B2B Customer Intelligence - Flexibility

This is the second post in a new Customer Intelligence blog series. Please add comments if you’d like me to expand on any of the points, or have suggestions for other topics. I also welcome any feedback on the content.

Customer Intelligence takes a lot of work. To implement it properly you or your company will need to invest time and resources on technology, data quality and ETL projects. You will also need to hire or contract with analysts who will help to derive the insights needed to properly target your best customers and prospects.

As soon as you’ve invested all of that money, the target will change. Sales goals are redefined yearly at most companies, and even the most stable change companies their approach every couple of years. The Customer Intelligence program you’ve defined needs to flex with your corporate goals to be effective. Rather than build a rigid model that’s engrained into your production systems, a flexible model will allow for longer term success and save money.

Strategic Flexibility

From your first brainstorming session on Customer Intelligence, you need to think about how the program can meet multiple corporate goals. Let’s say that today your company is dedicated to growing it’s customer base through prospect acquisition. If you’re successful at that approach, your company will likely mature into a customer retention focus over time. Seems logical, right? So from the start how do you create prospect scoring approach that will also translate into customer lifetime value models. Which factors are shared in determining prospect conversion and customer longevity? If you were the CEO, which direction would you take the company in next? Now ask others in the brainstorming session what they would do. Set 3 or 4 targets and be sure that your customer intelligence plan can support all of them with little adaptation.

Insourced vs. Outsourced Analysts

One of the first decisions you will need to make is who is going to do the work. I assume if you are reading this that you’ve got a solid background in data quality, project management, and a few other relevant skills that will vary from reader to reader. Are you a PHD statistician? I’m not. Are you willing to spend 80 hours next week writing ETL code? Having done that for a few years (anyone else remember clicking “run” at 2AM on a 4 hour denormalization package, then cuddling up under their desk for a nap?) I’m happy to hand that work to someone else. You will need support, and it will be expensive.

Depending on the size of your Customer Intelligence program and the size of your company, you will need to decide on the best approach. Personally, I’ve used both internally staffed and external analyst firms and I’ve had success and failure with both. I’ve found the key is to have more than one modeler working on the problem to share ideas and help you keep track of the other. There’s no bigger wake-up call than hearing from a statistician that their peer is incapable of doing the work assigned. This stuff is really hard. Not everyone who has a good resume can build a model from scratch that drives your business to the next level.

To stay flexible, you need to be able to grow your team quickly during development and then shed excess analysts while Customer Intelligence programs prove their success. I would recommend a hybrid aprpoach with some full-time employees to maintain models and external consultants available when new models are needed or when your internal folks hit roadblocks.

Developing a Library of Models

Another key to Customer Intelligence flexibility is to build your models on top of a standard data set that can be used over and over again. There are nuances to using corporate data. It’s time consuming to build a dataset of input variables for a statistical model. There are issues based on business definition, geography, time calculations, and data sparsity that all require both business analysis and technical work to get right. By building a single set of denormalized data (think one big table where the transactional keys have been resolved to their text values in the data set) you will save time and money the next time you try to build a new model. Of course you can add variables over time, and even include different versions of the same values. But by building it once and using it repeatedly you will get more results with less effort.

B2B Customer Intelligence Series - Introduction

My job title and primary focus is Data Governance, however the data I spend the most time managing is B2B data. As a former consultant, I constantly find myself suggesting new applications to drive Sales and Marketing programs. Sometimes it’s as basic as dipping into database reserves to find new prospects, but often my extracurricular projects are pure Customer Intelligence, also referred to as CI. Customer Intelligence can cover a lot of ground, so let me define it.

Customer Intelligence is the intersection of Sales, Marketing, and Analytics that helps present the best customers and prospects to inform company strategy and tactical approach.

Customer Intelligence work manifests itself as Analytic projects including predictive modeling, cross-functional leads programs, and Sales and Marketing strategy projects. In most companies I’ve worked with, Customer Intelligence is distributed across several functions and that group collectively defines the companies Customer Relationship approach. In others a central group focuses on Customer Intelligence and coordinates the distribution of related information to drive strategy.

Other companies have no Customer Intelligence. They are Customer Ignorant. In that case, the Sales and Marketing teams approach customers based on generic approaches and anecdotal history of which customers are the “best”. With good products and excellent customer service, this approach can work. However, over time it will open the door for competitors to take over and dominate the market.

Here are a few thoughts for the upcoming series. The plan is for each of these topics to be developed as an entire post over the next few weeks. If you have other suggestions, please comment on the post or tweet a reply to @markgoloboy.

1) Customer Intelligence Flexibility

No matter how successful it is, every sales strategy will be retired someday. It may be next week. It may be next year. But be sure, the target will be set somewhere other than it is today. How do you build a Customer Intelligence strategy that allows your organization to stay agile?

2) Aligning Against the Opportunity

Once properly implemented your Customer Intelligence approach should become an integral part of the sales reps’ day. How do you integrate the information with your existing processes and systems to align against the opportunity?

3) Direct Marketing Impact on Sales

Customer Intelligence can help define a set of high value prospects that deliver return on Marketing investment. Determining which factors influence purchasing requires advanced analytics. What’s the best way to develop analytic models and measure whether they predict a prospect’s likelihood to become a customer?

4) Defining Lifetime Value

Determining Lifetime Value (LTV) of a customer and potential LTV of a prospect can align Sales, Marketing and Product goals with corporate strategy. However, LTV analysis requires that your product and sales data has been collected consistently for your company’s history. How do you develop and use LTV to drive product direction and Sales and Marketing focus?

CRM Marketing Strategy To Drive Sales Revenue

For the purpose of this post, I’m defining Marketing Strategy to mean the analytics that complement the work done by the Database / CRM Marketing team and the influence that those analytics can have on Sales and Marketing programs. This article isn’t about Brand, Creative, or Media Purchase. It’s also not about Social Media, SEM or SEO. Just to be clear, I’m writing about using Data, Technology, & Analytics to reach out to customers and prospects, improve program results and drive revenue. Generally my recommendations are specific to Email Marketing, but may also be applied to Direct Mail or TeleMarketing.

Marketing Strategy should be the heading under which your company defines it’s analytically based sales programs. But you don’t need a fully developed internal analytics shop to be strategic. In fact, that isn’t the first step at all. First, focus on your data. It doesn’t matter how good your analysts are, they won’t be effective or efficient if you give them low quality data to work with.

So, let’s begin with the data. The following three areas of Marketing Strategy rely on Data Governance, so for my opinions there read up on Lightweight Data Governance.

Opportunity Analysis

If your data isn’t clean, you can’t get a feel for how well you are doing. For instance, duplicates in your data when analyzed in aggregate cause double-counting within buckets, double-counting across buckets, and general noise that diminishes the value of analytics and modeling. If you have the same company listed as both a customer and a prospect, and they both fall in the same industry or geography, your analytics driven programs will fail.

The solution is to develop a source system down approach to gathering data for your Marketing Strategy work. If you collect the data with reporting and analysis in mind, downstream processes will benefit. So work with your Sales, Service and Finance teams. It will take them an extra few seconds to correct the problem. Every customer facing employee should spend a few seconds each time they speak with a customer to ask, “Is our contact information for you still correct?” That change, which may require a culture shift that starts with the executive team, will dramatically improve program results.

Targeting and Personalization

Do you use Marketing Strategy and leverage data to drive campaign results? Test and Learn is a sneaky term for Marketing Strategy. It sounds less expensive, so use it.

Your content should be specific to your target’s business problem. It should be tailored to their purchase history, or lack thereof, and include next (or first) best product or cross-sell specific messaging. You should personalize the email to the targets name, title, company name, and other fields in your CRM system. If you have analyzed the lift generated by personalized vs. non-personalized messaging, you understand the need to target your emails. 

When you segment targets and personalize the subject line and body content of your emails, your open and click-through rates will increase significantly. Also, your opt-out rates will go down. The quality of these key fields must be monitored and improved to take advantage of that lift.

Compliance and Customer Interaction

If you are not familiar with CAN-SPAM, read my take on your responsibilities as an email marketer. You are required by law to do a great job at managing email opt-outs and confirming that your business is transparently represented in your email marketing. Even forgetting about email compliance, you should still consider the customer impact of bad marketing. “Mark Goloboy or Current Resident” = Trash Can. Trash Can = Poor return on direct mail spend. Read: Not enough revenue to justify your program.

So how do you mature your processes quickly to take advantage of the potential sales impact of Marketing Strategy? You must develop quality processes that remove bounced emails from your lists and opt-out all instances of an email address when requested. You also need to be sure your sales reps are honoring your opt-outs. Lastly, you must review your data for patterns of Sales or Service rep entered garbage. What’s the most frequently occurring word in the First Name field of your contact records? I bet it’s not “John.”

Data Acquisition

Whether you are in B2C or B2B Marketing, there is value in purchasing reference data for your existing customers, and lead data for prospecting. How much value is there? You need to test purchases from different vendors to determine that. There’s no other way. Buy a small list. Or better, ask a list vendor for a free list to test. They aren’t selling a lot now. Take advantage! Then predict ROI by scaling the results and associated spend across your whole universe of customer and prospect data.

You can also measure the impact of cleaning, de-duplicating and referencing your data to a standard set by testing those processes against a statistically significant sample of your customer data. Again, you may be able to get some of these services for free by getting into a competitive sales situation. Your prospects ask you for this, right? Mastering this customer data requires assistance and cooperation from Sales, Marketing, Customer Service and Technology, which means Data Governance.

Lightweight Data Governance: A Starting Point

This expands on the previous article, Lightweight Data Governance. I’ll continue to add to the theory in upcoming posts. If there are any areas you would like me to focus on, please add a comment, or email me directly.

A few weeks back I met with Steve Sarsfield to discuss the upcoming MIT Information Quality Symposium (MITIQS). It will be my first time presenting to a Data Quality focused group, so I was excited when Steve offered to provide some background. My main concern was, “How can someone in the commercial space keep the interest of a combined business, government and research focused community?” We discussed my approach, and I think I’m on the right track. I’m going to describe how we initiated Data Governance at my company, kept it simple, and found early success.

So where did we start? Data Governance grew from an expressed need by the executive team for better data quality. Sounds simple right? Fix the data. It’s like the Kenan Thompson SNL character talking about the economy: Fix It. The company decided that Data Governance was needed, and that they would let me define the path to getting there. I set the scope to include any project where I have an opportunity to build credibility in data or reporting. I’ve formalized processes where necessary, but kept it “lightweight” in most areas. With the current state of the economy, I see no other way to get there.

I previously led the Marketing Analytics department, and we had responsibility for B2B and B2C Analytics. Most of our efforts were focused on the B2B side, since that’s where the most perceived opportunity existed. When I moved into the Director of Global Data Governance role, I built from my strength and worked on B2B issues first. I attacked the low-effort, high-value projects. I looked to expand on the local efforts that were working well. If teams or projects came up with creative solutions, I looked to expand their work globally. My thought was that it’s really hard to come up with the underlying process definition, but that an existing process was easy to expand. It doesn’t work for every existing process, but some are natural fits that resolve longstanding internal issues.

That became the basis for Lightweight Data Governance. Find the projects or efforts that are successful on a small scale, and expand them globally. That way you start with a base of knowledge, documentation, and executive support that’s very hard to build from scratch.

Grow Data Governance efforts organically

Start with existing processes. Find out which can be expanded, centralized or automated.

Focus on project level ROI

Don’t try to sell your management on a huge program to start. Build the business case at the project level. It’s easier for management to support small positive ROI projects.

Partner to be unobtrusive to ongoing work

Find projects that are already in flight. Would Data Governance add to their impact? If so, partner with their leadership to help craft the deliverables to create mutual benefit.

Build momentum from early successes

Get testimonials! If the project went well and the community benefited, you should be able to get the project sponsor to say so.

Measure initiatives on DQ impact

This step is further along the Data Governance continuum. Begin to show the impact on the organization when projects focus on data quality. This cultural shift will underscore the importance of future Data Governance work.

Follow with Formal Data Governance

Does it make sense for the enterprise? Does executive support exist? If not how do you build it? This is where the more traditional theory in most Data Governance efforts becomes relevant.

B2B vs. B2C Matching for Sales and Marketing

I recently read the KnowledgeBanks article Why is b2b marketing different from b2c marketing? The article works to disprove the common misperception that “B2B marketing is just marketing to consumers who happen to have a corporation to pay for what they buy.” I completely agree, and would like to extend the points made to include the differences in matching, sales, and marketing. I’ll also point out ways to know if you’ve reached your goals.

Some background before I begin. I spent two years working for Harte-Hanks implementing primarily B2C matching for Financial Services Marketing systems. Did you ever wonder how the banks knew that your accounts were linked even though you opened them as different names and addresses? Think about Bank of America’s history. It’s a collection of dozens of banks, and you could have opened accounts any time in the past. To merge that data which could include 10-100 million rows of accoutns, massive B2C matching systems are required, and equally complex logic. My work at Monster has centered on B2B data and how it’s used by Sales and Marketing teams. I’ve worked primarily with matching engines based on Trillium software, but I’ve gotten to know most of the other technologies used at an Enterprise level over the years. I’m a free-agent when it comes to technology, and I’d recommend all technologists embrace an open mind when it comes to vendor selection.

B2C Matching

B2C matching is absolutely nothing like B2B matching. The difference? Householding. In B2C the goal is to household different contacts at the same address. You household John’s accounts, then you household Mary’s accounts, then you merge them both together if they live at the same address. Sounds simple right? Well, sometimes John goes by Jack, and sometimes Mary uses her maiden name, and their phone company and credit card records are all listed as “M & J.” These are the troublesome records to figure out, and most of the records have some flavor of variation similar to this.

B2C Sales and Marketing relies on volume, and the companies focused on matching process huge data sets. You need to find many interested parties because the available money from each is very small. A person may buy $100 worth of software, a $2000 computer, or even a $20,000 car. But they will never spend $1MM with you unless you sell houses to the moderately rich or toys to the ultra rich.

B2B Matching

B2B matching is also about householding, but we don’t call it that. We call it Parent-Child Hierarchies. The goal is to determine all of the locations of a business that you are selling to, and try to figure out how they fit together. Of course this is an enterprise perspective, and SMB would be more focused on single locations. So the enterprise question is, which locations are headquarters of other branches? Does that headquarters control purchasing for the child branches? Or are the branches empowered to buy on their own? What does the sales history tell us about them? Do the reps know anything that can help, and how do we capture that data in an automated matching process. All that, and I haven’t mentioned that companies buy, sell, and merge all the time. Think about GE, Berkshire Hathaway and Tyco. Is each of those 1 business or 20?

For B2B Matching each sales and marketing person would like to know how much each location has purchased and which has the purchasing power. Some of those locations will purchase services that could result in multi-million dollar deals. When I worked for Accenture the philosophy shifted to “Big Bets” and the partners (who functioned as sales people) only targeted accounts willing to commit to $25MM per year. That year they sold several Billion dollar deals. Think about that for a minute. The sales reps will need appropriate level high quality contacts and contact information at each location. You can’t sell a $25MM deal to a line manager or team lead. There are many ways to get B2B contacts - list purchase, telemarketing, partnerships, etc. - but that’s a different post.

Measuring Results

If you think you’ve reached your goal and found success with your matching, you’re wrong! Matching is more of an art than a science, and as soon as you get to an acceptable level of completeness and accuracy you need to start looking for the next round of improvements. Matching (like data warehousing in general) must change with the business. As new products are developed, new technologies released, and new business processes implemented, the matching must be updated to dovetail with those changes.

To Address in Future Posts…

Preferences and Opt-out Management
B2B vs. B2C Analytics
Number of Services per Contact (B2C) vs.
Number of locations per Company (B2B)
Demographic and Firmagraphic Appends