondering if a single person can handle both Twitter and Facebook. And if one gets better results, should it be allocated more resources. To do that you will need data, likely from GA.

Let’s presume there is a PR plan, campaign objectives and a selection of measurable goals. These might be White Paper downloads, contact form leads, additions to a mailing list at the micro end or e-commerce sales at the macro level.

GA will allow you to set up each of these as goals and track the results. GA is quite keen on the concept of ‘valuing’ intangible actions on your site. For some this might be where things get a bit controversial. Now I am no fan of AVE’s, and the like. Setting a value on something like a White Paper download could be like trying to value of Twitter follower. And that was an argument which did not end well!

It would not be correct to value these interactions in the first instance. After a number of campaigns you might be able to make some fairly objective assumptions. Over time you may able to estimate how many of a certain type of interaction results for particular organisations in a sale if there is a tangible linkage.

For example, if you are a B2B provider you will probably have an idea that for every 20 service demonstrations you get a sale. You will know the average sale value and so can estimate the value of a demonstration, hence the value of that type of goal. But it does not work for everything so you must discriminate.

There are a selection of other facets you can check alongside you goal conversions including relative new to returning visitors, if they came by search what term did they search under and what it the general level of satisfaction with your site (bounce rate). There are a selection of attribution models you might consider to understand the progression successful buyers take and where others drop out.

My advice is to become familiar with Google Analytics. PR can benefit from access to web analytics if success is getting people onto a web property.  You really need to hold the keys to this stuff and not rely on other departments.

Thank you for viewing the article and I very much hope you found it interesting. Please don't hesitate to offer a comment, particularly if I get things wrong!  It would also be great if you wanted to subscribe to future blog updates.

Does Google Analytics offer PR a way of connecting their efforts with organisational outcomes?

- Put simply, yes(ish). 

Though it comes with lots of caveats, this is a possible way of tracking cross social media impact. Google Analytics (GA), as has been said before, is not easy. The interface is often confusing, often inconsistent and often failing the 'so what?' test.

But if your goal is for people to do certain things on your site, it might just prove to be the assist that puts the ball in the back of the net - not that I do football analogies.

Before I go any further I have a quick rant about GA. 

I do a lot of competitor comparison/analysis of media coverage. I hoped GA could help me with this, but it can't. While Google Search is the master at finding influential news topics and comment, GA does not integrate search results, unless (..and how great this would be) you can correct me.

Back on-topic; and the starting point is nowhere near Google Analytics. You need to think about what success might look like. Think carefully about the goals to achieve. If you run an e-commerce site and have a new product, its likely it involve sales. If you are B2B you might want sign up on your site, white paper downloads or contact forms. 

To appreciate the many processes involved, and you are a newbie to GA you would be best reviewing these tutorials: 

Firstly there is a series on GA Fundamentals. Then, if you run e-commerce (a shopping cart, but not etsy!) on your site you will need to enable that element and also review these additional tutorials.

I know this all seem ponderous, but there are concepts, terminology and bolt-on you need to know about if you are going to manage this this well. I promise you, if you really want to find the answers, this will not be a waste of your time.

However this is a gist of what's involved. Get a Google Analytics account or update an existing account (it changed quite a lot about 2 years ago, latest version known as 'Universal Analytics'). Prepare a list of your target social media networks. What you will probably want to know is which of these networks gets the most sales or conversions. 

Once you have prepared your post or tweet for sending out, you will want to add a link to the best landing page on your site. This is the magic bit...GA can add a bit of extra code to track lots of extra things to work out what works best.     

So you need to add a bit to your inbound link address and that's done in GA by using the Link Builder (put Link Builder into the Analytics Help bar). You will need to add 3 pieces of information, the Source or referrer, the Medium (banner, email, etc), and Campaign name. If that's not enough to isolate which of your efforts it is there are 2 other optional field you can use.

The link builder will create a unique link which you can then use as your link for that network and campaign, etc. Once installed the code will start returning activites to your GA account telling you things like what network and promotion was most successful in getting people to your site, making conversions and hitting your preset goals. It will also give you other useful information like what was the bounce rate (only visited 1 page), what type of device and browser was used. 

It will also tell you about what people do once on your site, like where did they go, for how long and if they were new or returning visitors. If its an e-commerce site, as long as you have enabled the right GA settings you can also see how people progress through the buying process and if there is an aspect making buyers drop out, or loop-back in the process. 

When you connect the initial tagging process with the other trackable things like product groups and categories, it is easy to image how easy it is to link the initial outreach with an end result.

This outcome connection is something PR has been striving to achieve. 

GA as a tool, has the feeling of being an instrument for the use of advertisers. It will quite easily integrate Adwords data, letting advertisers see how well different ads work. 

The worry I have is that too many PR's are discounting it for this very reason. Any sensible organisation will be using GA, or the even more confusing Abobe analytics package. 

PR's should fight to get and keep their GA logins; and study very carefully what they can do with it, as it has many PR applications. If I might be so bold, I might suggest it is one of the few ways you might genuinely connect PR input with organisational outcome.   

I hope you found this of interest and if you have a view please post a comment. It would also be great if you wanted to subscribe to future blog updates.     

  

PR vs Marketing - Let Battle Commence

I bumped into a tweet last week which mentioned the difference (or lack of) between PR and marketing. If you study the books and consult with the relevant trade bodies you will be very clear on the differences. But what of reality in this fast changing media environment?

Marketing is about selling stuff (er...simplified) encompassing sales and associated activities. PR on the other hand is more about communication with not only customer, but influencer and other stakeholders . It's about engaging with and convincing those various publics.

My angle on this is measurement. As the media environment has moved online so the arguments become louder over what are the best metrics. There now appears to be a lot more ways to show change. We used to talk about media output as the measure of PR success. Now we are told real insight comes from understanding outcomes. One of the outcomes which is most often touted is impact on sales. This is where I think the problems arise. Where marketing and PR are increasingly stepping on each others toes.

Both disciplines are starting to use very similar metrics. Sure, the online environment rather points you towards using a fairly prescriptive selection however PR is just a little bit more than just sales.

Auto Sentiment Analysis Infographic

I have created this infographic to help summarise my experiences with auto sentiment analysis programs. I have used a few over the years and while they are getting better, they are far from perfect, and indeed are sometimes described as being no better than tossing a coin!

One peculiarity that I have noticed is that the sentiment engines which have undergone some form of sector-specific 'training' often seem to be more accurate.

For example I was checking some social media coverage on cigarettes and it was notable how poor the sentiment scoring was. Often the same tweet got re-tweeted with a different sentiment score. Then a week or so later I was looking at coverage on mobile phones and it was amazing how much more accurate the sentiment scoring was. Whereas in the first example I has changing possibly every other item, with the mobile phone sector it was more like one in every four items.

Sitting in on the infuence debate

In the world of online measurement it is easy to think of it as similar to traditional media (well similar) but with the additional necessity to measure influence. In traditional media influence is not such a major factor as the medium is finite in volume, with the individual media titles having their own discernible character against which influence can ascribed.

In social media this walled-garden has been thrown open to all comers. The result is a melting pot of unlimited dimensions. In a weeks time I am presenting on media research at the CIPR and central to the presentation is some pointers on the measurement of social media and crucially some discussion on the measurement of influence.

To quote Shel Israel it is not easy to measure. He makes the point that it is easy to mistake popularity for influence, largely because popularity of easier to measure. You can consider the number of comments and the level of engagement as an proxy for popularity. The challenge to PR in the future is find who has influence in the space they are working in. The process has to start with a concise understanding of which areas of the social media space are relevant to the market place.

A key PR skill is communication and needs to be combined with the ability to research the media market. I'm imagining that we need to get used to asking the question 'who are the key media players in this space?'. Everyone will have a different take on this and only after mapping the responses will we be able to appreciate how the lines of communication run, taking us a valuable step nearer to a picture of influence.

The objective of favourability

The discipline of media evaluation has over the years found it hard to dodge the accusation of subjectivity. I have noticed that when executing research using the accepted forms of best practice it is never possible to make the process completely objective.

I have also noticed over the years that media research often returns to the core disciplines of the measurement of tone and the spotting of key messages. When it comes to establishing favourability there is always the possibility of some sort of ‘unfair’ bias.

There are ways to give the research rigour including proportional analysis, headline bias, other branding traits and potency of the wording. The default categories have tended to be ‘positive’, ‘neutral’ and ‘negative’, although it is possible to feel slightly uneasy about what the actual difference is between a neutral and positive item, for example.

If an article mentions an organisation by name is there some value to that, or should it just be classed as neutral? This could be resolved by introducing the categories of ‘slightly positive’ and, ‘wholly positive’ however there seems to a lack of research to support one categorisation over another. I strongley feel PR measurement would benefit from some form of discussion on this area.

Online adspend points the way for PR

As part of my duties to the International group of the CIPR it was off to the House of Commons last night to hear a talk by Baroness Buscombe, the new DG of the Advertising Association. Not immediately relevant to media research but what she said was of great interest to the PR industry.

While she is keen to maintain standards it was also very apparent that content has to be king and if this means pushing the boundaries, then so be it. With advertising revenues from traditional media falling like stones, so cutbacks in all areas of content are occurring. If magazines are unable to sell as many ads they are not going to make up the short-fall with more editorial content. Instead the two will be mirrored, in effect leading to less traditional media sources with which PR can engage.

During this meeting I had the sense that there was an elephant in the room – metaphorically that is…. and it goes by the name of Google, which sucks up a quarter of all online adspend, bypassing whole ad agency machine.

I really wanted to hear about the ways the ad agencies can deal themselves back into the online adspending and more importantly what PR needs to learn from this situation.