Service
Optimisation for the number that stopped moving
This page is for the specialised jobs: a stalled metric, a deliverability problem, a journey that stopped converting, a migration that hurt performance. If you are already on an Execution retainer, ongoing optimisation is part of it and this work is included.
Talk to usWe reply within one working day.
When you need this
A metric has stalled
Conversion, repeat purchase or engagement sat flat through the last few cycles. The obvious changes have been tried and nothing moved.
Mail stopped reaching the inbox
Delivery looks fine on paper, but engagement fell off a cliff at one provider. Something in authentication, list hygiene or sending pattern changed and nobody can say what.
A journey stopped converting
An always-on journey that used to carry revenue now runs quietly and returns very little. Nobody has opened it since it was built.
The reporting cannot answer the question
The board asks what CRM contributed and the answer arrives late, with a caveat. Two teams count a conversion differently and both are sure they are right.
What we optimise
Deliverability and data quality
- Authentication and sending domain checked against what the providers now enforce
- Sending reputation repaired: list hygiene, re-engagement and the volume that caused it
- Inbox placement traced provider by provider, to find where the mail actually lands
- Consent and suppression records reconciled with what is going out
- Identity and event quality fixed wherever two segments disagree
Journeys, segmentation and frequency
- A review of every always-on journey against what it was built to do
- Trigger, timing and entry-condition fixes
- Segmentation rebuilt where it no longer separates anyone
- Frequency capping and channel priority across journeys
- Retirement of journeys that no longer earn their place
Testing, reporting and attribution
- Test design: audience, split and the metric that decides it
- Holdout groups on always-on journeys, so you see incremental revenue
- Significance rules agreed before the test runs
- Documented definitions, so two teams read one number the same way
- A dashboard tying CRM activity to revenue, built where your team already looks
How we work
Diagnose
We take read access, reproduce the problem and separate the cause from the symptom before touching anything.
Fix
We make the change inside your platform and measure it against a control or a holdout wherever the journey is always on.
Hand back
We document what was wrong, what we changed and what to watch, then hand it to whoever owns it day to day.
Timelines depend on scope and on how we work together, so we estimate them after a scoping call rather than up front.
Platforms we work in
- Braze
- Bloomreach
- Insider
- Maestra
- Hightouch
- Snowflake
- Databricks
Related services
Tell us which number stopped moving
Send us how you measure it today and where the reporting stops. We will come back with what we would look at first.