— 13 search & organic growth intent-led, technical-first

we don't chase rankings. we build the asset.

a technical base that crawls cleanly, pages that answer what people actually search, and links earned rather than bought. one plan across search and content, and a weekly read that says one of three things: scale, hold, or refresh. an seo agency reports the numbers; we also write the pages that move them.

— minimum engagement
AED 1,250/mo entry
— review cadence
weekly read, monthly board
— content output
8–12 pages / month
— measurement
revenue + position
01. why this matters now

most seo programs report traffic, not customers.

search got harder to game and easier to lose. ai answers took the top of the page, and most agencies still send the same monthly deck full of vanity dashboards and "we've optimised the meta descriptions."

an seo program that works in 2026 isn't a link-buying engagement — it's three jobs at once: an engineering pass that makes the site legible, an editorial team that knows what each page is actually for, and a measurement layer that tells you which dollars made the next dollar. most agencies do one of those well, two badly, and bill for all three.

we run search the way a category challenger runs search: with a thesis on every cluster, an editorial volume that makes the site the obvious answer, and a measurement stance that survives algorithm updates and ai summaries. not "we'll get you rankings." rankings for what, against whose baseline, worth what per month.

and we make the cut calls. if a cluster doesn't earn its position inside the agreed window, we consolidate it or kill it. no agency paid per article will ever say "publish less here." we will. often.

~38%
— of organic clicks landing on "branded search" that customers were going to buy anyway
3:1
— typical refresh-to-new ratio. pages slip because the content went stale, not because the site is weak
+24%
— revenue lift from cutting the thinnest pages, not adding more
0
— agencies incentivised to tell you to publish less when they bill per article
— case studies

three projects, the numbers they produced.

the brief, the work, and what moved — with the client's own verdict at the end of each.

— case 01 Endeavour Service Group cleaning + maintenance 6 months
the brief

a commercial cleaning and maintenance operator whose site described services in its own language rather than the words customers were searching for.

what we did
  • rewrote 18+ service pages around the queries customers actually type
  • restructured targeting across 35+ keywords by intent, not volume
  • earned 80+ relevant links from trade and local sources
+42%
organic traffic
18+
pages optimised
35+
keywords targeted
80+
backlinks built

The SEO work was thorough, communication was clear and the whole process was easy to follow.— Endeavour Service Group

— case 02 Awards4U custom awards + promotional 5 months
the brief

a custom awards and promotional products supplier with a wide catalogue and almost no visibility on the specific product terms buyers search.

what we did
  • optimised 22+ pages against the catalogue's real search demand
  • built targeting across 40+ product and category keywords
  • added 90+ backlinks as a supporting layer, not the strategy
15+
keywords in top 10
22+
pages optimised
40+
keywords targeted
90+
backlinks built

The content improvements, keyword work and backlink strategy were handled professionally from start to finish.— Awards4U

— case 03 Bush Wonderers safaris + adventure travel 6 months
the brief

a Uganda-based operator running walking safaris and adventure trips, competing against aggregators for high-intent trip searches.

what we did
  • optimised 20+ key trip and destination pages
  • worked 45+ target keywords across trip types and seasons
  • earned 100+ backlinks while keeping the technical layer clean
18+
keywords in top 10
20+
pages optimised
45+
keywords targeted
100+
backlinks built

They kept the technical SEO side organised. A smooth experience from start to finish.— Bush Wonderers

02. how we think about search

four operating rules every engagement runs on.

we charge a flat fee, not a rate per page — so when we say "publish less here," we mean it. our incentive is your organic revenue, not your invoice.

01
— rule 01

every page needs a query. write it on the brief.

a comparison page is for evaluation. a money page is for capture. a cluster hub is for coverage. an faq block is for citation. if you can't say in one sentence what a page is for, you're spending without buying. we write the thesis up front, and pace against it.

— what we audit first
cluster mix vs. role — how much of your output is doing capture work vs. demand-creation. usually the ratio is wrong by half.
02
— rule 02

intent beats volume, every time.

the engines have solved retrieval. they haven't solved which answer deserves the click. we ship 8–12 pages a month written for a named query, not for a word count — with one or two clusters expanded every week. intent is the lever; everything else is the floor.

— what we make
money pages, comparison pages, cluster hubs, faq blocks, schema. written for the query, structured for the crawler, checked against the serp.
03
— rule 03

measure revenue, not sessions.

session growth is a story. organic revenue is the truth. we track the incremental revenue from the next cluster published, not the blended report card. and we run quarterly holdout clusters — not modelled, real — to know what the work is actually adding vs. what would have ranked anyway.

— what's in the read
non-brand clicks by cluster, organic revenue vs. forecast, position retention, brand-vs-non-brand split, cohort payback curves.
04
— rule 04

refresh before you publish more. always.

every cluster and page has a pre-agreed kill threshold: crawl window, index window, ranking window. if it doesn't clear, it's consolidated — the effort moves. most seo programs die from optimism, not bad ideas. we consolidate quickly so we can publish on what's working.

— what gets killed
pages stuck past position 20 after a full refresh, clusters that miss the revenue target two quarters running, links below the relevance floor. no "give it another quarter" without a written reason.
03. the search scoreboard

every layer, given a job and a number.

this is the operating doc we run for every account — one row per layer, the role it plays, the kpi we hold it to, the benchmark, and the effort share. no layer sits on the page without a job and a number it has to clear.

— layer
role
primary kpi
benchmark
effort share
status
technical seocrawl, speed, schema
foundation
indexed ratio+ core web vitals
≥ 95%
15%
hold
on-page + contentmoney pages, clusters
capture
non-brand clicks+ page-1 keywords
≥ 40%
30%
scale
local seoprofile, citations, reviews
proximity
map-pack visibility+ direction requests
top 3
12%
scale
link acquisitiondigital pr, partnerships
authority
referring domains+ topical relevance
+8 / mo
18%
hold
answer enginesllms.txt, faq schema
citation
ai citation share+ branded lift
≥ 25%
8%
test
ecommerce seofeeds, facets, pdp
revenue
organic revenueafter returns
≥ 3.1×
12%
scale
content refreshdecay, cannibalisation
defence
position retentionvs. 90-day baseline
≥ 90%
5%
hold
7
layers with a stated job
100%
clusters covered, no orphans
1.8×
avg organic revenue vs. forecast
3:1
refresh-to-new ratio held
04. how we wire it up

six weeks from crawl to first ranking movement.

we don't rewrite the site on day one. we crawl, we model the clusters, we fix what is blocking indexation, then we publish against a narrow first cohort — and scale once the read is honest.

01
— week 01–02audit + thesis

what's working, what's burning, what's missing.

we crawl the whole site, strip out branded queries, run a marginal-efficiency curve on each line, and write the cluster thesis — what each platform is for, what the kpi is, what the kill threshold is. delivered as a one-page board doc, not a 60-slide deck.

crawl audit cluster thesis scoreboard v1
02
— week 02–03content engine

the first 30 pages, on a cluster framework.

we don't ship a brand deck as a page. we ship 30 pages across formats — money pages, comparisons, cluster hubs, faq blocks — built off a cluster framework (problem-first, comparison, demonstration, proof, list). every page gets a tracked id so we can see which angle pays back.

cluster framework 30 page briefs tracking taxonomy
03
— week 03–04measurement layer

your numbers, in one place, daily.

we set up a single source-of-truth dashboard that pulls search console + your shopify/stripe/ga4 + a revenue-per-cluster model on top. by friday, you can open one tab and see what's climbing, what's slipping, what shipped today, and which cluster is breaking the line.

unified dashboard revenue-per-cluster model content tracker
04
— week 04go live, small

a narrow first cohort, learning mode.

we don't ramp on day one. we go live at half your usual output for two weeks — just enough to validate the model, calibrate the kill thresholds, and see which creatives have legs. then we scale into what's working, not what was working last quarter.

learning cohort kill calibration scale plan
05
— week 05–06scale + cadence

the weekly read becomes the rhythm.

monday: weekly read — positions, organic revenue, refresh signals, three calls (scale / hold / consolidate). thursday: content shipout — new pages live, thin pages merged. monthly: cluster holdout review, scoreboard v2. by week six, the program runs on rails — and you stop refreshing the rank tracker at midnight.

monday read thursday ship monthly board

four hundred pages ranked. flat-fee, every month.

— 01 / under management
400+
pages ranked across the portfolio in the last 12 months — flat-fee, no per-article billing
— 02 / output
~10
pages per account per month — written for a named query, not retrofitted from a blog calendar
— 03 / efficiency
+38%
avg organic revenue lift in first 90 days vs. previous engagement
— 04 / discipline
~17%
of accounts have recommended publishing less in the last quarter
"

first agency that's ever told us to publish less. and revenue went up.

D
deepak r.
— head of growth · dtc supplements
05. faq

questions, answered straight.

can't find it here? email hi@socialmafia.agency — replies in under 24 hours.

do you charge per article?
no — ever. we charge a flat monthly fee scaled to scope (cluster count, technical debt, link targets). paying per page puts our incentive against yours; we'd never recommend publishing less because we'd earn less. flat fee fixes that.
what's the minimum engagement?
AED 1,250/month on the foundation tier — growth is AED 2,499 and scale AED 4,999, all on a three-month minimum. below the entry tier, the technical work, content engine and weekly read aren't worth the overhead — you'd be better off with a freelance buyer + good creative templates. we'll happily refer you.
who owns the site access and the content?
your properties, your name on every contract, your billing. we work inside your search console, analytics and cms with vendor access — never our own. when the engagement ends, you keep every campaign, every audience, every learning. zero handover friction by design.
how do you measure seo when attribution is messy?
we run a three-layer read: rank-tracked (with appropriate skepticism), organic revenue vs. forecast (the truth), and quarterly holdouts on refreshed clusters (the validation). we don't rely on rank-tracker averages alone — that number is a marketing aid, not a financial one.
who writes the content? do we need a separate content agency?
we do the work ourselves. our search team includes a technical lead, editors, and a digital pr lead — built for performance, not brand campaigns. if you have a brand agency, we work alongside them: they own the voice, we own the pages that have to rank in it.
how fast do you typically see results?
technical fixes and the content engine take 4–6 weeks to stabilise. the read is honest by week six. material ranking shifts usually show inside 90 days. if nothing has moved by month four, the strategy is wrong — and we'll tell you, not paper over it.
which search surfaces do you actually work on?
google and bing organic, google maps and the local pack, google shopping feeds, youtube search, app store search where relevant, and ai answer engines. we'll refuse to chase queries that don't fit your category — so don't expect a wishlist. expect a thesis.

stop reporting traffic. start ranking for intent.

— flat fee · scope-agnostic · weekly read · technical first