PA Data Analytics

Marketing · Customer · Business Intelligence

Your data is already telling a story. Read it.

PA Data Analytics is an analytics consultancy based in Nairobi. We provide marketing analytics, customer intelligence, predictive analytics and business intelligence for e-commerce brands, growing businesses and organisations that want to make better decisions from their data.

Start a Conversation Explore Our Services
  • Founder-led
  • Nairobi, Kenya
  • Remote engagements
Illustrative dashboard
Sample data for demonstration only. Not a client or project result.
Marketing performance overview
Sample period
Revenue
$482K
Is growth on track?
Blended ROAS
3.6x
Which channels pay back?
CAC
$58
What does a customer cost?
Conversion
4.8%
Where does the funnel leak?
Revenue vs target
RevenueTarget
Return per $1 spent, by channel
  • Paid search54% of spend2.9x
  • Paid social28% of spend3.4x
  • Display13% of spend1.8x
  • Email5% of spend11.2x
  1. DataSpend, revenue and conversions by channel
  2. InsightEmail returns the most per dollar but gets 5% of spend
  3. DecisionTest a gradual budget shift and measure before scaling

From our portfolio case studies. Figures describe the scale of the demonstration analyses.

  • 6
    End-to-end case studies
  • 11,292
    Touchpoints in attribution modelling
  • 5
    Attribution models compared
  • 104 wks
    Of data in a marketing mix model
The business problem

You have the data. The hard part is turning it into a decision.

Most businesses collect more data than ever: ad platforms, GA4, Shopify, CRMs, spreadsheets. The difficulty is joining it up, trusting it, and knowing what to do differently as a result.

“We spend on marketing, but we don't know what's working.”

Channel performance, attribution and marketing ROI, measured on the same basis.

“We have customers, but we don't know who to retain.”

Valuable customers, churn risk and retention opportunities, identified from your own data.

“We're growing, but we can't see what's driving it.”

Revenue, customer and operational data connected into one performance picture.

“We spend too much time preparing reports.”

Dashboards and automated reporting that free time for decisions.

Data → Insights → Decisions

Reports tell you what happened. Analysis should tell you what to do next.

A dashboard that only restates last month's numbers is not enough. Every PA Data Analytics project works through three questions and ends with a recommendation you can act on.

  1. DATA

    What happened?

    Clean, reconcile and structure the data so the numbers can be trusted.

  2. INSIGHTS

    Why did it happen?

    Use attribution, segmentation, cohorts and models to find the drivers behind the numbers.

  3. DECISIONS

    What should happen next?

    Turn findings into priorities, tests and budget choices, with the uncertainty stated plainly.

Services

Analytics services built around business decisions

Four core services, from understanding marketing performance to predicting customer behaviour and building the reporting your team uses every week.

Marketing Analytics

See which channels, campaigns and customer journeys actually drive revenue, and where next month's budget should go.

  • Campaign analytics
  • Attribution
  • Marketing mix modelling
  • Funnel analysis

Outcome: Clearer marketing ROI and better-informed budget allocation.

Marketing Analytics services

Customer Intelligence

Understand who your valuable customers are, when customers stop returning, and where the retention opportunities are.

  • Customer segmentation
  • RFM analysis
  • Cohort retention
  • Customer lifetime value

Outcome: Know which customers to retain, win back and invest in.

Customer Intelligence services

Predictive Analytics

Go beyond reporting what happened to estimating what is likely to happen next, so you can act earlier.

  • Churn prediction
  • Forecasting
  • Predictive modelling
  • Experimentation

Outcome: Earlier action on churn risk and more informed planning.

Predictive Analytics services

Dashboards & Reporting

Bring the right numbers together in Power BI, Excel or Looker Studio so your team can make decisions with confidence.

  • Power BI
  • Excel
  • Looker Studio
  • Reporting automation

Outcome: Less manual reporting and one trusted set of numbers.

Dashboards & Reporting services

Also available: e-commerce analytics, A/B testing and experimentation, data cleaning and reporting automation, and NGO impact and M&E reporting.

Audiences

Who we work with

Organisations that already collect data and want clearer answers from it, whether or not they have an analytics team of their own.

E-commerce brands

Understand customer journeys, conversion funnels, product performance, retention and the channels that really drive revenue, not just the ones that got the last click.

E-commerce marketing analytics

Growing businesses & SMEs

Get the analysis you need without hiring a full analytics team: clear KPI reporting, one view of performance, and answers to specific commercial questions.

Dashboards and reporting

Marketing teams

Understand campaign performance, attribution, channel efficiency and marketing ROI, and reconcile the numbers each ad platform reports differently.

Marketing analytics services

SaaS & startups

Track recurring revenue health, spot churn risk early and understand which customers and plans drive retention.

Churn prediction

NGOs & impact organisations

Turn programme and survey data into indicator tracking, M&E reporting and donor dashboards you can stand behind.

Impact reporting dashboards
Case studies

From raw data to meaningful decisions

Portfolio analyses showing how we approach a business question: the data, the method, what the analysis shows and how a business could use it.

Portfolio projects built on demonstration datasets. They are not client engagements, and their figures describe the datasets and models, not real business results.

SHARE OF REVENUE CREDITShapleyLast ClickPaid SearchPaid Search: 35.6% of credit under Last ClickPaid Search: 41.9% of credit under Shapley35.6%41.9%DirectDirect: 8.4% of credit under Last ClickDirect: 51.9% of credit under Shapley8.4%51.9%EmailEmail: 18.9% of credit under Last ClickEmail: 0.0% of credit under Shapley18.9%0%Paid SocialPaid Social: 15.0% of credit under Last ClickPaid Social: 6.2% of credit under Shapley15%6.2%InfluencerInfluencer: 7.6% of credit under Last ClickInfluencer: 0.0% of credit under Shapley7.6%0%Portfolio project · Synthetic dataset
  • Portfolio project
  • Synthetic dataset
  • Marketing Analytics

Comparing five attribution models to understand channel contribution

A portfolio analysis that builds five attribution models on the same customer journeys, to show how much a channel's credit depends on the model chosen.

Synthetic dataset: 11,292 touchpoints · 3,500 customers · 5 models

Read the attribution case study
MONTHLY REVENUEModel-fittedActual$0.6M$1.0M2023-01: actual $780,659, model-fitted $777,9402023-02: actual $756,909, model-fitted $756,3382023-03: actual $724,723, model-fitted $728,0512023-04: actual $705,401, model-fitted $705,1562023-05: actual $945,021, model-fitted $952,0832023-06: actual $752,169, model-fitted $730,9202023-07: actual $873,979, model-fitted $891,5472023-08: actual $791,576, model-fitted $790,1582023-09: actual $729,243, model-fitted $724,5182023-10: actual $894,399, model-fitted $913,6452023-11: actual $933,780, model-fitted $924,6182023-12: actual $948,259, model-fitted $952,0402024-01: actual $846,935, model-fitted $858,9172024-02: actual $794,305, model-fitted $805,7142024-03: actual $766,408, model-fitted $763,6472024-04: actual $948,479, model-fitted $936,6722024-05: actual $789,934, model-fitted $797,6692024-06: actual $787,235, model-fitted $784,7892024-07: actual $928,345, model-fitted $942,9262024-08: actual $826,781, model-fitted $803,0612024-09: actual $963,461, model-fitted $963,2772024-10: actual $752,190, model-fitted $763,5462024-11: actual $978,515, model-fitted $975,4352024-12: actual $1,004,106, model-fitted $980,147Jan 2023Jan 2024Dec 2024Portfolio project · Synthetic dataset · R² 0.957 in-sample
  • Portfolio project
  • Synthetic dataset
  • Marketing Analytics

Marketing mix modelling and budget allocation analysis

A portfolio marketing mix model that estimates each channel's contribution to revenue from two years of weekly data and tests how a fixed budget could be reallocated.

Synthetic dataset: 104 weeks · 4 paid channels · R² 0.957 (in-sample)

Read the marketing mix modelling case study
RETENTION BY COHORTOct 232023-10 cohort, month 1: 39% retained2023-10 cohort, month 2: 22.9% retained2023-10 cohort, month 3: 18.5% retained2023-10 cohort, month 4: 17.1% retained2023-10 cohort, month 5: 15.1% retained2023-10 cohort, month 6: 10.2% retained2023-10 cohort, month 7: 14.6% retained2023-10 cohort, month 8: 10.7% retained2023-10 cohort, month 9: 12.2% retained2023-10 cohort, month 10: 10.2% retained2023-10 cohort, month 11: 7.3% retained2023-10 cohort, month 12: 8.3% retained2023-10 cohort, month 13: 7.8% retained2023-10 cohort, month 14: 5.4% retained2023-11 cohort, month 1: 40.5% retained2023-11 cohort, month 2: 20.3% retained2023-11 cohort, month 3: 21.2% retained2023-11 cohort, month 4: 16.2% retained2023-11 cohort, month 5: 12.6% retained2023-11 cohort, month 6: 12.2% retained2023-11 cohort, month 7: 11.3% retained2023-11 cohort, month 8: 9.9% retained2023-11 cohort, month 9: 10.4% retained2023-11 cohort, month 10: 10.4% retained2023-11 cohort, month 11: 6.3% retained2023-11 cohort, month 12: 9% retained2023-11 cohort, month 13: 8.6% retained2023-12 cohort, month 1: 43.1% retained2023-12 cohort, month 2: 23.9% retained2023-12 cohort, month 3: 22.3% retained2023-12 cohort, month 4: 12.2% retained2023-12 cohort, month 5: 16% retained2023-12 cohort, month 6: 9% retained2023-12 cohort, month 7: 13.8% retained2023-12 cohort, month 8: 9.6% retained2023-12 cohort, month 9: 9.6% retained2023-12 cohort, month 10: 7.4% retained2023-12 cohort, month 11: 9% retained2023-12 cohort, month 12: 9.6% retainedJan 242024-01 cohort, month 1: 35.2% retained2024-01 cohort, month 2: 20% retained2024-01 cohort, month 3: 20.5% retained2024-01 cohort, month 4: 15.2% retained2024-01 cohort, month 5: 14.8% retained2024-01 cohort, month 6: 9% retained2024-01 cohort, month 7: 9.5% retained2024-01 cohort, month 8: 9% retained2024-01 cohort, month 9: 12.4% retained2024-01 cohort, month 10: 10.5% retained2024-01 cohort, month 11: 10.5% retained2024-02 cohort, month 1: 39.1% retained2024-02 cohort, month 2: 25.5% retained2024-02 cohort, month 3: 21.1% retained2024-02 cohort, month 4: 17.4% retained2024-02 cohort, month 5: 15.5% retained2024-02 cohort, month 6: 9.9% retained2024-02 cohort, month 7: 12.4% retained2024-02 cohort, month 8: 8.1% retained2024-02 cohort, month 9: 6.2% retained2024-02 cohort, month 10: 7.5% retained2024-03 cohort, month 1: 33.9% retained2024-03 cohort, month 2: 22.4% retained2024-03 cohort, month 3: 18.8% retained2024-03 cohort, month 4: 10.4% retained2024-03 cohort, month 5: 9.4% retained2024-03 cohort, month 6: 13.5% retained2024-03 cohort, month 7: 10.9% retained2024-03 cohort, month 8: 8.3% retained2024-03 cohort, month 9: 10.4% retainedApr 242024-04 cohort, month 1: 34.1% retained2024-04 cohort, month 2: 22.5% retained2024-04 cohort, month 3: 17.6% retained2024-04 cohort, month 4: 13.7% retained2024-04 cohort, month 5: 15.4% retained2024-04 cohort, month 6: 9.9% retained2024-04 cohort, month 7: 12.1% retained2024-04 cohort, month 8: 10.4% retained2024-05 cohort, month 1: 36.8% retained2024-05 cohort, month 2: 20.3% retained2024-05 cohort, month 3: 13.4% retained2024-05 cohort, month 4: 11.7% retained2024-05 cohort, month 5: 13% retained2024-05 cohort, month 6: 10.4% retained2024-05 cohort, month 7: 6.5% retained2024-06 cohort, month 1: 37.8% retained2024-06 cohort, month 2: 27.9% retained2024-06 cohort, month 3: 16.9% retained2024-06 cohort, month 4: 12.9% retained2024-06 cohort, month 5: 9.5% retained2024-06 cohort, month 6: 16.9% retainedJul 242024-07 cohort, month 1: 38.8% retained2024-07 cohort, month 2: 27% retained2024-07 cohort, month 3: 19.4% retained2024-07 cohort, month 4: 15.8% retained2024-07 cohort, month 5: 13.8% retained2024-08 cohort, month 1: 39.5% retained2024-08 cohort, month 2: 20.6% retained2024-08 cohort, month 3: 14.9% retained2024-08 cohort, month 4: 15.8% retained2024-09 cohort, month 1: 38% retained2024-09 cohort, month 2: 25.9% retained2024-09 cohort, month 3: 20% retainedOct 242024-10 cohort, month 1: 38.2% retained2024-10 cohort, month 2: 21.3% retained2024-11 cohort, month 1: 36.9% retainedM1M3M6M9M125%45%+Portfolio project · Synthetic dataset · Avg M1 37.9%
  • Portfolio project
  • Synthetic dataset
  • Customer Intelligence

Cohort retention analysis: when do customers stop coming back?

A portfolio cohort analysis of 15 months of e-commerce orders, with a reusable SQL pipeline and retention heatmaps showing when customers stop returning.

Synthetic dataset: 2,810 customers · 6,404 orders · 15 cohorts

Read the cohort retention case study
View all six case studies
How we work

A clear process from question to decision

Every engagement starts with the business question, not the tool.

  1. Discovery

    Agree the business question, the decision it supports and what a useful answer looks like.

  2. Data audit

    Review the data you have: sources, quality, gaps and structure. We confirm what is feasible before any building starts.

  3. Analysis & build

    Clean and analyse the data, develop models where they are justified, and build dashboards or reports.

  4. Delivery & handover

    Present the insights, recommendations and next steps, with documentation so your team can keep using the work.

About PA Data Analytics

A data analytics consultancy built around better decisions

PA Data Analytics is a data analytics consultancy helping businesses turn complex data into clear insights and practical decisions.

We work with businesses, e-commerce brands, marketing teams, growing organisations and impact-focused organisations to improve how they understand and use their data. Our work spans:

From fragmented spreadsheets and marketing data to interactive dashboards and analytical models, we help organisations move from data to insights to better decisions.

DataInsightsBetter decisions

Based in
Nairobi, Kenya
Service area
Remote & international
Model
Founder-led consultancy

About PA Data Analytics and our approach

Illustrative business intelligence dashboard with sample data: revenue, ROAS and CAC cards, a revenue-versus-target trend, ROAS by channel and a customer segment table
Illustrative dashboard · sample data

Analysis & modelling

  • Python
  • Pandas
  • NumPy
  • Statsmodels
  • Scikit-learn

Data

  • SQL
  • BigQuery
  • Excel

BI & visualisation

  • Power BI
  • Tableau
  • Looker Studio

Measurement

  • GA4
  • Attribution
  • MMM
Digital products

Excel dashboard templates for teams without a BI budget

Practical Excel workbooks for sales forecasting, marketing ROI, business performance and content tracking. Add your own figures and the dashboards calculate the rest.

Browse all dashboard templates
Insights

Practical writing on marketing and customer analytics

Explanations and worked examples for people making data-driven marketing and customer decisions.

Attribution   · 7 min read

Why Last-Click Attribution is Killing Your Ad Efficiency

Last-click reports make retargeting look brilliant and prospecting look wasteful. Here is how that distorts budget decisions, and which attribution models and tests to use instead.

Customer Intelligence   · 9 min read

A Beginner's Guide to RFM Segmentation in SQL

Build recency, frequency and monetary scores from a transactions table, assign customer segments, and turn them into actions your marketing team can use.

Read all insights

Let's turn your data into your next decision.

Tell us what you are trying to understand: marketing performance, customer retention, forecasting or reporting. We will reply with practical next steps.

Prefer email? [email protected]

Start a Conversation Book a Meeting