Marketing Libero

The Loyalty Paradox

 

Trawling through some client research I am struck by how consistent customers are in their definition of loyalty.

It would seem for the majority of customers it is about longevity of relationship.

However, for most brands loyalty (and hence rewards) is about intensity of relationship, or in other words - spend levels. The more you spend the more cash you get back or the more points you accrue.

I accept that many marketing practitioners would wish to split the assumption that a competent reward scheme equals a loyal customer base, wishing to focus instead on creating magic moments with the brand and driving up the holy grail of loyalty metrics - net promoter score (NPS).

(The uncomfortable truth for many service brands is that their NPS scores are mostly negative, put simply they have more customers who would not recommend them than customers who would).

But nonetheless this sets up a classic loyalty paradox between customers and brands.

There are some excellent reward schemes out there undoubtedly helping with customer retention (a very different thing to loyalty) and I accept / even extol the CRM 101 pragmatism of focusing your resources on your most valuable customers.

I have blogged previously about the holy trinity of loyalty; structural, emotional and rational. There must be a way to resolve this paradox using this important and consistent customer insight on the definition of loyalty.

As practitioners we surely now have the data & CRM capability to build longevity, as another dimension to value, into both our reward schemes and our customer service policies.

Welcome

Having just bought a new iPhone from CPW on Orange I am reminded of the importance of a good welcome process.

During the first 12 months of any relationship between a customer and a service brand it is estimated that 75% of all X-sell & up-sell happens and particularly in the early months churn risk is at it's highest.

Also the relationship is set. For example whether marketing permission is given for contact, whether key data is collected such as email addresses, etc.

In short welcome is the cornerstone of any CRM programme. Yet more resource is often focused on retention and mature customer up and cross sell.

Five quick rules:

1. Ring fence new customers and make welcome someone's responsibility, they will have to work X-functionally of course.

2. Work out what a successfully welcomed customer looks like, preferrably using behavioural rules to gauge engagement - these should be in line with your business aims and set your communication objectives.

3. Synergise service and marketing communications, ideally making them feel like part of the service.

4. Choreograph your channels and have a single contact strategy, preferably with the website at the heart of your welcome strategy.

5. Create a welcome dashboard with a holistic KPIs to track your progress.

Data visualisation rules

 
 

This is a subject getting increasing conversation time on various blogs. One of the consistent challenges is to avoid visualisation for the sake of it.

In thinking about our rule set I like very much Manuel Lima's information visualisation manifesto:

1. "Form follows function
2. Start with a question
3. Interactivity is key
4. Cite your source
5. The power of narrative
6. Do not glorify aesthetics
7. Look for relevancy
8. Embrace time
9. Aspire to knowledge
10. Avoid gratuitous visualisations"

However in avoiding Lima's points 6 & 10 Edward R Tufte is still the authoritative source. Below is Tufte's own manifesto quoted from his excellent book page 13 - the second bullet encapsulates it all really.

"Excellence in statistical graphics consists of complex ideas communicated with clarity, precision and efficiency. Graphical displays should

• show the data
• induce the viewer to think about the substance rather than about methodology, graphic design, the technology of graphic production, or something else
• avoid distorting what the data have to say
• present many numbers in a small space
• make large data sets coherent
• encourage the eye to compare different pieces of data
• reveal the data at several levels of detail, from a broad overview to the fine structure
• serve a reasonably clear purpose: description, exploration, tabulation or decoration
• be closely integrated with the statistical and verbal descriptions of a data set.

Graphics reveal data".

The image above is one of my favourites from his fantastic collection of visualisations - Napoleon's march into Russia. The brown line is proportionate in size to his army as he begins his campaign. The advance of the army is shown across a map of Russia. The black line is Napoleon's retreat with date and temperature plotted - revealing with great articulation the human tragedy.

Being creative with data - IPA event

Link to a recent IPA event I helped to organise. Thanks again to all those who participated.

Do You Get Value from Sharing Your Data

Clearly this is a rhetorical question.

One of the constant themes that interests me is the value consumers receive in exchange for providing their personal information to large brands.

A few companies aside, evidence would suggest that the value exchange is stacked in favour of the Big Brands.

This asymmetry takes many forms - most commonly:

• Repeating your details multiple times when you call the contact centre
• Spamming / Junk-mailing you with inappropriate offers that you have failed to respond to time and time again
• Etc

Those brands that will be successful in the future get the following:

• How you use your customer's data is part of the brand promise - this goes beyond simple security and appropriate use
• Through demonstration not declaration there needs to be a clear value exchange for supplying your personal details
• Finding ways that customer data can be applied to help them manage their service with you rather than just for your company's benefit

What's provoked my latest tirade?

I called my mortgage provider. At first the voice recognition software was excellent, far better than fighting through a push button IVR system, only to have to repeat every single detail to the agent in Mumbai!

I love wordle.net

I am fairly obsessed with data-visualisation and telling stories. A part of the digital age is the creation, storage and analysis of different kinds of data. We need new techniques.

Excell is great for numbers and graphs but it lacks a certain creativity. 

What about analysing and visualising words?

At wordle.net you can do just that. By simply cutting and pasting in a section of text it creates graphic "clouds" of words where size is proportional to frequency. It is surprisingly effective.

One of my own favourites is below. You have a number of options for analysis when considering the question - who are the most successful international football teams in the history of the World Cup?

Looking at teams who have made the semi-final you could  - produce a table, build a complex score and graph it or simply do this...

Loyalty's Holy Trinity

This week both Virgin Media and Sky communicated their results for the last quarter. This is a service sector driven by investment analyst sentiment to be fixated by the number of "net-adds". In simple terms the number of new customers minus those customers who left.

Sky stated 80,000 new customers driven by their HD-offer as they hurtle inevitably to their stated target of 10m households with a Sky product.

Virgin's figures were more modest but far more interesting citing customer churn at an all time low of 1.1%

Once you sort involuntary churn (bad debt, etc) from voluntary churn for most service brands there is a generally held belief in the Holy Trinity of loyalty...

• Emotional engagement
• Reward schemes
• Structural entanglement 

Emotional engagement

By far the most powerful loyalty factor and the one we all strive for. It is made up of all those emotional and intangible aspects of brand experience. 

It takes time to build and can be re-enforced by genuine "magic moments" and can be lost in a single interaction. It is often expressed in advocacy - a customer's potential to recommend a brand and measured in terms of net promoter score (NPS).  

Reward schemes

Sorry I can't bring myself to call them loyalty programs. These are a very useful tool particularly where there is a genuine value exchange between the customer and the brand. There are of course good examples and bad. 

Often a key objective is to remove the anonymity of retail or third party transactions through data collection as well as to reward the right customer behaviours. 

It is often overlooked that the loyalty is to the reward scheme not the brand. For example, people follow AirMiles not necessarily the specific credit card. 

Whilst useful reward schemes can never overcome "structural faults" in your service proposition or customer experience.

Structural entanglement. 

What is meant by this is often referred to as "locking-in" customers through a 12-month contract for example. This is often justified (not unreasonably) where equipment is needed to access the service and is given for "free" e.g. a mobile device, modem / router, set-top box, etc. 

It is not common practice to use price promotion periods as a lock-in as regulators see this as anti-competitive. Other forms may include early payment clauses and depth of relationship. 

A good example of the latter is current accounts where the hassle of changing direct debits / standing orders, re-setting up of internet passwords, etc are a formidable barrier. Here structural entanglement is made exponentially more powerful by the second most powerful force in the universe - apathy. (The first is serendipity). 

However, whilst it was once said that you were more likely to get divorced than change your bank account this is perhaps less true these days. The opening up of utility markets to competition, the decline in trust of big-brands, the commoditisation of once intermediated services (home insurance), the transparency facilitated by the internet, aggregators and the current global financial crisis are changing customer attitudes to switching for good.

Also this week I considered changing my standard Sky wireless modem router to improve the coverage in my house. Not unsurprisingly my wife is "sick of wires everywhere". However, to my chagrin I found out that as a Sky customer I can only use their equipment on their broadband service. 

They justify this by the fact that the equipment is free and they do not want people canceling within the introductory period and using their modem routers on another supplier. 

After quick but not exhaustive research none of the other providers seem to have this stipulation. Now their Netgear equipment is OK but N+ standard it is not. Apparently this is in the T&C's and a quick review of the relevant blogs found examples of customers who had hacked-in to get the relevant details to use another modem router and been subsequently cut-off.

Now its about a value exchange. This T&C would be fine for the first 12-months but after that it's plain unreasonable. I was an advocate of Sky broadband but now because of this insidious piece of structural entanglement - I'm open to offers! Will I leave tomorrow? Probably not but the seed of discontent has been sown.

Walking the Data Talk

Tesco's recent announcement that it intends to relaunch and further invest in its Clubcard confirms its position as the number-1 company that gets data is the future.

The card has always been more than just simply about loyalty (Nectar are you paying attention?). The points / vouchers are positioned as a "Thank You" rather than as a reward for custom. Most importantly it's core purpose is to remove the anonymity of transactions by collecting customer data. (You can read more about the Clubcard story in this book).

The latest manifestation of its data driven strategy is the announcement that they will be opening as a bank offering current accounts from 30-branches. (previously they sold only secondary financial products). 

Whilst the Big Banks may not yet be quaking in their boots, with Tesco's ability to disproportionately amplify its proposition compounded with post credit crunch consumer sentiment and their ability to execute, the Old-Guard should be very concerned.

They "walk-the-walk" whilst other brands just talk, talk, talk.

Big Brother or Marketing People Want?

The news that the BBC is to use behavioural targeting on its global websites outside the UK (Marketing Week 27.04.09) is an important milestone for this (relatively) new and highly controversial technology for 2-reasons...

1. It is seen as a key enabler to commercialising their websites
2. If the ultra-cautious BBC are happy to adopt it should your company be doing the same?  

  
Behavioural targeting uses knowledge of the websites you visit to serve you more relevant online advertising. It is seen as a crucial technology for both ISPs and web owners to monetise their sites, and to revitalise online display advertising.

It's controversial because of privacy - do you want your web-surfing habits to be tracked and used in this way? The answer is of course - it depends! If it is genuinely useful then yes. There is a distinct generation gap in the attitude of those consumers (often younger and social-networking literate) who understand the value of sharing their personal information.

BT got is spectacularly wrong last year in its test on the Phorm behavioural targeting platform. Rather than invite customers to participate in a beta-test (Google style) it picked customers randomly. Whilst their anonymity was assured the customers in the trial did not appreciate being selected without permission.

Online display advertising is at a bit of a crossroads - despite comprising 19% of online advertising expenditure in 2008 (IAB) the effectiveness of this media is coming under intense scrutiny. Network buys have become the norm where you pay for impressions but are blind to the web sites you might end up on with the organisational risk and wastage that entails.

On a more positive note the combination of behavioural targeting and the interactive display formats that are increasingly available promise greater customer relevancy and a better branded experience - potentially moving from the interruptive to the genuinely engaging.

If this sounds to you a bit like the cookie debate of 5-years or so ago - you'd be right - using much the same technology. Whilst consumers are beginning to accept that cookies are useful for improving their internet experience there is a way to go to win the debate on highly targeted advertising. 

Brand and digital media owners need to demonstrate to consumers that they can use behavioural targeting responsibly to genuinely deliver marketing people want. The IAB's good practice guide is an excellent place to start.

How to cheat at direct marketing ROI

Calculating how to measure direct marketing ROI (return on investment) is fairly simple:

Sales X Value / Cost = ROI

Below is a guide as to how you can improve your ROI in 5 simple steps...

1. Measure response at the GROSS level

Rather than count only incremental sales (target cell sales minus control cell sales) from your campaign simply count all the sales.

Control cells can be tricky to set-up so why bother?

Further measuring incremental response can be hazardous to your business case as for most campaigns it is only 20-40% of gross response. This is particularly true for brands which have other sales channels such as a retail network.

2. Massage your value

If the sums are getting tricky simply inflate the value of the product or service.

Don't think seriously about whether you should use a 1,3 or 5-year NPV (net present value) just use the highest one! Never mind that to achieve the higher 5-year value might involve further customer investment.

Whatever you do don't look back 6-months later at the actual NPV of the customers your campaign acquired and re-adjust your ROI expectations for the next campaign. This so called Total Value Management approach can make any business case dead in the water.

3. Drive down your pack cost

Yet another way to make that business case is to drive down the cost of the communication. For a letter and A5-leaflet it is perfectly possible to get something produced for 24p including postage.

This is about economies of scale. Just mail a huge volume well past the point of customer relevancy (see step 5) and get the stuff printed in Eastern Europe.

Don't let the allure of the agency brains trust fool you into buying any of that expensive strategy & planning stuff.

Just ignore those brand communications or CRM guys who insist that direct marketing has a role to play not only as a highly optimised sales channel but one that can build relationships too.

4. Don't do test and learn

It just drags down your overall ROI.

It's tricky to set-up. If you want to learn anything worthwhile you have to really think hard about it. Plus it takes ages to get results back.

5. Evaluate the campaign at an overall level

I have left the most important tip till last. Make sure you present your results at the highest aggregate level - blended results can hide many sins.

For example, don't look at the results disaggregated by response decile. Whilst you can get positive ROI on mailing the top-6 deciles (60% of the base) it is very inconvenient that the top-2 deciles only create all the profit whilst the next 2 only break even and the next 2 have negative ROI.

Don't let anyone see (especially those finance guys) that by mailing half the volume you could achieve c85% of the sales, 95% of the profit and double your ROI.

You never know what might happen - they could halve your budget and give it to those web freaks!

Of course if you want to measure your direct marketing properly you'd do the opposite of this guide but then that involves doing communications that genuinely engage your customers creating additional demand, rather than simply hoovering-up sales you would have got anyway through other channels.