Our SFMC implementation failed

This is unfortunately not an unheard statement, which I get from some organisations who tried to implement Salesforce Marketing Cloud.

How do one define a failure in this context?

It is first of all, a question of not getting the expected value from the significant investment, which Salesforce Marketing Cloud is.

Definition of failed SFMC implementation

While this is the overall issue, it is derived from few underlying problems, like these below:

1. We mainly send weekly newsletters to our entire base, with little personalization and no segmentation.
2. We struggle to migrate all our existing features from legacy systems, and hence are paying for two solutions running concurrently.
3. We can’t measure the commercial uplift derived from switching to SFMC

These issues are not uncommon, and can be addressed. But addressing them before you start on the implementation, will provide you with a much smoother process and greatly increase the probability of success.

Pick your strategy

What is your commercial goal of investing in Marketing Cloud? And how do you want to get there? Having a CRM strategy in place, makes it much easier to scope and plan an implementation. Start with your customers and the experience you want to provide them:

• what type of campaigns will you be building on SFMC?
• map your always on journeys (welcome, winback, birthday, etc)
• identify your channels (email, sms, push, landing pages)

Once you have done this, you must define what criteria you will be using to segment and personalize every single piece of content (e.g. email) within your list above. This will create your data shopping list:

• birthday journey? We will need a date of birth for our customers.
• app push? Can you identify your customer in your app? If not, all the messages will be generic
• winback? you will need engagement data, that will most probably be their purchase history
• and so on…

Each of these items probably originates in different systems, hence it will be a good indicator on what teams you will need to collaborate with, in order to bring this data into Marketing Cloud. This should also be your main input for building a data model in Marketing Cloud.

And, when speaking of data model and customer data – how do you define your customer? Selecting the right identifier for a customer is one of your most important decisions. It is so important, that I have created a post just on that topic.

But hey, now it starts becoming complex. Are you sure you want to continue? I promise, it will only get worse. Unless you have existing experience with Salesforce Marketing Cloud, to a degree where you are familiar with SQL, APIs, data modelling, Ampscript, you should probably ask for help. But more on your options later.

You have your strategy, let’s make a plan

Your strategy should define the full foreseeable scope of your implementation, while your plan will break it down into manageable components.

You want each of these components to be able to provide some value to your business along the way. This could be by defining them as milestones:

1. All data extensions created and mapped in the contact model
2. All customer data in Marketing Cloud as a one-off import, to validate feasibility of the data model
3. Inbound integrations up and running for customer data
4. Inbound integrations up and running for transactional data
5. Email templates and content blocks created as per brand guidelines
6. Outbound integrations for bounces and unsubscribes to customer/consent system of record
7. Segmentation rules defined and implemented in Automation Studio
8. Welcome journey running
9. First ad-hoc campaign sent
10. Inbound integrations for transactional emails
11. Adding SMS to transactional notifications

Above is just an example, but illustrates realistic items you can adapt to fit your own strategy.

Measure your success

Before starting your implementation or migration, define what commercial success factors are critical to you. If you are migrating from an existing solution, make sure you are tracking these KPIs already, as you will otherwise don’t have a baseline with which you can compare your upcoming implementation.

Who you gonna call?

There is nothing wrong with doing your own implementation, provided you have significant familiarity with Marketing Cloud and all it’s components. Some degree of knowledge when it comes to enterprise architecture is also helpful – especially when designing your integration landscape.

If you feel a need for help, you can connect with your local Salesforce Implementation Partner, who has the rights experience to help you. Do keep in mind, that Salesforce experience doesn’t always imply Marketing Cloud experience. Sales/Service Cloud and Marketing Cloud are two very different technologies, suited for very different business needs, and being an expert in one, does not necessary make you an expert in the other.

Here is some good material on Salesforce Partners, and how to find one. This Trailhead Trail also has some more details.

When you find a partner, or maybe a freelance resource, it is always a good idea to validate their skills by checking their certifications in this tool. After provide a name or email address of your candidate, you are being shown the certifications currently held by the individual. When selecting a Marketing Cloud resource, you should aim for one having at least three certifications relating to Marketing Cloud, Salesforce Marketing Cloud Consultant being one of them.

You could also engage with Salesforce Professional Services, who has very senior resources, with numerous complex implementations behind them. They are the real SMEs, and being close to the product teams, they are the first ones to get updates on product releases, roadmaps, etc. Additionally, they have access to Salesforce internal tools, making debugging and other task much more efficient than other consultants. They might not be your cheapest option, but they are indeed some of the very best out there.

Good luck with your project!