Here at 1000heads we often extol the virtues of using human analysts to derive meaningful insights and recommendations from WOM listening. We also talk about how brands need to be more human when interacting with people in social media (and beyond). For a large organisation to achieve this, and to deal with any customer service issues that arise, there must clearly be some sort of process involved.
But what happens when that process actually makes the humans seem like robots?
Watching the recent Tesco employment story evolve, we saw a well-run and personalised customer service Twitter stream begin repeating the same message to multiple people for hours on end. But regardless of your views on the issue itself something went very wrong with the way Twitter was used to respond to people’s concerns. What’s more, I’ve no doubt that it was a human in charge of the Twitter feed. So what went wrong?
I imagine what we saw here was that Tesco has a prescriptive set of KPIs telling operators how many people they should reply to and a flowchart telling them how to respond in a crisis. These two processes perhaps worked together to make Tesco suddenly appear less human and more like a robotic call centre. Like other Twitter outbursts such as the GAP logo change, this may not have a lasting effect on Tesco’s reputation in the short term. But all brands should remember that human operators are more than capable of appearing machinelike in certain situations. A coordinated, sincere, human response to a situation like this is really hard to pull off, but a modern social business has to aspire to it.
How do you encourage the people running your social presence to come up with a better solution? For me it has to start with measuring the right things. For example, if you measure the % of tweets you are responding to as a success metric, your team is left with no incentive to demand a change to the script when it’s really needed. They’ll just keep on tweeting to hit their quota. It’s the same reason why direct marketing can become labelled as junk mail. The need to hit volumes outweighs the ambition to be targeted and relevant, and brands only entrench themselves deeper when social interactions go wrong.
This problem of blind process getting in the way of delivery has been solved elsewhere. In his excellent book “How to Measure Anything” Douglas Hubbard describes the early days of agile software development, where people measured the speed of work simply because it is an easy thing to quantify and optimise. But when these developers produced a large amount of features that no-one wanted to use, the realisation struck that a gauge closer to a consumer-facing outcome was needed.
So if our supposition is correct the first thing Tesco needs is a shift from speed-of-work based reporting to measurement based on outcomes. In doing so its Twitter team would be empowered with the flexibility to identify an issue and elevate it internally (with the benefit of a linear organisation structure to provide the speed and level of authority needed), which would ultimately allow the brand to respond in a human and empathetic way.
In keeping with our ‘celebrating our staff’ brithday week, we’re handing over today’s post to one of our team you haven’t yet heard from: Rajeev Talwar, our rather fantastic Head of Social CRM, who is going to update us on the latest in WOM measurement and ROI… ^Molls
This week I’ve been reading a timely article from McKinsey about WOM measurement which includes some interesting statistics. The article introduces the concept of ‘WOM equity’ which is described as ‘an index of a brand’s power to generate messages that influence the consumer’s decision to purchase’, and which is represented as the number of WOM messages multiplied by the average sales impact.
I’ve extrapolated six key points from the study which I think are most helpful:
Word of mouth is the primary factor behind 20-50% of all purchasing decisions and its influence is greatest when consumers are buying a product for the first time or when products are relatively expensive.
A high-impact recommendation—from a trusted friend conveying a relevant message, for example—is up to 50 times more likely to trigger a purchase than a low-impact recommendation.
About 8-10% of consumers are ‘influentials’. Influentials typically generate three times more word-of-mouth messages than non-influentials do, and each message has four times more impact on a recipient’s purchasing decision.
Typically, messages passed within tight, trusted networks have less reach but greater impact than those circulated through dispersed communities.
To turn consumers into an effective marketing vehicle (yuck), companies need to outperform on product and service attributes that have intrinsic word-of-mouth potential.
Marketing induced consumer-to-consumer word of mouth generates more than twice the sales of paid advertising.
That’s pretty strong stuff, enough to make any brand owner sit up and take notice.
I was also glad to see that Miele gets a mention as one of the companies that generate successful buzz around new product launches. Having worked with Miele since 2008 we feel that they really do deserve props for the work they do.
However, although the above stats are great, I feel that we cannot use a single metric across all markets. And like may WOM studies, the article is heavily focussed towards an FMCG scenario and does not take into account other B2B or B2C scenarios.
So what are we doing to improve WOM measurement? Well, we’re currently working on a new ROI model.
Our Insights Manager Jacqui Hill explains that “the ROI model 1000heads is testing will give our clients a monetary figure representative of the value inherent in stimulating advocacy among consumers in the online environment. It will assign a pound sign to every single conversation we capture – and that value will be adjusted for sentiment, influence, trust, audience, reach and significantly, the heightened potential for long-term damage associated with detractor negativity.
1000heads’ research has established the ultimate value of a conversation to be based on the number of people who will take the intended action upon engaging with social media content. Without going into too much detail – X amount of consumers read a highly positive blog post about brand A, Y amount of consumers go out and actually buy one of Brand A’s products.
This will enable our clients to not only assess the return on their social media marketing activities and consumer engagement, but to refine their strategy, targeting resources at engaging communities/voices who offer the highest return on investment.
We’re currently testing the model across all of our clients’ online conversations and will formally introduce it once testing is complete.”
As I’ve said before, 1000heads don’t shout much about what we do. But as we move into our tenth year of executing word of mouth campaigns (blimey), it seems about time we start to get a bit better at sharing some of our successes, challenges and learnings with you.
So to kick things off we’ve pulled together a case study for one of the disruptive trials we run for Nokia – in this instance to help launch their E72 device. Hopefully the deck speaks for itself, but please do comment below, spread it around and get some debate going, or get in touch if you have questions or thoughts.
Having arrived back in Heathrow at midnight, I’m still dogged by jetlag and trying to assimilate the great eclectic beast that was WOMMA (Word of Mouth Marketing Association)’s 2009 Summit in Vegas, no less. With three days of keynotes, panels and case studies from some of the biggest US brands active in the WOM space such as Ford, HP and Coca-Cola, as well as research from the likes of Forrester Research, Nielsen and The Keller Fay Group, it was a mindblast of the latest theories and commercial applications of WOM.
Measurement was predictably high on the agenda. Clients are crying out for industry-wide standards, but there was an acknowledgment that meaningful metrics will be be different according to client objectives (visibility, sales, loyalty etc) and therefore project-specific education is still essential. Conversation relies on context, while most ad metrics are stand-alone and focused on scale alone. Consequently, the most successful examples of effective measurement involved a brand combining insights and figures from other departments (sales, eyeballs, customer services calls etc) with a broad range of qual and quant WOM data.
Internal ownership was also a massive issue, with some great sessions from IBM & Newell Rubbermaid and Mars on how they’ve integrated WOM listening and advocacy programmes into their existing structures and processes. This was related to an ongoing conversation about how Customer Services links with WOM. A panel including Pete Blackshaw from Nielsen, Frank Eliason from Comcast, Tom Asher from Levi Strauss, Denise Morrissey from Toyota and John Bernier from Best Buy looked at examples such as @TWELPFORCE and @comcastcares which fully integrate Twitter into CS. The main takeaway was: just try, keep communicating, and help employees learn and progress from their mistakes. Take the risk, and as long as your approach has integrity and strategy behind it, the benefits will be enormous.
Another highlight was Steve Knox from P&G’s Tremor using cognitive psychology to explain why customers talk – apparently if you disrupt their schema (the model in their head of how the world works and their assumptions about a brand) it’s WOM gold. And the panel of WOM academics tackling the toughest questions in the industry had some powerful messages, in particular the importance of overlooked visual, aural and offline WOM triggers; the need for research into geographical and cultural differences in behaviour; and the use of future visioning to sell in the value of WOM to brands: if we do or don’t engage this talkative customer, what will the impact be?
Steve Knox from Tremor on cognitive psychology. Spot me earnestly taking notes on my Mac…
On the flipside, some of the examples I saw were still too based around an old-school marketing approach. Isn’t a moderated, branded page or forum in an independent community (such as Tropicana for BlogHer) really just a microsite dressed up in social clothing? And as President of WOMMA’s sister organisation WOM UK, it was interesting to observe the differences in approach between the US and UK. I’m not sure that some of the more gung-ho, blatantly branded adovacy groups such as the Feld Family Activators at Mom Central would gain much traction in a nation that tends to be highly sceptical of associating itself so strongly with commerce. And some agencies were even stipulating time limits whereby participants were ‘expected’ to talk in return for goods or experiences – where’s the spontaneous, independent and heartfelt advocacy in that?
Overall it was a rich and stimulating event and I’m sure more thoughts and observations will trickle through across the next few weeks. For more, check out my live tweets from the Summit @WOMUK, as well as video highlights here and photos here. And if you want a more detailed lowdown on insights and issues raised, just drop me a line and I’d be happy to take you through it over a coffee… or indeed a Vegas-themed cocktail.
Back in the day, bloggers were heralded as the new journalists, and subjected to all kinds of spammy PR approaches that didn’t respect their unique culture and aims. Nowadays, marketers are much more skilled at approaching them in a more personalised and mutually beneficial way, but the new fear is whether bloggers are the ones brands should be pursuing after all. Measurements of blog influence are becoming increasingly nonsensical, but it can be even harder to define the influence of those using other venues.
As always, a good rule of thumb is to look behind the tools to the individuals using them. Bloggers still have a unique and defined position in social media, and microblogging isn’t just an ‘evolution’ of blogging. Historically, bloggers have identified themselves in a more self-conscious way than tweeters, or social networkers, or forum members. They invest more time and space in constructing their own view of the world, in a more sustained and permanent way. They encourage people to migrate to their own property, rather than co-existing on one shared platform, and maintain a more rigid hierarchy of author and commentor as opposed to conversational free-for-all.
This doesn’t mean that blogging is in danger now that we’re all more semiotically promiscuous than we were when blogs first came on the scene. Just as in real life we sometimes want an in-depth discussion about Socrates on the sofa, and at other times we want to toss out a truism about Madonna at the bar, there will always be space for all forms of self-expression on the web. Brands need to always look at who is talking and what their passions are, and then tailor their word of mouth strategies to engage with different people at different stages of the campaign. A link to a twitpic of an exclusive new product has a different appeal, audience, and home on the net than a detailed review of a two-hour hands-on trial session organised by a brand.
We’ll probably be writing a nostalgic piece about Twitter in a couple of years, but I can guarantee that people will still be talking, sharing, recommending, on all the different platforms at their disposal, just as they always have. The fuss about ‘hot’ and ‘obsolete’ platforms is actually driven by anxiety over measurement – how marketers can keep a handle on the influence and effects of new tools. That’s understandable, but that fear shouldn’t stop brands from embracing and utilising all platforms in a flexible and inclusive way.
Measuring the value of a word of mouth strategy always involves two elements: getting some reassuring metrics to execs who think in terms of reach, eyeballs and direct sales; and educating companies about the new and different ways in which WOM brings long-term value to them.
The metrics can in fact be more precise than traditional measurements such as vague print circulation figures or possible viewers of a TV ad. We can track the volume of WOM directly stimulated by a project; measure the wider engagements with that content to show how the WOM ripple has spread; calculte the financial value of a positive unit of conversation (and the cost of a negative unit of conversation) for a brand using sales data and industry research; collect statements of intent from those who have come into contact with those conversations and been influenced by them; and overlay internal sales and marketing figures to show correlations between their activity and WOM.
The second type of measurement is a little more tricky, as it asks companies to question some of their entrenched beliefs about ‘value’ and move away from a reassuring but possibly misleading reliance on numbers. It includes demonstrating the power of quality engagement with fewer, more passionate and vocal advocates rather than ‘viral’ spread to a large number of disinterested people; examining how word of mouth has impact on all areas of a business, not just in marketing but internally; and – something we are particularly evangelical about at the moment – demonstrating how an effective WOM strategy can shift the emotions that adhere to a brand in social media, stimulating a range of positive emotions that are key in securing long-term loyalty and sales.
Each day more businesses are awakening to the power of the social web and there is one thing separating the successful from the unsuccessful: Knowledge. Without knowledge a business can quickly get lost in the “sea of social technology” losing precious time doing unproductive activities without achieving a desired result. Money is a derivative of knowledge. The lack of knowledge is a derivative of cost.
Deragon’s belief that brands need to cultivate this learning internally is spot on – internal knowledge and skills transfer is an essential part of our work. It not only ensures that a WOM strategy has longevity and cut-through within a company, but helps execs and employees alike understand the sometimes unfamiliar yardsticks of value that a WOM strategy should be measured against.
Lots of talk about Twitter and its effect on link sharing. A little while back, Brian Solis and Dorrion Carrol debated whether Technorati’s “authority” model was being affected by micromedia (mainly Twitter). Today another similar post on SEOmoz.
The wider point being considered is whether property development – a more traditionally PR focused blogging practice of growing audience and building traffic – is decreasingly relevant as a pin to hang influence on.
It is. Tracking and measuring “influence” and “authority” based upon venue based link structures is limiting. Social media is about presence over platform and identity over institution – we are everywhere, not somewhere and influence is a hobo.
Interesting ‘WOM espresso meeting’ this morning with various agencies and companies considering ‘What alternative media can do for your brand in a recession’.
Ivan Palmer from Wildfire presented a coherent and practical deck (which you can see here) about approaching word of mouth as a pervasive discipline, and Steve Barton from WOM UK (in which we are a governing member) chaired discussion around challenging assumptions around metrics, influence and quick wins. My favourite quote from Ivan: “you can’t just dip your toe in anymore. The time for experimentation is over, this is about long-term strategy”. If you’re interested in coming to the next one, focused on measurement, let Steve know. The bacon butties are particularly good.
A WOM ripple can be defined as the spread and exposure of a piece of conversation or content across online and offline scapes, as it is relayed, repurposed and used to inspire further conversations. There’s lots of commentary devoted to identifying effective ripple triggers, such as the whitepaper on Spreadable Media I mentioned yesterday; key qualities include the relevance, creativity, adaptability, originality, informative value and emotional engagement of the trigger, as well as who is distributing it, where, and why.
However, it can be notoriously difficult to help clients envisage exactly how that conversational trigger then ignites in the space, accelerating through search; social networks, communities and interconnections; and aggregation, distribution and bookmarking tools. With the right software and research layer, it’s possible to measure the numbers, but it’s still tricky to illustrate the extraordinary, unexpected and synaptic spread of ideas and images across a host of different communities and scapes.
For example, brands rarely take into account offline visual cues. Mike R relates the story of his new jet-powered car cleaner, which quickly appeared on several driveways on his street without him uttering a word of recommendation to anyone. We’ve all had similar experiences; I’ve been looking for months for a pair of boots I saw on a girl walking down High Holborn. Any truly pervasive word of mouth project should be finding ways to tap into these incredibly powerful offline triggers and platforms for the WOM ripple, even if they’re a little less definable to a CEO.
The concept of the Social Graph has historically been used to examine personal interconnections online, and Beth Kanter’s new post The Conversation Graph: The Social Life of A Blog Post attempts an update by examining various tools that track the journey of a piece of content. But there’s still no clear illustration of how a trigger spreads both on and offline, with all the corresponding complexities and unexpected factors that invariably pop up when you’re dealing with human behaviour – not least the fact that content spreads between people who aren’t visibly connected.
To accompany the quantifiable figures, there’s a real need for some in-depth case studies, taking individual examples of a product, service or brand experience and tracking in minute detail how it spreads. This should use social media measurement tools but also interview and examination of the family, community, social context and behaviour of the person involved – as well as where, how and if it translates into a sale. Watch this space.
Interesting discussion going on over at the Economist’s Technology Quarterly about just how important online reviews have become, with new research from Bazaarvoice showing that “visitors are more reluctant to buy until a product attracts a reasonable number of reviews and picks up momentum”. But with the new ‘helpfulness’ rating on Amazon reviews emphasising quality as well as quantity, measuring the ‘value’ of these conversations can be a complex beast.
I spent a rather excellent couple of hours with our strategy team yesterday, discussing a unique new model they’re working towards for measuring the influence, impact and financial value of word of mouth – exciting times.
For the moment, check out the full Economist post here.