Capability 03 · Marketing Technology
AI campaign optimisation that moves budget on measured results while campaigns run.
We build conversion tracking, incrementality tests and marketing mix models that show where budget should go. An agent proposes each move with its reasoning; a named person approves any move above your limit.
Day 7: retargeting is cut by the maximum again and sits at its spend floor. Search category asked for 26 percent more; it moved the 20 percent limit and the rest waits for a person to approve.
Illustrative figures in any currency. The dashed tick on each bar is the spend floor.
What we offer · 6 parts
Accurate data first, then models that move budget.
Experiments calibrate each model, so the budget moves it proposes rest on measured lift.
-
Measurement foundations
Server-side tagging, consent mode, deduplicated conversion APIs and offline conversion import, so ad platforms and your warehouse agree on the numbers.
Clean, consented data -
Incrementality & geo experiments
Holdout and geo-split tests that measure what your advertising caused and calibrate every model that follows.
Lift · holdouts -
Marketing mix modelling
Bayesian models of what each channel adds to sales, calibrated against experiments, with budget scenarios by channel and market.
MMM · scenarios -
Budget allocation & bidding
Daily or weekly budget proposals across channels and campaigns, each with expected effect and confidence, applied through platform APIs within your approval rules.
Proposed · approved · applied -
Creative & audience testing
Test plans using sequential tests or multi-armed bandits, alerts when creative wears out and a library of past results.
Tests · bandits -
Anomaly detection & spend caps
Alerts within hours on pacing, cost and conversion problems, plus spend caps, rollback and a kill switch your team controls.
Alerts · caps · rollback
How we use AI
Spend moves on evidence, with caps and approvals built in.
Each budget move goes through these five steps; the model does its work at the highlighted one.
- 01 · Data Conversions in Server-side conversions and offline sales, deduplicated and consented.
- 02 · Model Forecast Expected return per ad set and audience at the next unit of spend.
- 03 · Guard Automatic checks Daily shift limits, spend floors, brand safety and frequency caps. Campaign AI
- 04 · Approve Above threshold Moves above the agreed limit wait for a named person to approve.
- 05 · Test Incrementality Geo and holdout tests check what the platforms report.
How it works · 4 stages
Baseline, tracking and models, then automation.
Each stage relies on the one before. Automation starts once tracking and models can be trusted. Timings are typical, set per engagement.
Stage 01 of 4Wk 01–02
Baseline
We review spend, reporting, attribution and data quality. We measure gaps between platform numbers and your records and design the first experiments.
Produces
- Tracking audit
- Spend baseline
- Experiment plan
Stage 02 of 4Wk 02–05
Track
We build server-side tagging, consent mode, conversion APIs and warehouse feeds, with deduplicated events and one conversion definition.
Produces
- Tracking build
- Warehouse pipeline
- Metric definitions
Stage 03 of 4Wk 05–09
Model
We fit mix models, calibrate them against experiment results, plan budget scenarios and write up what each channel contributes.
Produces
- Mix model results
- Experiment read-outs
- Budget scenarios
Stage 04 of 4Ongoing
Optimise
Budget proposals and creative tests run weekly, spend caps are monitored and models are refitted as data grows.
Produces
- Weekly proposals
- Test log
- Monthly performance review
Holdout read-out · what you will see
Test regions against control regions, then move budget.
Each read-out compares matched regions with and without the change, shows the difference and states how certain it is.
| Measure | Test regions (8) | Control regions (8) | Difference | 90% interval |
|---|---|---|---|---|
| Orders per 10k people | 41.2 | 37.6 | +9.6% (on track) | +4.1% to +14.8% |
| Cost per incremental order | ₹ 612 | —not applicable | Below ₹ 700 target (on track) | ₹ 520 to ₹ 890 |
| New customers share | 38% | 36% | +2 pts (needs attention) | −1 to +5 pts |
| Returns within 30 days | 6.1% | 6.0% | No change (on track) | −0.8 to +1.0 pts |
What you get
Models, tests and read-outs, built in your warehouse.
Each comes with the code to rerun it. What we make for you is yours once it is paid for. Tools we already had stay ours, and you get a free, permanent licence to use them.
| # | Deliverable | Format |
|---|---|---|
| 01 | Measurement plan & metric definitions | Doc · sheet |
| 02 | Server-side tagging & conversion APIs | Container · code |
| 03 | Experiment designs & read-outs | Plan · report |
| 04 | Marketing mix model & scenario tool | Model · dashboard |
| 05 | Budget allocation proposals | Dashboard · log |
| 06 | Creative & audience test library | Sheet · board |
| 07 | Anomaly alerts & spend caps | Alerts · config |
What changes
Three measures of success that finance can check.
Success means lift measured against a control, a lower cost per outcome and budget moved on evidence. These are aims; no result is guaranteed.
-
01
Numbers that agree
One conversion definition across ad platforms, analytics and your warehouse, so every team reports the same figures.
-
02
Budget that follows evidence
Budget moves based on experiment results and each channel’s modelled contribution.
-
03
Mistakes caught in hours
Pacing and cost problems are flagged the same day, with caps and rollback ready to use.
Cost per acquisition by week, optimised vs fixed split. Lower is better. Shape only: budget moved during the campaign against budget kept at its launch split, with the gap checked by a holdout test.
Technologies we work with
Your warehouse first, then the models on top.
The analytics, warehouse, modelling and reporting tools we use to measure and move spend. Listed as technologies we work with; no partnership is implied. Model providers run only inside the approval step.
- Google Analytics
- Google Tag Manager
- Google BigQuery
- Snowflake
- Airbyte
- Python
- scikit-learn
- Looker
- PostHog
- Mixpanel
- dbt
- Power BI
Frameworks we build to
Measured without tracking people further than they agreed.
Each model and tag plan is built to these frameworks. Listing them does not mean we are certified.
-
GDPR — General Data Protection Regulation (EU) 2016/679
General Data Protection Regulation (EU) 2016/679
Consent recorded per purpose and channel, tags fired only after consent (Consent Mode v2 where Google is used) and erasure requests passed to every connected tool.
-
DPDP Act 2023 — Digital Personal Data Protection Act, 2023
Digital Personal Data Protection Act, 2023
Notices and consent records for India-based customers, with withdrawal honoured in every journey and channel within the same working day.
-
NIST AI RMF 1.0 — AI Risk Management Framework
AI Risk Management Framework
Each model that scores, targets or writes is listed with an owner, a measured error rate and a threshold that stops it.
-
EU AI Act — Artificial Intelligence Act (EU) 2024/1689
Artificial Intelligence Act (EU) 2024/1689
AI-generated content labelled where required, targeting and scoring use cases classified by risk, with a person able to override every automated decision.
-
ISO/IEC 27701 — Privacy information management systems
Privacy information management systems
A current list of every marketing tool holding personal data, with its purpose, retention and processor terms.
Services & packages
Optimisation services, from tracking to budget automation.
We build the tracking, experiments and models that show where budget should go, then automate the moves within limits you set.
- Categories
- 05
- Services
- 15
- Packages
- 04
How to buy
- 01Pick services. Enquire about one, or add several to a brief.
- 02Choose a package. A sprint, a fixed project or an ongoing team.
- 03Send the brief. We reply within one working day.
How we work with you
Ways to engage, from a question to an RFQ.
Ask a quick question, send a project brief or issue a formal RFQ. The lead for the work reads each one in full, and any services already in your brief go with it.
Or book a thirty-minute call-
01
For a first conversation, a press request or anything that does not need a scope yet.
You get A reply from a named lead
-
02Recommended
Goals, audiences, a budget band and timing, so our first reply can outline the work.
You get Options and a first scope after one call
-
03
Your documents, deadlines and the procurement and security rules the work must meet.
You get Receipt confirmed and a named bid lead
How it is priced
Each package shows its pricing model. Work starts once a written scope and quote are agreed.
Opens a project brief with this package chosen.
-
In your brief
- Typical length
- 1–3 weeks
- Pricing
- Fixed fee
-
In your brief
- Typical length
- 4–12 weeks
- Pricing
- Fixed price
-
In your brief
- Typical length
- Ongoing · 6-month minimum
- Pricing
- Monthly fee
-
In your brief
- Typical length
- 6–18 months
- Pricing
- Programme fee · by statement of work
Compare what each package includes
| Package | Every engagement includes | Best for |
|---|---|---|
| SprintA short, fixed-scope engagement that answers one defined question. |
|
Discovery, a diagnostic, a prototype or a decision you need to make soon |
| ProjectA defined scope, delivered for a fixed price. |
|
Work you can describe up front: an identity, a platform or a set of tools |
| RetainerReserved monthly capacity to run, improve and extend what we built. |
|
Live brands and products that need a steady team without hiring one |
| EnterpriseA multi-workstream programme with a dedicated team, governance and agreed service levels. |
|
Large organisations running change across markets, portfolios or business units |
Questions
What teams ask about campaign AI.
Anything else, ask us directly; a short call is usually quicker than email.
01Is this the same as Performance Marketing?
No. Performance Marketing, part of Campaign & Content Design, plans and runs the media. This service builds the measurement, models and automation that decide where budget goes. It works with our media team or your current agency.
02Do we need attribution, incrementality testing or mix modelling?
All three answer different questions. Attribution shows customer paths and helps day-to-day operations. Experiments measure the lift one channel causes. Mix modelling covers the whole budget, including channels you cannot track. We calibrate the model with experiments and use attribution as a diagnostic.
03Does lost tracking from privacy changes break this?
No, it changes how we measure. Consent mode, server-side tagging and conversion APIs keep the data people agreed to share. Mix modelling and geo experiments need no individual tracking.
04Will AI be allowed to change our budgets?
Only within limits you set. Each proposal shows its expected effect, confidence and reasoning, and a named person approves anything above your threshold. Caps, exclusions and rollback are always on. Narrow, well-tested decisions can run automatically once the evidence supports it.
05How much data does mix modelling need?
Typically two to three years of weekly history across channels and markets, plus spend, price and seasonality. With less, we start with experiments and lightweight models and build towards it.
Works best with
Better spend needs better creative and better leads.
Creative AI gives the model more ads to test; Lead Gen turns that traffic into sales pipeline.
Tell us what you need built.
You will speak to a lead who would run the work, and get a straight answer on fit.
Three ways to start
Every engagement starts with a written scope and a quote agreed before work begins.
Choose one of the three ways above