The measurement debate continues. Niall Cook makes the case on the Hill & Knowlton Collective Conversation blog that Social media influence cannot be measured.
It all depends what you want to measure. Influence cannot be measured without direct conversations with your audience before and after a campaign to determine its attitudinal and emotional impact. Its expensive and is likely to dwarf your budget and is why such a rigorous approach is rarely undertaken to support a PR campaign.
The digital environment does enable us to identify audiences with relative ease and get close to a measure of influence by monitoring web traffic at points along a communication framework (brand) or buying cycle (sales). Crucially at low cost and in real time.
Niall calls this reach and reckons that it is not an accurate measure of influence. In absolute terms he is correct, but this metric in the form of readership or opportunities to see has been been a staple of the ad industry for decades.
We have the opportunity to dig deeper using analytics to monitor traffic within a network and a web site destination. By capturing IP addresses we can scrutinise reach to an incredibly granular level (location, company, time, frequency, and more). Better still if the audience is sufficiently motivated to respond to a campaign it will provide data that provides the start of an ongoing dialogue, or sales process.
Some PR agencies are starting to share examples of their work. Steve Loynes at Chameleon PR shared an example this week of how his campaign for Siemens generated a pre-qualified telesales database. Its a smart tactic that works in both the business-to-business and business-to-consumer markets.
I’ve previously blogged measurement examples using web analytics. Have a look at Speed launch stats: social networks trump online ads and Measuring the National Work from Home Day digital communications campaign.
Web analytics is one of the most potent tools at the disposal of the PR industry. Finally we can prove that a given input led to an output and it enables us to connect with the language and metrics of the marketing department.
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=0e8e9324-7fce-48e2-a245-7a38d5c6e9fa)
I totally agree that it’s measurable and disagree with hiding behind the ‘you can’t measure it properly, so why try?’ argument. Advertising would have died out a long time ago, if not for measurement, and PR should be no different.
Even if the forms of measurement, CTRs, conversion to downloads, etc, aren’t perfect, they still go a long way towards justifying spend on marketing programmes. In my eyes, PR is much closer to advertising in that regard nowadays, with digital channels enabling standard measurement tools.
I do however ask – whose responsibility is it to undertake the measurement? I think the days of clients relying on its agencies to do all of the measurement are over and it really needs to be an inhouse activity. My clients who measure well never have issues getting their budgets signed off.
I blogged earlier this week on the very topic of the different ways that PR measures its impact http://www.showmenumbers.com/measurement/the-value-of-pr-measurement-part-1. I think there are two ways people seem to talk about measuring “influence”.
1.Potential to influence – I think this is what Niall is referring to. Trying to assess whether a person or publication is likely to be able to influence a particular community and to what extent. (A recent example of trying to distill this into a measure was put together by Jonny Bentwood at Edelman with regards to Twitter http://technobabble2dot0.wordpress.com/2009/05/05/jcpr-twitter-index/. At RealWire we have designed our Influence Rating to do something similar where coverage is concerned – will be posting more about this in the near future.)
2. Outcomes – a sale, improved sentiment, deeper relationship etc – and whether these were influenced by a particular activity e.g. blog post, article, Twitter conversation.
The Potential to Influence can never really be measured in isolation as it is a prediction. It also requires evidence from the Outcomes measure to truly inform and refine it. For instance if a particular person’s blog posts consistently result in a demonstrable change in the level of sales of particular types of products then it is likely to be reasonable to conclude that person has a significant level of potential to influence in that area.
In reality such evidence may be difficult to gather but it is often possible to observe indicators of influence if not outcomes themselves e.g. the extent that certain sites/people drive more traffic to your website for instance or using econometrics you can take a number of these indicators and build a model that provides a more comprehensive measure of what has driven certain outcomes http://www.metrica.net/MeasurementMatters/post/2009/05/22/Review-of-ROI-techniques-Part-6.aspx.
This might sound strange coming from an accountant, but just because you can’t measure something with certainty doesn’t mean you shouldn’t try!
Hey Stephen, you should have been at the Convergence Conversation last night where the same measurement issues were raised in relation to advertising in a converged world.
Talking of convergence, I believe there is going to be a convergence of advertising measurement and PR measurement; ultimately “influence measurement”, and of course following my ebook on social Web analytics you won’t be surprised to hear that I believe this nascent capability will play a key role.
Check out my post on last night’s event here:
http://bit.ly/mAc90
[...] spune ca se poate, cel putin metric/analytics: The digital environment does enable us to identify audiences with relative ease and get close to a [...]
I’m talking about measuring influence, not just measuring things that might (or might not) contribute towards influence.
I don’t disagree with any of the theories here, but I do wish that agencies and metrics vendors would stop trying to tell me they can measure influence when they’re simply measuring things that they think can influence.
In my mind (and admittedly my mind is a bit wonky sometimes) you can only measure influence properly if you do it at the receiving end, not the transmitting end. But that’s proper market research that costs big money, not something you can package up and a piece of technology and flog.
That said, if someone can show me a calculation that takes all these other metrics and turns them into a measure of influence, then I’m all ears.
Buzz and sentiment monitoring is the start. We also run spreadsheets for all our clients linking to project-critical KPIs.
This argument was settled a long time ago. If you can’t measure it, why do it?
[...] Social web analytics key to proving value – Wadds’ PR Blog [...]
[...] media – another tool or a state of mind? Following on from a debate I was reading from Wadds into how do you measure PR and a question Silky asked me about the difference between digital PR [...]