The growth engine, as I currently see it
I scored drivers on what I can actually see. Organic search is the highest-quality scalable engine: 11% of sessions, 42 seconds. Paid Social is a proven volume driver with weak engagement. LinkedIn is a proof-led click signal; I cannot yet call it a cause of site traffic.
01
Direct / brand discovery
Most durable engine: medium evidence that it is purely brand
What I can see. 8,120 sessions (37.09%), 65.23% engagement. Led most 2026 months after the July 2025 paid burst faded. August Direct reached 677 sessions.
Where I am less sure. Untagged campaign leakage and LinkedIn without UTMs may inflate Direct. No branded-search series.
What changed. Became the recurring base after August 2025. Still cannot separate brand demand from untagged traffic.
Fact
02
Organic search
Highest-quality scalable engine
What I can see. 2,461 sessions, 54.00% engagement, 42 seconds average time. Peaked July 2026 at 253 sessions when paid was immaterial.
Where I am less sure. Share is only 11.24%. Residual archive vs 2026 commercial pages is not separated.
What changed. Small but steady, then a July 2026 quality peak. Blog publishing remained zero throughout.
Fact
03
Paid Social
Major volume driver with weak engagement
What I can see. 4,773 sessions, 23.63% engagement, 3 seconds average time. Created the December–January and March–April bursts.
Where I am less sure. Placement, geography and in-app (Android WebView 36.1% of users, 9 seconds) still need a campaign audit.
What changed. Emerged December 2025; large Mar–Apr burst; smaller August return (199 sessions).
Fact
04
Paid Other
Burst driver: source detail required
What I can see. 3,614 sessions (16.51%), 45.71% engagement. Dominated July 2025 with 1,924 sessions (76.6% of that month), then vanished, then returned December–Q1.
Where I am less sure. Channel label obscures campaign, placement and geography.
What changed. Looks campaign-led, not organic baseline growth. Fades after Q1 2026.
Fact
05
LinkedIn / founder brand
Case-study signal is strong; attribution is weak
What I can see. 18 dated rows Apr–Jun 2026; 352 clicks; 169 (48%) on the 10 June Gulf banking case study. Organic Social was only 7.34% of sessions.
Where I am less sure. Impressions blank; n=3 correlations unusable; months outside Apr–Jun unknown.
What changed. Tracker cadence rose 1 → 4 → 13 while Organic Social stayed 94 / 73 / 74.
Hypothesis
06
Commercial website (2026 IA)
Observed structural change: not joined to a traffic step-change
What I can see. July–August 2026 lastmod cluster. High-intent pages (Sales Squad 52s, Pricing 47s, Case Studies 32s) are underexposed versus 19,419 homepage views.
Where I am less sure. Lastmod is not a deploy timestamp. Traffic cannot be attributed to the redesign from current-state pages alone.
What changed. Positioning is stronger than distribution and conversion plumbing.
Fact
07
On-site publishing
Inactive inside the window
What I can see. 0 original posts from 1 Jul 2025 to 30 Sep 2026; last post 21 May 2025; blog index 207 views.
Where I am less sure. Historical posts may still attract residual search; that residual is inside the Organic Search total, not isolated.
What changed. Fell from 17 posts in Jan–May 2025 to zero for 15 months.
Fact
08
Referrals / partnerships
Cannot claim partner-led growth
What I can see. 633 referral sessions overall (90.84% engagement). January 509-session burst at 99.2% engagement and 0.79% key-event rate is implausible for commercial traffic.
Where I am less sure. Anomaly unverified (bots, internal, low-intent). Case-study logos are not referral sessions.
What changed. Little observed until January; low after.
Hypothesis
09
Repeat visitors
Not yet a compounding engine
What I can see. New users 18,383 vs active users 18,435. Monthly returning users ranged 26–178; January peak 178, then 46–112 through August.
Where I am less sure. No cohort report. Blog cadence that might create return reasons was off.
What changed. Still small and inconsistent.
Fact
What changed
Publishing engine off
Before. Jan–May 2025: 17 original posts, mixed English/Arabic.
Change. No original publish after 21 May 2025.
After. Jul 2025–Sep 2026: 0 posts. Blog hub copy later refreshed (Aug 2026) without new articles. Blog index: 207 views.
The numbers. Blog publish dates vs /blog/ lastmod vs GA4 page report.
What this means for us. Inbound thought-leadership was not the 15-month growth programme. Residual search from the archive is inside Organic Search, not a new-article engine.
Acquisition mix became bursty
Before. July 2025 Paid Other-led burst (1,924 of 2,512 sessions).
Change. Direct became the recurring base; Paid Social added December–January and March–April bursts.
After. Late period is more Direct and Organic between campaigns. Q1 2026 +208.7% sessions; Q2 −66.8%.
The numbers. Monthly channel rows transcribed from GA4.
What this means for us. Lower paid dependence is possible, but not yet secure. Quality improved when paid receded (July 2026).
Offer architecture rebuilt
Before. Legacy /home/: managed sales provider, Closed Won language, older proof metrics.
Change. July–August 2026 IA lastmods; home-2026 as front page.
After. Squad vs full squad, retainer + commission, live in 4 weeks, market validation as pre-sale. High-intent pages engage well and remain underexposed.
The numbers. Sitemap lastmods, live vs legacy homepage copy, GA4 page views.
What this means for us. Positioning is stronger than distribution. Digital should be judged as a conversion site for the squad offer: once key events are rebuilt.
Proof metrics restated
Before. Legacy page: 495+ vetted deals, $95M, $5M recognised, 600M+ contacts, 45+ solutions sold.
Change. Current homepage: $100M+ pipeline, $10M+ revenue, 25+ deployments, 20+ industries; packages cite 20M+ GCC contacts.
After. Two live URLs currently advertise two histories.
The numbers. / vs /home/ on 17 Sep 2026.
What this means for us. Sales conversations can be contradicted by Google. This is a credibility risk independent of GA4.
Reported conversions are not decision-grade
Before. Brief assumed key events could stand in for leads.
Change. 13,250 key events coexist with 28 form-submission users; careers 97.5% key-event rate; revenue £0.
After. The usable funnel is 18,435 → 930 CTA → 313 form start → 28 submit.
The numbers. GA4 events and user counts in the 17 Sep 2026 report.
What this means for us. I would not brief ROI from the key-event total. I want a qualified-lead definition and a CRM join before scaling paid.