Independent Braze consultant for Nordic CRM teams

A Braze consultant who builds in the workspace

If your Braze work goes out as tickets to an agency, the brief and the build sit with different people. I’m one person. The consultant you talk to on the call is the one who builds the Canvases, the segments behind them, the Liquid that personalises them and the data they run on.

I was Head of CRM at IKEA Ingka Group and at Bauer Media Group, and I’ve run CRM from the inside for fifteen-plus years. My first questions about any Canvas are who it should leave out, and how we’ll know it helped.

Based in Malmö, a short train ride from Copenhagen. The call is on Google Meet.

What I check first in a Braze workspace

When I start with a team on Braze, these are the first things I check, in this order.

• Canvases and their exits. Which Canvases are live, which event puts people in, and what takes them out. Someone who has done what a Canvas asks for should leave before the next step sends.
• Segments. Whether they run on behaviour, or on attributes that stopped updating when the tracking changed.
• Frequency capping and suppression. Which Canvases and campaigns are exempt from the caps, and whether anyone remembers why.
• Consent. What each customer agreed to, in each market, and whether Braze knows it.
• Liquid. Whether personalisation has a fallback, so a missing first name doesn’t go out as “Hi ,”.
• Measurement. Which Canvases run with a holdout, and whether Currents streams events out to where your data team can read them.
• Intelligent Timing and BrazeAI. Where they’re switched on, and whether anyone has checked them against what they replaced.

When a propensity score should decide who enters a Canvas, I build the model in Databricks, with XGBoost and MLflow.

On the 20-minute call, we start with your Canvases.

What I’ve built on Braze

At Tibber, energy-tech in four markets with about 550,000 customers, I built the referral program on Braze. Referral grew to roughly a third of new customers (client-reported). The receipt below has the first-week numbers.

Also at Tibber, I built Spark in-house, an AI email and content tool integrated with Braze. It wrote brief-to-send email and SEO content, and email won by a distance. The tone-of-voice scorer went unused. The receipt below has why.

So I’d ask BrazeAI, or any model in your workspace, what job it does and whether anyone needs that job done.

Braze is one of three marketing automation platforms I build in hands-on. The other two are Klaviyo and Salesforce Marketing Cloud Next.

Bringing me in on Braze

There are three ways in.

• Someone to own Braze and the program on it. That’s the Fractional Head of CRM: part-time, inside your team, deciding what ships next in the workspace and building it. Two clients at a time, never three.
• One program built properly in Braze. Onboarding, retention, win-back, referral or loyalty, as a fixed-scope lifecycle marketing program. By the time the scope closes, its Canvases, triggers, segments and measurement are sending in your workspace.
• Where BrazeAI and your own models pay off. The AI marketing audit puts the candidates in a build order, with what I’d build first at the top.