🚀 We're hiring, check out our careers page to join our team!

Customer data model: How the customer stack simplifies everything

Bewildered by 5000+ Martech tools? Unsure of best practices for using marketing technology? Questioning how you use dozens of separate tools together effectively? This customer data management guide is for you.

First step. Stop calling it a “Marketing Stack. Focusing on technology doesn’t inform you of gaps (any salesperson will tell you you’re missing something) and what to do next. This is a key mindset change.

Focus on your customer instead.

Customer data management is NOT optional.

Your tools, teams, and data should focus around a customer, who they are and what they’ve done. Your operations and “stack” should be customer centric, a tie together your data and interactions across their entire lifecycle.

If we draw out the key “moments” with customer data, we find a circle. A closed loop.

  1. Sending messages
  2. Tracking reactions
  3. Recording to profiles
  4. Deciding actions

… to send more messages. With this closed loop, you can scale, automate and personalize your sales and marketing.

But how do you assemble and orchestrate all this customer data at your company?

Customer Stack


First, figure out your messaging strategy.

Before you commit to what tools to buy for your customer stack, you need to consider what channels you can operate in.

Yes, you can research where your audience is, do keyword research, social profiling, buy data lists etc. around your ideal customer profile. Of course this is important.

But, you’ve also got to consider the operational impact of managing many channels. A single cake shop wouldn’t open one hundred new stores overnight.

Businesses have operations. Marketing needs marketing operations

Yawn...

"Operations". Seriously?

You had to take something interesting like Marketing tech and make it. Totally. Boring...

Well, when you've expended thousands on tools you don't use, campaigns that don't perform and teams that are fighting amongst themselves, you'll come back to answer all these following questions.

If you get your marketing operations plan right, it plots a clear roadmap to how you can scale and automate highly personalized sales and marketing. Oooo ah! Got you back? Let's continue...

Before diving in everywhere, it’s important to understand why you might not want to be everywhere that your ideal customers are. There are three things to consider:

1. Is it a high-touch or low-touch channel?

How much human involvement is needed in sending each message? How much can be automated? Can it be fully automated?

High-touch enables more human care, creativity and responsiveness. But, this comes at the cost of hours of work billed and growing headcount in your organization.

2. Is it a broadcast message or a conversation?

Is it sent to many people at once, or to an individual? Can people reply to the sender personally?

Personal conversations have more chance to be relevant and motivate people to take action. But, this comes at the cost of crafting, optimizing and measuring many more messages.

3. Is it hard or easy to get wide reach?

Is it ten times the work to reach ten people? What if it’s 10,000 people? And they’re spread all over the world?

The best converting tactics might not be scalable, so you might be worse off pursuing them.


If we take each of these questions and lay them over each channel, we start to see the operational impact of trying to maintain them.

Medium Human involvement? Conversation type? Ease of reach?
In person meeting Hi-touch Personal Low
Website Low-touch Broadcast High
Live Chat Hi-touch Personal & Broadcast Moderate
Social Media Hi-touch Personal & Broadcast High
Ads Low-touch Broadcast High
Email Low-touch Personal & Broadcast Moderate
Phone call Hi-touch Personal Moderate
Mail Low-touch Personal & Broadcast Moderate

Now, we see the trade off…

We want to maximize our sales and marketing effort and reach as many relevant people in as many relevant places as possible… but we don’t have the resources to do that.

As a marketer, have you ever felt you’re spread too thin? That you’re doing too much? That you’re losing focus?

In the ideal world, we'd be able to combine the best of all of these:

  • Low-touch (with the power of hi-touch)
  • Personal (with the ease of broadcasting)
  • High reach (without an extraordinary amount of resources)

If we can do all this, we get way more bang for our buck. It’s operationally more efficient, means we can reach more people, and it leads to better, more effective marketing too.

The answer to all of this is personalization. Crafting emails, ads, webpages, phone calls and more to be sent to an individual - for an audience of one.

Creating your personalization strategy means answering five questions:

  • Who are you targeting?
  • What message are sending them?
  • Where are you messaging them?
  • When are you messaging them?
  • Why are you messaging them?

The key takeaway? To scale and automate your messaging, you don’t just need to think of which channels you operate in. You also need to consider how your customer data will flow so you can effectively personalize your sales and marketing messages.

Customer data management is the answer.

To create and send highly personal messages, we need to think about how our customer data flows: messaging, tracking, profiling and actions.

Let’s start with tracking.

The trouble with tracking

Tracking is about the reactions to a message. Things like…

  • Viewed a web page.
  • Submitted a form.
  • Answered a phone call.
  • Subscribed as a new customer.

This data gives us the context from which to talk to someone. Are they a customer? If not, have they requested a demo or looked at our pricing page? Do we have their email address? What is their job title?

All of these simple for the user perspective, but they cause three problems for data-driven marketers and sales team to track.

The first problem with tracking customer data is it’s needed in multiple places. As marketers, we’re familiar with tools like Google Analytics, but “reactions” are also tracked elsewhere.

  • Website tracking tools (client-side)
  • Databases (server-side)
  • Other tools (like emails, surveys, ads)
  • Offline (phone, events etc.)

Second, tracked customer data is often in multiple types and formats which aren’t all consistent.

  • UTM parameters - tracking external clicks to your website
  • Analytics events - tracking actions on your page, backend
  • Databases tracking other interactions - searches, purchases, logins etc.

And that’s just for the web - not thinking about the different tools and systems like email and phone calls.

Finally, tracked data is often inconsistent. Too often, tracking plans come with different, cryptic naming conventions, so it’s hard to quickly see what’s going on.

It's hard for tracking not to be done in multiple places with multiple tools. So long as the reactions are recorded neatly and you have a reliable stream of data, we can combine all your customer data together later.

In other words, don’t deal with the tracking when it’s tracked. Deal with it later on.

Which brings us into...

Profiles! Profiles everywhere!

All this tracked customer data needs to go somewhere. This is where customer profiles and customer profiling come in - storing everything about someone and what they’ve done.

Contact profiles will feature details like a name, email address, phone number, job title and so on. Usually, you’ll find these amongst your:

  • CRMs (like Salesforce)
  • Marketing automation and email tools
  • Databases
  • Data enrichment tools

The first problem is all these different tools (which are tracking different things) have different different data.

Siloed tools = siloed data = siloed teams = chaos, politics and inefficiency. Your customer profiles needs to be integrated together.

But, that’s only the first challenge to customer profile management…

Lots of data is about unidentified, anonymous people! For instance, Google Analytics tells you about actions on your site, but it doesn’t tell you who did them.

This anonymous lead and customer data is often full of helpful, often behavioral insights like the pages they looked at before signing up. Maybe they looked at your enterprise pricing? Who knows?

Alas, anonymous data is hard to connect to your other data (like a CRM profile for our sales team).

The most common way to combine and resolve customer profiles is with hidden fields on a form submission. Slipping a little extra data through with each form submission. When you click that submit button, more information goes through than what you typed in.

For instance, when a user with the anonymous user ID 12345 submits a demo request with their email address through a form, you can set it up to pass their anonymous user ID 12345 at the same time, so the two can be associated together.

Which brings us back to the first challenge of customer data management - we’ve still got data siloed across many different tools. Siloed tools = siloed data = siloed teams.

Most companies use these seven customer data integration methods, each with their pros and cons, to build one unified view of each customer.

Whatever method you use, the goal is to have all your customer data tracked in many tools in many different ways be unified around a common customer profile, and all the tools being in sync, and every tool having everything about each person. Hoorah!

And so every tool has all the all the context to send a personal, relevant message. It’s as if you’ve given EVERY team “x-ray vision” to precisely target and craft the optimal message to send to each individual person.

Now, we just need to decide what we do with all that customer data. Which leads us to…

Automate ALL the actions!

With all this data within a customer profile, it's time to decide what actions are to be taken. From the personalization framework we have five questions to answer - who gets what message, when, where, and why?

Remember, we started with the goal being to scale and automate messaging. Most messaging might start out as pretty manual, like pitching a startup idea on stage, writing a blog post or emailing your friends and family.

But as you scale, mature and find things that work, you codify the messages and logic that works into processes like call scripts, templates, and “who does what” into products like CRMs and marketing automation tools.

From the five questions in the personalization strategy framework, four of them come from grouping people together - segmentation powers the who, the when, the where, and the why.

To automate actions, you need to create and manage effective segments. This enables you to target the right people, at the right time, in the right channel, at each stage of the customer lifecycle. Leads, customers, email lists, ad audiences - they’re all segments.

What message gets sent is content - the last piece of the customer stack framework. As your messaging matures and your audience grows, not everyone should get the same message. Your sales team will adapt calls and emails for different personas and companies. This is where dynamic content and templates prove their value.

Templates and scripts enable you to “fill in the blanks” with dynamic content so you can to personalize messages based on names, job titles, locations, and so on. Isn’t it nice to be treated like an individual person again, {firstName}?

Both segmentation and dynamic content in templates are built from data in your profiles. So it’s the data in your profiles that enables you to personalize, scale, and automate your customer communications. This is why it's important to aggregate all your customer data together and sync it between all your tools.

… and back to messaging

We now know how we can personalize, scale and automate our messages!

The way we can do this is with segmentation and templates, driven by unified customer profiles, not by hand. By pulling all our tracked data and different profiles into one place, we have all the context to do this.

Now how do you make that work? Remember, customer data management is NOT optional.

Hire your tools to do a job!

At Hull, we’re a big fan of the jobs-to-be-done framework. In the same way we can define problems and roles to hire people for, we want to define problems and roles to hire tools for.

With our customer stack framework, we’ve clear roles to fill. Some tools might be able to fill many roles at once.

  • Messaging across the channels you want to target
  • Tracking across all those channels
  • Profiles from all that tracked data
  • Actions to decide what messages to send next

Let’s look at some common tools and see how they help our customer data management strategy.

Examples of Common Tools In a Customer Stack

Buffer

Social scheduling tool

  • Messaging: Many social social channels
  • Tracking: Nothing tied to people. Tracking is per message, not per person
  • Profiles: None.
  • Actions: Manually scheduling messages.

Social messaging is for broadcasting, not personal one-to-one. Buffer misses the user-centric view. You can’t schedule messages to reach specific individual people.

Whilst you can tie campaign data to the URL, so you know if a Buffered message was responsible for a conversion, it’s still a closed loop.


Unbounce

The landing page builder.

  • Messaging: Landing pages, with templates
  • Tracking: Form fills, page views and events.
  • Profiles: Data from form fill only
  • Actions: Manual form fill only. Some dynamic content possible.

Unbounce captures valuable lead data, but keeps it siloed away until it’s send and combined with data from elsewhere. It can’t trigger another message like an email or a sales call until this happens.


HubSpot

Marketing automation tool and CRM

  • Messaging: Web pages, social, email and CRM for sales people.
  • Tracking: Page views, events, form fills and email engagement tied to contacts for identified contacts
  • Profiles: HubSpot Contacts
  • Actions: Lists and Workflows for sales and marketing.

HubSpot has a closed loop - success! Messages tie to tracking tie to profiles to actions which delivers more messages. This is the key benefit of marketing automation platforms and CRMs which have profiles of identified people - and this is why these “closed loop” tools become core to your sales and marketing team.

The challenge comes from associating other customer data with this “source of truth”.

For instance, anonymous user data (since every contact is tied to an email address, HubSpot only tracks a profile from the first time they filled in their email address).

HubSpot users also need to aggregate marketing activities from other channels like ads, live chat, and offline. Most companies tend to have other tools with duplicate functionality too.

Finally, though HubSpot has a closed loop, it only has a closed loop in the pre-customer stage of the lifecycle, and needs to integrate with sales, support, success, and product tools.

HubSpot needs to be connected to your other tools too.

Hull - ETL Diagrams - All-in-one CRMs and Platforms

Now evaluate your tools, spot missing roles too fill, and tie it all together!

From the examples above, it’s quick to spot the gaps missing in each channel and loop.

What about your company? What tools do you use? What channels are you targeting? How are you tracking all the reactions? How are you profiling all that data and centralising it? How are you creating segments to trigger action and “fill-in-the-blanks” in your messaging templates to make dynamic content?

(Pssst. If you’re looking for one easy template to work through this, grab our complete Personalization Pack on the bottom here!)

To recap.

Stop calling it a "Marketing Stack". Instead, think of the customer, the entire customer lifecycle, and the closed loop between interactions and your customer data.

  1. Messaging needs to reach your audience through as many channels given how much human involved is needed, whether it’s broadcast or personal, and how easy it is to drive high reach. To personalize, scale, and automate this, you need to close the loop with the other elements, including…

  2. Tracking needs to combine identified and anonymous data from multiples places, in multiple formats in a consistent way. This data should all feed into…

  3. Profiles need to combine all the data around each customer into one place to cover all the context needed to deliver the most relevant, personal message. Using all that data you can then trigger…

  4. Actions with rules to group people together into segments and insert dynamic content into message templates. Share these with your tools to deliver highly personal messages at scale.

Remember, you need to tie all your tools together to make this work. Customer data management is NOT optional. Not sure if your customer stack up to the job? Worried it’s a FraNkenSTack? 😱 Take the free assessment to get your grade and advice.

frankenstack

Ed fry
Ed Fry

'Ed of Growth at Hull, working on all things content, acquisition & conversion. Conference speaker, flight hacker, prev. employee #1 at inbound.org (acq. HubSpot).

If you've questions or ideas, I'd love to geek out together on Twitter or LinkedIn. 👇