What do we even do?

If you’ve ever worked with a CRM team, or heard about one somewhere in the depths of a campaign meeting, you might still be wondering ‘what do we actually do?’  

For starters, we’re not creatives in the ad agency, traditional sense. We’re also not data scientists or IT. But somehow, we touch all parts, and none of it at the same time.  

Young,Crazy,Bearded,Man,Confused,Expression.,Copy,Space,Concept

After a few years of working in the world of CRM, across both brand and strategy sides, I’d like to think that I’ve seen a lot. The processes don’t change much, the platforms might, the politics does. But at its core the job is about connecting the dots between tools, people and plans to make marketing happen.  

It all starts with what I like to call the CRM lifecycle. * 

*There is a glossary below for any funny looking terms or acronyms that have you scratching your head. 

Step 1 – The Brief

Mixture of concrete and vague ideas with missing ingredients 

 

Every campaign is supposed to start with a brief. Ideally it tells us: 

  • What the message is  
  • Who it’s for  
  • What the goal is (conversion, retention, engagement)?  
  • Which customer channel should we use, and do we have the right data to accurately segment, personalise and target?  
  • Where the data model lives 
  • When it needs to go out  

What we actually receive varies wildly. Sometimes we’re briefed on audience segments that don’t yet exist. Sometimes we’re told to use a predictive model we’ve never been given access to. Other times, we are asked to run a campaign that overlaps with others already in progress.   

It’s like trying to build Ikea furniture with missing parts and no instructions. 

Still, the deadline remains firm, and with limited hours and many urgent requests, we simply can’t say yes to everything, no matter how much we’d like to.  

Step 2 – The Build

The CRM maze of access and SQL 

 

The next step is to actually build the campaign.   

Typically, we would use a marketing automation platform to build the campaign in. This usually involves: 

  • Pulling or building the right audience segment (via SQL or visual builders) and often the exact customer segment doesn’t exist in the tools 
  • Connecting to the right data sources  
  • Applying pressure rules and frequency caps  
  • Building the campaign flow: entry logic, exclusions and variations  
  • Testing message logic (dynamic content conditional rules etc.)  
  • QA’ing the full customer journey  

 Sounds pretty straightforward right? One of the biggest challenges that happens early on in this process is access.  

We often don’t have access to the data we’ve been briefed on using. The model might sit in a completely different environment, locked behind permissions we don’t have. Getting access can take 2-3 days, if the right person is even around to approve.

And while the SLA (service level agreement) might say “4 days from brief to deploy” the timeline doesn’t account for delays like waiting for access, the analysts building the necessary data points to finalise the audience file, validating the data or fixing broken models. 

Step 3 – Deployment

Please don’t hit send yet!

 

So, you’ve finished building your campaign now it’s time to publish.  

In theory, we would just press send, but in practice, there is a whole checklist before we hit the button: 

  • Final content checks – so we don’t send another “Hi Linda” email to 200 people. 
  • Link tracking validation 
  • Device rendering QA (Outlook quirks)  
  • UAT from marketing or legal teams 
  • Last-minute copy tweaks (always right before sign-off)  

Just to note that the deployment interfaces are often built with technical users in mind not marketers.  

So, if you are not well versed in HTML or how to spot broken tokens or why your audience size has dropped by 30%, then you’re stuck. Some tools require SQL (Structured Query Language) just to pull a simple audience. Others need CSS (Cascading Style Sheets) skills to align your email copy properly.  

If you don’t have those skills in-house, you’re paying freelancers or training existing team members which can quickly become time-consuming and expensive. 

Step 4 – Analyse

The stage we don’t spend much time on…

 

Now your campaign has been launched, the final step would be the campaign analysis. 

Now it’s the time to ask: 

  • Who opened it? 
  • Who clicked? 
  • Did it drive the intended behaviour? 
  • Did it hurt the performance of another campaign?  
  • Did it hit the wrong audience segment? (Yes, it happens)  
  • What can we change in our checklists to avoid repeating mistakes? 

This is where CRM overlaps with analytics. Where some teams have dashboards, others rely on the BI teams to provide post-campaign reports.  

Either way this is still a critical step in the cycle that is often overlooked in the rush to move on to the next campaign.  

Without this feedback loop it makes it difficult to know whether a campaign is genuinely impactful or whether we’re just shouting into the void. 

So… what DO we do?

CRM is more than just sending emails. It’s the behind-the-scenes engine that makes marketing happen. We sit at the intersection of marketing, data, tech, building automated journeys, enforcing contact rules and ensuring the right message reaches the right person at the right time. It’s a balancing act of access, logic, timing and empathy. If no one’s noticed us, that usually means that everything has gone to plan.  

Signed,  

A former Marketing Campaign Manager 

CRM TERM GLOSSARY 

  • SQL (Structured Query Language) – Basically the language we use to talk to databases. If you want to pull an audience to find out who bought trainers in the last 6 days, this is how you ask. Yes, one missing comma can ruin your day. 
  • CSS (Cascading Style Sheets) – This is what makes emails and websites look pretty. Without it everything looks Times New Roman and sad.  
  • Predictive model – A fancy data-powered guess about what a customer might do next. 
  • BI (Business Intelligence) – They turn the data in dashboards and reports, tell you what worked what didn’t and whether a 0.2% CTR was worth the 18-hour build time.  
  • CRM – Customer Relationship Management 
  • SLA (Service Level Agreement) – The timeline that we promise to deliver something in. Often optimistic, occasionally accurate.  
  • QA (Quality Assurance) – We test, test again and then test some more to make sure nothing breaks. 
  • UAT (User Acceptance Testing) – The final “does-this look, okay?” check from marketing and business teams. Usually where last minute “can we please change the copy” tweaks happen.