> ## Documentation Index
> Fetch the complete documentation index at: https://docs.synscribe.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Running the client review

> The end-to-end monthly review procedure — pull GSC, separate branded from earned traffic, build attribution evidence, and co-create next month.

The end-to-end procedure for the monthly SEO review: gather the data, separate real earned traffic
from branded noise, fill the review with attribution proof, and co-create next month. Every review
follows the same arc: overview first, then credibility, then next steps.

> 🎥 **Video:** [Prepare a good client review](https://www.loom.com/share/2faf469c941d4739870e68a1b632f2b9)

> **Companion page:** this SOP is *what you do*; [3.7 Preparing client-review material](/sops/client-review-prep)
> is *how you write the doc* you present. Gather here, write there, present.

> **Plan next month before you start.** Clients become emotionally invested once they've co-created
> the plan. Moving keywords to the following month signals what they'd lose by not continuing, and
> that's the retention mechanism. See [Part 1 §1.10 Attribution](/theory/attribution).

## Inputs checklist (before you start)

* [ ] GSC access to the client property (log in as `sam@synscribe`); see [connect: 2.1.3](/platform/connect-gsc)
* [ ] The Synscribe keyword planner for this account (for rank cross-referencing)
* [ ] PostHog / analytics access if attribution is set up; see [connect: 2.1.4](/platform/connect-posthog-mixpanel)
* [ ] Last month's review doc (to compare against)
* [ ] The full list of pages you've **delivered** this cycle (to check indexing against)
* [ ] A Notion page for this month's review (screenshots land here)

***

## Step 1 — Pull the GSC overview, separate branded from earned

* [ ] Log into the client's GSC as `sam@synscribe`; set the date range to **compare last 28 days
  vs. previous period**
* [ ] Screenshot the top-level metrics (clicks, impressions, CTR, avg. position)
* [ ] Identify **branded** terms in the query report (brand name, product name, domain name)
* [ ] Subtract branded clicks/impressions from the totals, and record the adjusted non-branded count
  as the SEO baseline
* [ ] Note whether the trend is up / flat / down vs. the previous period

> **Report the real number.** Branded searches inflate the totals and must be separated out before
> you claim anything.

> **Rently example (Mar '26):** overview showed 3.81k clicks / 50.4k impressions (up from 1.78k /
> 23.2k). But of the \~2,030 non-branded clicks noted, at least 1.5k came from branded queries like
> "Rently" / "Rently UAE", leaving roughly 530 clicks actually attributable to SEO discovery terms.
> That 530 is the number you report.

## Step 2 — Drill into blog performance

* [ ] Filter GSC pages by URL containing `/blog`
* [ ] Screenshot the blog-level summary (clicks, impressions, CTR, avg. position)
* [ ] List each post with current clicks, previous clicks, and the difference
* [ ] Cross-reference ranks in the Synscribe app to confirm which rankings are **your** content
* [ ] Flag posts that went from 0 to measurable clicks; that's the strongest proof new content is
  indexing and ranking
* [ ] If the client ever questioned content quality, spotlight the highest-ranking posts as
  counter-evidence
* [ ] Screenshot the live SERP for the top 2–3 posts

> **Rently example:** `/blog` showed 495 clicks (up from 132) and 49.2k impressions (up from 12k).
> Seven of the top nine posts went from zero clicks to measurable traffic, direct evidence that
> newly published content is working.

## Step 3 — Drill into every other page category

* [ ] List every URL segment the client has (`/uses`, `/features`, `/solutions`, landing pages…)
* [ ] Filter GSC by each category; record clicks / impressions / CTR / avg. position vs. previous
* [ ] Screenshot each category for the review deck
* [ ] Note any category with 0 / near-0 data and investigate why (indexing, thin content…); this
  feeds Step 4

> **Rently example:** `/uses` showed 17 clicks / 1.07k impressions (up from 0/0), early traction.
> But only 2 use-case pages were indexed of the full published set; pages like `/uses/12-cheque-rent`
> and `/uses/rent-loan-alternative-dubai` weren't yet in Google's index, which explained the low numbers.

## Step 4 — Verify indexing status (the process checkpoint)

* [ ] Run `site:domain.com/category` on Google for each URL category to see what's indexed
* [ ] Cross-reference indexed pages against the full list you've **delivered**
* [ ] For anything missing, find the root cause (trailing slashes, sitemap gaps, `noindex`…)
* [ ] Check the client's `sitemap.xml`; confirm all target pages are actually in it
* [ ] Manually request indexing for missing pages via GSC's URL Inspection tool; see [how-to: 2.5.3](/platform/index-gsc-sitemap) *(forward ref)*
* [ ] Set a follow-up date to re-check (typically 2–3 days)

> **Shortfalls in your deliverables are a chance to surface client-side blockers. Don't bury them;
> raise them clearly.**

> **Rently example:** the `/uses` section had a trailing-slash issue blocking indexing, and the
> sitemap at `rently-uae.com/sitemap.xml` contained only 2 URLs (homepage plus apply page), so no
> use-case or blog pages were included. Root cause was client-side: the web team hadn't given SEO
> access outside the `/uses` package to update the sitemap. Action: manually request indexing for the
> affected pages, then re-check in 2 days.

## Step 5 — Build attribution evidence

The goal is to fill the review with proof that your content ranks for the target terms.

* [ ] Export the keyword list from Synscribe and sort by current rank; refresh ranks first so the
  data is current
* [ ] Cross-reference each keyword's rank in GSC's **Performance › Queries** tab
* [ ] Search the top 10–15 keywords directly on Google; screenshot each SERP
* [ ] Highlight in every screenshot which result is the client's content
* [ ] Be transparent: if a ranking page is the client's homepage or an older page you didn't create,
  say so and do not attribute it
* [ ] Compile all screenshots into a "SERP Evidence" section of the deck

> **Rently example:** SERP screenshots confirmed blog content ranking `best rent app uae` #1,
> `rent installment plan uae` #1, `no security deposit rent uae` #1 & #3, `employer housing advance
> uae` #1, and both `keyper vs ezy` / `keyper vs rently` at #1 (a Reddit thread the only non-Rently
> result on page 1). The `rent app uae` query gave Rently three of the top three positions
> (homepage #1, Play listing #2, "3 Best Rent Apps" blog #3).

Real prompt, validate impact in the analytics tool:

```text theme={null}
find out from [client]'s [analytics tool] what growth looks like over the last couple of
months, and whether it can be attributed to the blogs and /uses pages. do a full exploration
of events and flows first to build a picture of the funnel. then illustrate how much conversion
is attributed to /uses or blog per month vs other channels (split it) — show funnel input,
output (converts), and contribution. i'm validating whether these two things had significant
impact on the org.
```

Pi builds the funnel picture first, then splits conversion contribution by content type. (See
[prompt library §5.4](/_sources/prompt-library).)

## Step 6 — Surface opportunities and gaps

Call out what could be better: incomplete deliverables, missing infrastructure, untapped potential.
Frame each as an opportunity rather than a failure, with a clear next action and owner.

* [ ] List every incomplete deliverable and why it's blocked
* [ ] Identify client-side blockers (sitemap access, CMS limits, tracking setup) and document them
* [ ] Note pillars attempted vs. planned; for live pillars, confirm they're ranking (proof of concept)
* [ ] Recommend an attribution setup if none exists (PostHog / UTM) to close the loop on what
  content drives signups

> **Rently example, three opportunities:** (1) most LP pillars not yet created (2 of 2 attempted
> ranked, but the web team hadn't granted access outside `/uses` for full deployment); (2) sitemap
> fix (only 2 URLs, all new pages needed adding, client-side); (3) no attribution layer, so no way
> to tell if SEO-driven leads convert; setting up PostHog would close the feedback loop.

## Step 7 — Establish next steps (co-create the plan)

* [ ] List next month's content pieces (topic + target keyword + URL slug)
* [ ] Confirm the cycle's output capacity covers the plan; anything that doesn't fit gets pushed to
  next cycle intentionally (this is the FOMO lever)
* [ ] Explicitly move deprioritized keywords to the following month so the client sees what's queued
* [ ] Flag head terms that would need link building (out of scope for content) so the client is
  aware; see [3.9 Link building](/sops/link-building) *(forward ref)* and [3.10 Press release](/sops/press-release)
* [ ] Schedule a check-in with the client's product/founder to surface upcoming features; new
  features become anchors for next month's keyword clusters
* [ ] Set follow-up dates for any indexing/technical fixes; confirm the attribution setup has an owner

> **Capacity framing:** a cycle covers a set number of pages, split between landing pages and blogs.
> ❓ \[needs Raymond / \[DECIDE]: standard-tier output per cycle — pricing rework in progress]

> **Rently example, April '26 next steps:** re-check indexing for the 3 manually-submitted use-case
> pages; follow up with the web team on sitemap access; plan 4 new pieces from planner keywords not
> yet targeted; discuss new product features with the founder for fresh anchors; set up an attribution
> layer for the May review.

***

## Acceptance checks

* [ ] Branded traffic subtracted: the number you report is earned, non-branded clicks
* [ ] Every published category checked for indexing; missing pages have a root cause + follow-up date
* [ ] Attribution: top 10–15 keywords have SERP screenshots with the client's result highlighted
* [ ] Nothing attributed that isn't ours (homepage / old pages flagged honestly)
* [ ] Each gap is framed as an opportunity with a next action and an owner (you vs. client)
* [ ] Next month is co-created: content pieces listed, deprioritized keywords explicitly moved forward
* [ ] The written deck follows [3.7](/sops/client-review-prep) (clusters not pages, no numbers in commentary)

## Edge cases — quick lookup

| Scenario                                  | What to do                                                                                      |
| ----------------------------------------- | ----------------------------------------------------------------------------------------------- |
| Growth looks huge but it's mostly branded | Subtract branded queries; report only the earned delta (Step 1)                                 |
| A page category shows near-zero data      | It's likely an indexing problem; go straight to Step 4 and find the root cause                  |
| Deliverables are missing / blocked        | Surface as a client-side blocker with a clear owner; don't bury it (Steps 4, 6)                 |
| A ranking page isn't yours                | Say so; do **not** attribute it (Step 5). Transparency protects credibility                     |
| Head term needs link building             | Flag as out-of-content-scope; route to [3.9](/sops/link-building) / [3.10](/sops/press-release) |
| Everything planned won't fit the cycle    | Push the overflow to next month explicitly; that's the retention lever, not a failure           |

## Appendix — strategy for mature sites

Once a site has accumulated enough first-party GSC data, shift the keyword strategy toward its own
"golden" keywords:

* [ ] In GSC Performance, sort queries by **impressions** (descending)
* [ ] Filter by **average position 4–20**, where the site is visible but not capturing clicks yet
* [ ] Identify the top \~20 by impression volume in that range
* [ ] For each, decide: does an existing page target it, or is a new page needed?
* [ ] Prioritize where a small position gain yields the largest click gain (e.g. position 5 to 3)
* [ ] Add these to next month's content plan

> **First-party GSC data beats third-party tools (Ahrefs) for this.** The progression is to spread
> out and collect first-party data first, then deepen where the data says. It's the same signal
> logic as the [SERP cliff §1.4](/theory/serp-cliff), applied to your own footprint.

## See also

* [3.7 Preparing client-review material](/sops/client-review-prep): writing the doc you present
* [3.1 Onboarding — Month 1, Week 4](/sops/onboarding-month-1#week-4-—-review-prep-&-calibration): the first review
* [Part 1 §1.10 Attribution](/theory/attribution): the attribution philosophy
* [2.6.3 Reconcile GSC + PostHog into a performance review](/platform/performance-review) *(forward ref)*