Taking a project to MOSCOW

Moscow, as everyone knows is a city in Russia. Hot in summer, very cold in Winter. What you might not know is MOSCOW is a good way if breaking down project requests into the absolutely necessary and the nice to have. In this regard, dropping the letter "O" from Moscow gives us MSCW -

M = Must Have
S = Should Have
C = Could Have
W = Wish List

This is a convenient way of managing unwieldy project specifications by prioritising requirements, particularly those where there are either: a) many objectives, b) many stakeholders, or c), development or delivery complexities.

M = Must Have
These are the only requirements that the project must deliver. A requirement can only have this priority for one of the following reasons:
a) They are mandated by law or an (internal or external) regulatory body.
b) They are high value business requirements delivering the benefits on which the Business Case for the project is founded
c) They are requirements which must be met to enable requirements of the above types to be met.

S = Should Have
These are medium value requirements, directly delivering business benefits, but the Business Case for the project is still valid even if these requirements are not met.

C = Could Have
Requirements which would be useful but do not directly deliver any of the business benefit on which the Business Case is based.

W = Would Like to Have
These are either:
a) Stand-alone requirements that could be postponed to a later project (or project phase) without impacting on the benefits or usefulness of the core functionality being delivered.
b) Minor requirements reflecting personal preferences, e.g. cosmetic

For example, regulatory and compliance issues would be tagged "M", whereas minor cosmetic details not affecting operations would be tagged "W". This allows a business case to be more focused and ensure that scare IT development resources are applied more intelligently, in pursuit of the most ecomonically beneficial tasks.

Managing Brief Creep

In Customer Management, like many other fields, there is always something more that can be done. There are always new initiatives that can be rolled out to customers, there is always more analysis that can be done to improve targetting and there is always more information that can be provided to advisors (or customers through self-service channels) througk Knowledge Management, training and systems. Consequently, managers in customer management are never short of development briefs. When you add to the complexities of managing cross-channel interactions with, say, a technology refresh or a merger incorporating two diverse platforms, there is rarely a time when managers in this are conclude that they have all of the systems and support that they need. So, specifications and briefs abound. Managing these briefs and delivering greatest economic value back to the business is a testing role.

The factors above and the constantly changing world of marketing communications means that briefs evolve as knowledge and understanding evolves and as the business responds to the dynamics of the marketplace (new products, regulation, competition, etc). This results often in brief creep. Using a MOSCOW prioritisation scheme is helpful in managing down the extent of the creeping and focusing attention back onto that which is really delivering economic value.

Copyright 2006 Robert A Innes

Guidelines for Personalising Content with Real and Assumed data

First, some definitions: real customer data is discrete data elements that can be related to an individual with certainty. Types of real data:

Personal Date: name, address and telephone number data (commonly referred to as NAT) is real and associated directly with individual customers.

Transactional Data: Contract Start Date, Last Order Data, Offer Code Used, etc. which come form a brand’s interaction with customers

Profile Data: information about the interests and preferences of customers, gained from survey information or during telephone contact. This might include, Yes/No selections for subscriptions to different email newsletters, Opt Out from 3rd Party Marketing, Source of Acquisition, etc.

Depending on your customer data set there may be additional data elements that are relevant for particular sectors, e.g. airlines might gather seat and meal preferences, whereas, sports clubs might gather dates of fitness assessments or personal health indicators such as resting pulse rate.

NB Some assumed data can look like real data and be misused as a result. For example, a data element in your website data set may be Data of Last Visit, to track frequency of customer visits. If a sign-in process is used, e.g. for online banking, then Date of Last Visit is reliable can be used as real data for personalisation, thus: “Since your visit on DD/MM/YY we have made some changes. Click here to learn more about these changes.” Located behind the sign-in process, this customisation is valid. However, if the Date of Last Visit is dependent on cookies to identify individual repeat visits, then use this data with care. Two problems here: firstly, generally cookies identify the machine used to visit the website, rather than an individual, and secondly, visits made to the website from different machines by the same individual will not be related. Consequently, customisation such as the example above, cannot be relied upon and it may be more appropriate to refer to the changes more generally, such as “Recently, our website has undergone some updating, please click here to learn more about these changes.” In severe cases using assumed data as real data can have negative effects on customer engagement.

Where does “Assumed Data” come from?

Many sources exist including: Transactional analysis, RFV Models, Click-thru analysis, profiling systems and possibly from data mining, etc. Examples of assumptions that could reasonably be made based on customer interactions:

For a newspaper web site, a customer who clicks on Golf and Tennis links, might reasonably be assumed to be interested in these sports, and a successful use of this insight might be to present these links higher up in the order of priority for the customer, so, upon future visits, the links are earlier in the content stream or are more prominent. An unwise interpretation is that the customer actually plays these sports. This is an important distinction as a possible use of the data might be to include “Customers who are golfers” in a selection list for an email that offers a discount voucher for golf clubs. Interest in golf is a correlate of interest in golfing, but the relationship cannot be relied upon.

For a travel agency website that books airline flights, hotels and car hire through travel providers, they will gather much information on customer interests by observing the searches and the bookings made. Bookings are a subset of searches as not all searches will result in a booking (no availability, too expensive, inconvenient, etc.). Data from prior searches could potentially be used to inform display options in current searches, for example, if in the past, the customer has rejected flight options that are indirect, perhaps a “direct-only” filter could be added as a recommendation or applied by default. This might be a convenience for the customer and reflect learning. There is, however, a possible wrinkle in this logic: what if the customer books for business and also for pleasure? Perhaps direct flights are important on business but less so for pleasure. In which case, the filtering might be inappropriate.

The best guide is to use assumed data carefully and not overly specifically. There are many opportunities to intelligently reflect likely customer interests in a content stream and the increasingly common digital content streams we are exposed to now gives more options that ever before. This can be very usefully deployed, for example to present advertising alongside content, to provide recommendations for other content that might be of interest based on current consumption, or to up-weight / down-weight certain content areas based on observed behaviour.

Permission Based Marketing

Permission marketing deserves a mention here. Capturing data from customers increasingly leads to concerns over how the data will be used. The more specific and personal the data, the more commercial the future use, the more customer effort required to provide the data all lead towards more use of permission. Nothing wrong with saying to a customer – “you seem to visit our site often, tell us what you like and we’ll gather more content that you’re interested in and hide content you’re less interested in.” It is important to continue to do some varying of content (people’s interests change, special events can spark interest in previously ignored areas) as novelty is a great thing and may positively disrupt content consumption and extend customer engagement. That said, using an email address without permission absolutely does not fall into this category. A good example, would be a mobile phone operator who has enabled customer bill preview on their website. Here, sign-in is required to access the bill so there is good integrity on identifying the individual. Placing some low-noise content alerts for novelty content, say “World Cup Offer” to those not interested in sport, might work.

Recommendation engines

Software that takes customer community intelligence used to offer suggestions for an individual based on selection decisions previously made by a community of similar customers. The mass of User Generated Content now appearing, gives more data for these algorithms to work with, hopefully generating more utility and becoming less visible - ambient technology - it's there but it's so tuned to your needs, you don't notice it.

Finally, it is important to note that this can all be accomplished through databases, campaign management and content management, not through building individual plans for individual customers. This can only work commercially if it doesn’t tie you knots trying to achieve it!

Copyright 2006 Robert A Innes

Weaving Contact Strategies across Channels

For years consultants have been advising organisations to move away from product based systems to customer based systems. This is a basic, necessary step that all organisations must take before they can start to operate intelligent marketing processes. Intelligent in this context means, using data from customers and their interactions with the brand to create ever more engaging, compelling and relevant marketing communications. Why? Well, if you want to lift return on investment you need to be smarter – consumers are better informed and more savvy than ever before. They are also promotionally promiscuous and have absolutely no notion of “relationship” to the activity you refer to as Customer Relationship Management.

Now we have the World Wide Web (the only thing I know of whose shortened
form - www - takes three times longer to say than what it's short for). - Douglas Adams

Smart contact strategies will be relevant and engaging for customers, rewarding them for their investment in time with either better value products and services or higher utility options. This improves overall return to the brand, either through increased product spend, increased product penetration or reduced churn. The purpose is absolutely to lift ROI.

With a portfolio view (looking at one customer’s record and seeing all of their products in one place), organisations are able to engage with customers in a more intelligent manner. This is however, only the starting point of intelligent marketing systems, not the end of the road. A path is formed by laying one step at a time, and consequently an intelligent marketing system will develop over time as the building blocks are put into place. Organisations should be prepared to invest for the long term and build a capability in individualised marketing communications. The investment in time and resources will be well rewarded with significantly higher returns.

A contact strategy is the term given to a series of actions and events that enclose a marketing communication. This uses information about the individual customer to create a personalised experience with standardised templates. By taking the correct content, the correct products and the correct offers together and presenting them in the right order and format for different customers is the contact strategy.

A contact strategy may be as simple as selecting all customers who do not currently have insurance and sending them a particular mail-pack with a 10% discount as the offer. Or, it may be significantly more complex, including many different customer selection groups (cells), multiple contact types (mail-pack, email, telephone, etc.), multiple contact instances (mail-pack plus postcard or mail-pack plus telephone call, for example), multiple offers (10% off, 1st month free, £50 Amazon voucher, etc.). Managing the processes of overlapping communications across channels, is what we refer to as Channel Weaving, and is the intelligent application of personalised data to build on the dialogue (i.e. add information, engagement, sense of brand values, etc) established through earlier communications.

For example, let’s look at a contact strategy for a Hotel Group. They have a good database of previous visitors and enquirers to which they send direct mail packs and electronic direct mail (emails). For instance, an email that follows a mail-pack would be consistent with the message of the mail-pack, but will build on the communication, offering new information and extended content. Using email allows the company to exploit more wonderful resort photography which is naturally limited in scope due to the cost of the production process. In digital media, however, a click-thru on a follow-up email can offer rich media options such as 360 degree views of hotel features. This allows the Hotel Group to play to the strengths of the different media options and, critically, for each media to be “aware” of the building dialogue, using profile information to tailor content and extend the use.

More intelligently yet, information that is gained during one round of communication is acknowledged and developed in the second round. So, for example, take an Airline Frequent Flyer Programme, where a series of emails are used to reinforce the brand values and the services available to individual members. In such a scheme one email may generate a certain click thru pattern that the second email would refer to, such as redistributing content areas around the email to reflect earlier click thru patterns. Say, a particular customer clicks through on item seven in the first email to content about long-haul holidays, then the next email in the sequence would up-weight Long-Haul content for that customer. Obviously content selection models will be very much more complex than this and cover many different customer types, but the point should be clear – brand personalisation leads to customer engagement which leads to marketing success.

Yet another example would be to use information from a telephone conversation to intelligently customise a follow-up email. This is particularly important as the customer likely remembers some parts of the call and a lack of personalisation in the follow-up becomes very obvious. For example, take a business-to-business example of an office products company that manufactures and sells imaging products (like photocopiers, work-group laser printers, etc.). The buying decision for different companies will be different – for some, speed of production will be important, for others it might be quality or finishing options, and for others it might be cost per copy. Delivering one communication to all types of companies which gives all a relevant and engaging presentation of the services offered would be exceptionally difficult given the widely differing buying criteria. Consequently, the marketing team uses information from earlier communication (the telephone call) to log interests and requirements and then customises emails based on need and sector. So, for instance, one email might lead with Quality and Finishing Options and give a case study in the Logistics and Distribution sector, whereas another email might lead with Cost per Copy and give a case study in the Government sector. This provides different prospects with a tailored experience, yet is driven from one marketing system.

This approach gives maximum flexibility and control to the brand and relevance to customers. Personalising communications is easy with a handful of prospects but with thousands it is impossible without a good marketing communications system. Channel Weaving can deliver quantum leaps in marketing performance for those brands that invest in serving their customers and prospects intelligently.

Copyright 2006 Robert A Innes

Video Call Considerations

As they say, “online, no-one can tell you’re a dog”. So with call centres, in a voice-to-voice communication, visual information is missing from the dialogue. This has traditionally led to problems in communication, for example, removing the visual cues/queues that we all use to add “colour” to our conversations by way of expressive gestures. When you only have voice, the remaining elements of the communication, such as content and tone take on more importance. When Video is added to the call centre mix, one might think that it simply adds back the missing visual element of the communication; however, video is much more subtle and complex than that.

Indeed, what is there that does not appear marvelous when it comes to our knowledge for the first time? How many things, too, are looked upon as quite impossible until they have been actually effected? - Pliny the Elder

You have a great face for Radio

With a video-enabled communication, for example between a mobile 3G customer and a video agent with a screen mounted webcam, there are some team considerations that are unique to video:

Selection – do you select new staff based on physically attractive (whatever that means) criteria? – of course not, this would be highly provocative and unfair and de-motivate the entire centre

Presentation – how do you dress the advisor and the advisor environment? – keep it simple! Avoid high contrast combinations and avoid bright colours, esp. Red.

Formality – video calls are uncommon and there is a novelty value associated with them – as a result, customers are even less likely to stick to the script and the call structure is likely to be less formal than with voice calls

Responsibility – the video team may well become a desirable team to be a member of – this is perhaps due to the less formal nature of the conversations and the greater perceived variety – this can be used to motivate the rest of the team to achieve membership

Video Call Examples

In the United States, there is a fantastic example of video calling working and having a positive impact on society. CSD was established in 1975 and provides communication services to the deaf community. They employ over 4000 people and service 1525 call center seats. CSD created the first FCC approved Video Relay Contact Centres in 2001 with the CosmoCall Universe platform.

Video Relay Services (VRS) allows a deaf person to make a phone call by using an Internet video connection. VRS interpreters then permit conversations to occur at close to normal speed in sign language, for deaf and hard of hearing consumers. Previously, deaf subscribers typed their request on a TTY device to a remote agent who conducted a voice call on their behalf with a call centre (say, the deaf customer’s bank). With video, the deaf person can sign to a video agent and the communication is significantly speeded up and provides greater convenience to the deaf community.

In the United Kingdom, since November 2004, Vodafone have made Video Welcome Calls to new 3G contract subscribers as part of their welcome process. Working with Vodafone, we developed a process to educate new customers about video-calling. The best way to do that is to show them the capability and hopefully encourage trial. The 3G video call has huge novelty value and is very well received by customers. There is an education element – getting customers to use the camera correctly to project their image, but mainly the call focuses on how cool the new technology is. Customers instantly see application and are enthused about trialing the technology with their friends (those with 3G devices).

Further Use of Video

One contact centre provider – Cosmocom – is developing a Video IVR service which should make adoption of IVR self-service technology far greater. Traditional IVR where you listen to a script and then push buttons according to the choices available is a universally hated technology. Customers perhaps do not realise that getting an IVR response is far better than no response at all, and the cost of enabling all calls to be answered by a live person at all times is grossly prohibitive. This does not change customers viewpoints – they seem to consistently be frustrated by the technology. With Video IVR, however, rather than listening to a series of options, the options can be displayed as a menu onto the customer’s handset making the use much more intuitive and faster. This has many benefits for customers – primarily speed and ease of use making the experience pleasurable rather than frustrating. For organisations, they can still deploy self-service technology to ease inbound transaction flows but can add neat new features, such as playing advertising or video-enabled comfort messages to waiting customers, or they can uplift adoption of self-service by making data entry easier than voice IVR and more like a web page for customers. The ability to mix simple messages with rich media creates numerous new service options, and no doubt, new revenue options for the future.

Copyright 2006 Robert A Innes

Common IT Phrases Translated

There are many indexes across the web to help you unravel the acronym-heavy world of IT. A good one is found at . These indexes, however, do not cover the really important code that IT people use to communicate with users. This "code" is designed to be:

oblique - difficult to interpret
non-commital - sounds like a firm date but actually it's variable
mildly derogotary - if you're not a geek you are slightly inferior

A perfect example of the code is when a user asks for a delivery date from an IT person. Anyone who has spent any time in an organisation of any kind will know that IT people do not do timescales and they most certainly do not do deliverables. They are creative and ingenious and need freedom from the boring work realities of the rest of us in order to build wonderful applications and systems. Consequently, when pressed, they will respond in the following manner:

User: I need to add a new checkbox to the online application. How long will that take?

Developer: Hmmm. That could be tricky. I'll need to look at it.

User: But it's just one checkbox?!

Developer: Yes, but the database will need to be updated and tested.

If you press, then more technical constraints will be come apparent and the acronyms will start flying!

User: It's just like the checkboxes that are already on the form. How difficult can it be.

Developer: Well the recordsets will all have to be updated and the XML feeds might have to be redeveloped.

User: How long will that take?

Developer: (Sucking in of air) Give me a couple of days.

User: What?!!

Developer: You want it right don't you?

User: (sighs) Yes, that'll be fine. I'll tell Marketing it'll be a couple of days.

One-nil to IT.

Generally, any timescale issued after the sucking of air sound is going to be longer than you hoped for. Timescales are padded by between 50% and 100% to allow the developer to press on with more urgent activities, such as building their own website. "That's pretty easy" means less than an hour to fix. "Pretty sure that can be done" means half a day. "You're kidding" means weeks!

This behaviour is quite easy to understand. IT are held accountable for the performance of systems (and usually, increasingly complex systems) and with so many technologies at work, often they genuinely do not know how long something will take until they start work on it. Similarly, existing systems may be the work of other developers, some of whom may no longer work at the organisation. Documentation may be sparse and one of the systems to be integrated with may be known to be in need of a re-write, but it never gets high enough on the IT agenda.

The best way to handle this complexity is not to fight it. That way leads to madness. Engagement is key and involving IT early in the process can significantly cut overall development time. The dreaded "brief creep" which is common in marketing and customer interaction systems needs to feed into realistic and reassessed project timescales.

Dealing with clients who do not appreciate the engineering work behind modern systems is difficult but you must persevere and work with IT to present back a menu of options. Many, many times "must-have" features can be reduced in complexity or removed altogether if managed with the resource-stick - "Yes, you can have multiple over-layed option boxes feeding into different remote databases, but it will take $1M and 2 years to complete. Or, you can have a lookup table that uses the existing remote database feeds and that will take $0.1M and 2 months to complete."

Splitting feature release into phases and managing phases based on customer acceptance is a good way to demonstrate success and get the client onto the road of progress, but building systems in a manageble way. There is a direct correlation between the number of objectives for a system and the development time. Plus, the more objectives, the higher the risk.

Copyright 2006 Robert A Innes

Give me some power Nikon

Nikon offer a superlative range of quality imaging and photographic products. The range extends from simple point-and-shoot digital cameras to ultra-expensive professional imaging systems. Nikon have managed very well to maintain premium positioning of their professional gear, whilst opening the market at the lower end, seemingly without canabalising sales of expensive products, which is difficult to do when targeting a wide range of users (with widely different spend patterns).

An object in possession seldom retains the same charm that it had in pursuit - Pliny the Younger

The new (at the time of writing) D50 is a digital SLR which accepts the full range of Nikon digital lenses but packages up the great optics, metering and image management features from the middle of their range, into an entry level digital SLR. Many of the complex features found in other models have been incorporated in automatic settings. This provides great results for novices.

Out of the box the Nikon D50 is superb and the customer experience is wonderful. You truly think you have purchased a great product. There is a great web resource that provides information and support to which all customers have access.

Whilst browsing support web resource, this customer notices that there is product recall on the battery used in the D50. A check of the batch number of the battery finds that the recent purchase is included in one of the batches covered by the recall. It is easy to log a support issue online and shortly after, an acknowledgement email is received with recall redemption instructions. These involve returning the potentially faulty battery to Nikon and then after approximately 7 to 10 days Nikon will send a replacement. Sorry? What? Send the battery back and do without power?

Surely the battery recall is an opportunity to demonstrate Nikon’s commitment to it’s customers, rather than insist that customers do without their system for effectively two weeks? This seems an odd way to thank customers for investing in the Nikon system.

Presumably the recall is organised in this way to prevent a deluge of requests for “spare” batteries based on spurious qualification criteria. Holding onto replacements until the customer returns an item that can be verified is a smart way of holding down false redemptions, however, it seems to have an accounting rather than marketing mentality driving it. Sending the message that Nikon considers its customers are not to be trusted is a dangerous step. What about using the network of retail partners that Nikon has up and down the country? Presumably retail partners would seek some compensation from Nikon for administering the recall, in which case it is understandable that Nikon is hesitant. Similarly, what about distance selling partners – how would their customers redeem?

Acknowledging the difficulties in managing the redemptions, it is still difficult to understand how Nikon settled on the current redemption regime. If the fault in the battery was down to customer issue, a remote solution with two week turnaround would be a reasonable response, but an issue due to manufacturing weaknesses where potentially faulty products are still being actively sold through the channel, insisting on a customer return first redemption model seems overly harsh and designed to frustrate customers.

Postscript

After complaining, a new battery arrived. No letter, just a battery. Still, I got what I wanted.

Copyright 2006 Robert A Innes

A Tale of Two Banks

Case 1: Morgan Stanley

Customer has their credit card refused and refers to the bank to see what went wrong. Morgan Stanley quickly establish that the customer has been, how can we say it? … stupid. No other word for it – having previously reported a lost card, the customer (it’s me, dear reader) received a new card from Morgan Stanley. Rather than use that for the transaction in question, the customer finds the old card and, forgetting about the new card, used the old card, which was quite rightly refused (as it had been cancelled). When the MS call centre advisor unravels this and explains to the customer, everything becomes clear. MS Advisor handles the call incredibly sensitively, and despite the obvious customer error, never makes the customer feel stupid and goes out of her way to help. Immediately a new (third) card is ordered.

This seems to be an excellent response despite the error being the responsibility of the customer. Well done Morgan Stanley.

Case 2: Smile – The online bank from The Co-operative Bank

Because it’s an online bank, Smile do not have high street branches and rely largely on The Post Office to provide services, such as depositing cheques.

Customer deposits cheque at Edinburgh Post Office and after four business days, cheque has not shown up in account. Customer phones Smile to seek some reassurance that the cheque is in the system and will show up in due course. However, on explaining request to Smile advisor, the advisor immediately goes on the defensive and abdicates responsibility for the process to The Post Office – accepting no blame for any delays. This is despite the customer pointing out that Smile had previously compensating this customer for losing a cheque in 2003. Despite this prior instance, advisor still insists problem can only be the cause of the Post Office. States that turnaround time could be eight to ten working days! After twelve minutes on the call advisor offers overdraft (not interest free) and when accepted, suddenly becomes customer friend, empathising and offering support. It’s all a little late but one gets the impression advisor’s hands are tied by process.

Customer writes to complain, particularly in regard to lack of ownership taken by Smile – who am I going to report to? The Post Office? What are they going to tell me? The alternative to paying in the cheque at the post office is to post the cheque to Smile. Somehow using the traditional postal service for letters is reliable, however, bizarrely, passing the cheque to a Post Office employee across the counter and receiving a receipt is less reliable? How could a smart organisation think up this kind of logic?

Response to complaint arrives in good time (less than one week) but merely restates terms and conditions rather than addressing the complaint. No comment is made on the eight to ten days offered by the advisor and no comment on the absence of ownership of the issue. This seems not to be a fair response to a reasonable question.

Next day, as it turns out, cheque clears. So much for eight to ten days, however, was there no possibility of providing visibility of the imminent clearance of the cheque and thus avoiding all of the problems?

Copyright 2006 Robert A Innes

10 Things to do once your Call Centre campaign is live

1. Talk to the advisors about what customers think. No-one speaks with more of your customers than the advisors.
2. Be visible in the call centre. Calling teams have plenty of managers, coaches and trainers to call on, but seeing the campaign manager in the call centre is a positive thing. Make yourself available. You are not above talking with call centre advisors. We're all people!
3. Listen to calls. Not just on go-live day, but throughout the campaign. This can be a real pain but persist. You will find out stuff you never dreamed about.
4. Work with the call centre manager and the team manager to reward the team on a formal basis (planned and published incentives) and also on an ad-hoc basis (surprise pizza lunch, for example).
5. Resist the temptation to correct advisors for grammer or informal language. Leave call centre operations well alone!
6. Do not hesitate to correct advisors if they are off message. Be firm but please, please, be polite. Arrange a training update if widespread to cover the learnings so far from the campaign. Resist the temptation to lecture to staff. This never works. They are your customer ambassadors so be smart when dealing with them.
7. Walk the process. Check that what you think happens actually happens.
8. Check information capture for information formats. Surprising how often system glitches or briefing misunderstandings can create unexpected data values.
9. Measure your key campaign metrics and do not be afraid to pull the campaign if it is not working.
10. Make sure you can compare responses with other campaigns, with integrity. No point counting calls presented on one campaign and comparing it to calls handled on another. Try to leave operational MI to the call centre and concentrate on campaign MU, for example, look at cost to convert rather than average handling time.

Copyright 2006 Robert A Innes

10 Things to do before your Responsive Advertising hits the Call Centre

1. Share your media plan with the call centre! You'd be surprised how often call centres take calls for campaigns they were not expecting
2. Prepare a detailed briefing for the call centre. Include samples of the media used, the objective, the customer process, etc.
3. Decide on metrics. No use trying to figure our your MI once the campaign has finished.
4. Check all of your customer dialing numbers. Do this personally and do it twice.
5. Decide what information is really, really necessary. Remember, someone has to capture all of this information and the handling time for the call likely comes out of someone else's budget, so they may not be motivated to support long call durations
6. Role play the script or call guide with a colleague. Does it sound forced? Does it deliver a positive customer experience?
7. Make sure the post-call process is slick. Advisors will be making commitments to customers based on how you brief them - make sure this is accurate and that commitments actually happen.
8. Talk to the team who will be managing your call - too often they will feel like the twelth person on the team, but including them early might smooth go-live issues.
9. Understand the point at which you will pull the process if it's not working. Do not be afraid to kill a project that doesn't work. If the first 200 calls are bad, don't cross your fingers that it will get better - it won't. Something is clearly fundamentally wrong with the campaign.
10. Make sure you test your initiative before rolling out. Don't go large with a gut feel!

Copyright 2006 Robert A Innes