BaP Data Data Consultancy Agency · EMEA
Field notes · 11 published pieces · updated monthly

Field notes.

Long-form analysis on attribution, SaaS metrics, SME data infrastructure and pricing. Every piece is sourced to primary benchmarks (SaaS Capital, High Alpha, KeyBanc, Benchmarkit) or built from public-source audits. No thought-leadership fluff.

N° 11
BAP DATA · FIELD NOTES · N°11 E-COMMERCE · 12 MIN
Shopify to BigQuery:
one source
of truth.
NATIVE CONNECTOR
WHOLE WAREHOUSE
£5.35
▼
WITH AD PLATFORMS
MANAGED ETL ADDED
£79–199
▲
JIMMY OKOTH · BAP DATA Under £6/mo to start V.2.1 · JUL 2026
E-COMMERCE

Shopify to BigQuery integration: one source of truth without a data team

The three ingestion routes, what they actually cost per month, the schema you'll receive, and the reconciliation table that keeps your numbers honest.
JUL 2026 · 12 MIN
↗
N° 10
BAP DATA · FIELD NOTES · N°10 E-COMMERCE · 10 MIN
Platform ROAS
is lying
to you.
SUMMED ROAS
PLATFORM CLAIMED
3.96x
×
REAL BLENDED
AFTER DEDUPE
2.9x
▼
JIMMY OKOTH · BAP DATA 1.06x over-claimed V.2.1 · JUL 2026
E-COMMERCE

Platform ROAS is lying to you. Here is the number that matters

Meta says 4.2x. Google says 3.8x. Your bank balance says otherwise. The Media Efficiency Ratio, the formula, and a worked example you can copy.
JUL 2026 · 10 MIN
↗
N° 09
BAP DATA · FIELD NOTES · N°09 E-COMMERCE · 14 MIN
Consent Mode v2:
the hole in
your data.
EEA TRAFFIC
SHARE OF SESSIONS
50%
×
MODELLED
NOT MEASURED
≈27%
▼
JIMMY OKOTH · BAP DATA 50% × 55% decline V.2.1 · JUL 2026
E-COMMERCE

What Consent Mode v2 is doing to your attribution data

Most brands implemented Consent Mode v2, ticked a box, and moved on. The four mistakes almost every EEA setup makes, quantified — and the two ways to fix them.
JUL 2026 · 14 MIN
↗
N° 08
BAP DATA · FIELD NOTES · N°08 E-COMMERCE · 12 MIN
Why GA4 & Shopify
will never
agree.
GA4
BROWSER EVENT
£18,200
▼
SHOPIFY
SERVER RECORD
£21,400
▲
JIMMY OKOTH · BAP DATA Δ £3,200 · −15% V.2.1 · JUN 2026
E-COMMERCE

Why GA4 and Shopify numbers never agree

GA4 says £18,200. Shopify says £21,400. The gap is a signal. Direction tells you the cause, stability tells you the urgency — and the four-step method that closes it for good.
JUN 2026 · 12 MIN
↗
N° 01
BAP DATA · FIELD NOTES · N°01 SAAS · 08 MIN
Why your SaaS metrics
are lying to you.
SOURCE 01 · STRIPE
RAW EVENTS
£186,420
▲
SOURCE 02 · CHARTMOGUL
NORMALISED
£174,810
≠
SOURCE 03 · SHEET
INVOICED
£182,205
▼
JIMMY OKOTH · BAP DATA Δ £11.6K V.2.1 · APR 2026
SaaS

Why your SaaS metrics are lying to you

Three systems, three MRR numbers, zero confidence. Why Stripe, ChartMogul and your finance spreadsheet never agree — and the reconciliation that fixes it.
APR 2026 · 08 MIN
↗
N° 02
BAP DATA · FIELD NOTES · N°02 E-COMMERCE · 12 MIN
5 revenue leaks
hiding in your e-commerce data.
£1.42M
£1.06M
JIMMY OKOTH · BAPDATA.COM £353K UNTRACKED V.2.1 · APR 2026
E-COMMERCE

5 revenue leaks hiding in your e-commerce data

The attribution gaps, abandoned-cart signal and review decay every Shopify store is sitting on — and the SQL I use to surface each.
APR 2026 · 12 MIN
↗
N° 03
FIELD NOTES · N°03 · PRICING 14 MIN READ
An actuary
who stopped
pricing insurance.
// PRICING · IDENTITY
Price = E[Loss] + λ·σ + ε
LTV = Σ p(t) · m(t) · dt
CAC payback = CAC ÷ (ARPA · GM)
BAP DATA · PRICING NOTES ◐ V.2.1
PRICING

The actuarial approach to e-commerce pricing

Why insurers and pet-food brands have more in common than you think. A pricing model that gives you loss ratios instead of margin guesses.
MAR 2026 · 14 MIN
↗
N° 04
BAP DATA · FIELD NOTES · N°04 SAAS · 10 MIN
◐   ONE PAGE · NINE NUMBERS
The dashboard
your SaaS board
actually needs.
Board · Q2 · 2026
  LIVE
MRR
£186K
▲ +8.2%
NRR
112%
▲ +4 pp
CAC
£1,820
▲ −6%
LTV
£9,440
▲ +11%
PAYBACK
14 MO
▲ −2 mo
BURN ×
1.4
▲ −0.3
GROSS M
76%
▲ +1 pp
CHURN
2.1%
▼ +0.4
R40
48
▲ +3
STRIPE · POSTGRES · HUBSPOTV.2.1
JIMMY OKOTH · BAP DATA ◐ READ 10 MIN
SaaS

The dashboard your SaaS board actually needs

Nine numbers your investors keep asking for — MRR, NRR, CAC, LTV, burn multiple, Rule of 40. What to keep, what to cut, and the math behind each.
MAR 2026 · 10 MIN
↗
N° 05
FIELD NOTES · N°05 · SME 11 MIN
$2M
running on
spreadsheets.
// FOUR SIGNS
01 · MONTHLY CLOSE TAKES 3 DAYS
02 · NUMBERS DIFFER ACROSS SHEETS
03 · CASH FLOW SURPRISES YOU
04 · ONLY ONE PERSON UNDERSTANDS IT
BAP DATA · SME INFRASTRUCTURE ◐ V.2.1 · 2026
SME

You're running a $2M business on spreadsheets

Four signs your business has outgrown spreadsheets, and the cheapest viable path from workbook-of-truth to a dashboard you can check in 10 minutes.
FEB 2026 · 11 MIN
↗
N° 06
BAP DATA · FIELD NOTES · N°06 PRICING · 13 MIN
◐   LTV · COHORT-CORRECTED
Why your
dashboard
isn't making
you money.
RETENTION · C01–C06 X: MONTH · Y: P(ACTIVE)
C01 C02 C03 C04 C05 C06
LTVtrue = Σ p(t)·m(t)·dt n = 11,420 · NUTRISEED
JIMMY OKOTH · EX-ACTUARY ◐ JAN 2026 · V.2.1
PRICING

Why your dashboard isn't making you money

Most LTV calculations are wrong because they ignore cohort decay. A cleaner method, with a worked example on a nutrition subscription brand.
JAN 2026 · 13 MIN
↗
N° 07
BAP DATA · FIELD NOTES · N°07 E-COMMERCE · 11 MIN
◐   4 LAYERS · 40 CHECKPOINTS
What a
data audit
actually
includes.
AUDIT LAYER · CHECKPOINTS
01 · TRACKING
15
02 · ATTRIBUTION
10
03 · REPORTING
08
04 · DATA QUALITY
07
JIMMY OKOTH · BAP DATA 40 CHECKPOINTS V.2.1 · JUN 2026
E-COMMERCE

What a data audit actually includes

Our 40-checkpoint framework across four layers: tracking, attribution, reporting, data quality. What we look at, what we find, what the deliverable looks like, and what it costs.
JUN 2026 · 11 MIN
↗

Want the audit framework behind every piece?

→ Get the checklist