Case study

Ten hours an article became about 35 minutes.

An agent team that researches, writes, and publishes SEO content in one click

Book your free strategy call Read the strategy

~35 minhuman time per published article, down from ~10 hours

Client: Deployed on this very website
Engagement: Multi-agent content system
Led by: Prasad Iyer

The outcome

Where it started. Where it is now.

MeasureBeforeAfter

Human time per published article~10 hours~35 min

Content shipped per month4 articles, 40 hours of coordination16×same team

Approved draft to live URLCopy, paste and re-format in the CMS1 click

The situation

The writing was not the bottleneck. The handoffs were.

Content programs die in the handoffs: research in one tab, briefs in a doc, drafts in another tool, publishing in a CMS. Forty hours of coordination per month for four articles.

What we found

01
Research lived in one tab, briefs in a doc, drafts in another tool and publishing in a CMS.
02
Every move between them was a person copying, pasting and re-formatting.
03
Four articles a month cost forty hours of coordination, before anyone judged whether the article was good.

Researchbrowser · 20 tabs

copy, paste, re-format

Briefa shared doc

copy, paste, re-format

Drafta writing tool

copy, paste, re-format

Publishthe CMS

4 articles a month40 hours of coordination

Schematic of the old workflow · tool names are generic

The strategy

Give each handoff to an agent. Keep the judgement human.

We did not try to build one model that writes articles. We split the job the way a good content team already splits it, gave each step to an agent built for that step, and joined them so the output of one is the input of the next. A person stays at the one point where judgement matters, the review gate, and nothing goes live without them.

Starting point
The writing was not the bottleneck.
1
Agent 01 · Research
A research agent reads the whole results page
- Live SERPs
- People Also Ask
- Reddit questions
2
Agent 02 · Strategy
A strategy agent picks the angle that can win
- Keywords clustered by intent
- Angle chosen against page one
3
Agent 03 · Writing
A writing agent drafts in the site’s own voice
- House tone
- Internal links placed
- Schema markup included
4
Gate · Review
A person approves or edits. Nothing skips this.
- Approve, edit or reject
- No auto-publish path
5
Output · Publish
One click from approved to live
- Meta tags
- OG image
- Sitemap ping
Outcome
~35 min
human time per published article, down from ~10 hours

Execution

What we did, move by move.

Each move below is a piece of the strategy above, with a drawing of the work as it ran.

Agent 01 · Research

A research agent reads the whole results page

For each queued topic it pulls the live search results, People Also Ask questions, Reddit threads and the gaps competitors leave, and hands over one research pack instead of twenty open tabs.

Live SERPs
People Also Ask
Reddit questions
Competitor gaps

Research agent · topic queuedReading

how to choose a crm for a small team

SERPResult 1 · competitor-a.comThe complete guide to choosing a CRM
SERPResult 2 · competitor-b.comTop CRMs compared, feature by feature
PAAPeople also askWhat does a CRM cost for a small team?
PAAPeople also askHow long does CRM setup take?
RedditReddit · small business forumWhich CRM did you regret buying?
RedditReddit · sales forumMoved off spreadsheets, worth it?
GapGap in page oneNo result shows real setup time
GapGap in page oneNobody prices a five-person team

Research pack

What page one covers
Questions people ask
What buyers ask on Reddit
Gaps competitors leave

to the strategy agent

Illustration of the method · topic and sites are placeholders

Agent 02 · Strategy

A strategy agent picks the angle that can win

It clusters the keywords by intent and chooses the angle the current results are missing. An article that says what page one already says does not rank, so this step decides whether the piece is worth writing.

Keywords clustered by intent
Angle chosen against page one

Strategy agent · intentClustering

Unsorted

best crm small teamwhat is a crmcrm vs spreadsheetcrm pricingcrm setup stepscompetitor a alternativesaffordable crmcrm for beginnerscompetitor a vs b

Informational

Page one: What a CRM is

Commercial

Page one: Top 10 lists

Comparison

Page one: Feature tables

Angle chosenReal setup time and cost for a five-person teamMissing from page one

Illustration of the method · keywords are placeholders

Agent 03 · Writing

A writing agent drafts in the site’s own voice

The draft arrives with the site’s tone, internal links to the right pages and schema markup already in place, so the editor reviews an article, not a rough draft that still needs an hour of formatting.

House tone
Internal links placed
Schema markup included

Writing agent · draftReady for review

Choosing a CRM for a small team

Most guides list features. A small team needs two answers first: what it costs at your size, shown on our pricing page, and how long setup takes before anyone uses it.

Our setup walkthrough covers the first week. If you are weighing two tools, the side-by-side comparison puts the trade-offs in one table.

Schema · JSON-LDArticleFAQPage

{"@type": "Article""headline": "Choosing a CRM for a small team""author": { "@type": "Organization" }"publisher": "yourbrand.com"}

Illustration · yourbrand.com and the copy are placeholders

Gate · Review

A person approves or edits. Nothing skips this.

Every draft stops at a review gate. An editor approves it, edits it or sends it back. The system has no path to the live site that avoids this step.

Approve, edit or reject
No auto-publish path

Review queue · editor Gate on

Choosing a CRM for a small teamGuide
CRM vs spreadsheet: when to switchComparison
Setup checklist for week oneHow-to

Research Strategy Writing

Choosing a CRM for a small team

Most guides list features. A small team needs two answers first…

ApproveEditSend back

Agents EditorLive

No publish path skips review

Illustration · draft titles are placeholders

Output · Publish

One click from approved to live

Approval publishes straight to the site with meta tags, Open Graph image and a sitemap ping handled. The CMS copy-paste step no longer exists.

Meta tags
OG image
Sitemap ping

Publish run · approved draftApproved

Publish

Meta title and description
Open Graph image
Sitemap ping

Choosing a CRM for a small team | yourbrandyourbrand.com › blog

What it costs at your size and how long setup takes.

YOURBRAND.COMChoosing a CRM for a small team

yourbrand.com/blog/choosing-a-crmPending

Illustration · yourbrand.com is a placeholder

How it ran

The engagement, in order.

1
Step 1
Map the workflow
Every step from topic to live URL written down, with who touches it.
2
Step 2
Build the agents
Research, strategy and writing agents, each tested on its own step.
3
Step 3
Install the gate
The human review gate and the one-click publish to the live site.
4
Step 4
Run it here
Deployed on this website. About 35 minutes of human time per article.

The evidence

Where the hours went, before and after.

Human time per published article. The agents take the hand-offs; the editor keeps the judgement.

Human time per published article

Drawn to scale

Before~10 hours

Every step done and moved by hand

After~35 min

Editor’s time only

01Research agentAgent
02Strategy agentAgent
03Writing agentAgent
04Review gateHuman
05One-click publishOne click

16×
more content shipped per month, same team: 1 click
from approved draft to live URL

What carries over

Three things to take into your own business.

01
Automate the handoffs first
Most content hours go to moving work between tools. That is the cheapest time to give back.
02
One agent per job beats one big prompt
Small agents with one clear task are easier to test, fix and trust than a single model doing everything.
03
Keep a person at the gate
Speed is only worth having if nothing embarrassing goes live. The review gate is not optional.

The service behind this

AI content engine

See how we run this as a repeatable engagement, not a one-off.

View the service

Next case study

An outbound system that finds ideal prospects and books them onto your calendar

Running for TechDevs and 3 client sales teams

Read it

Who led it

Prasad runs this kind of work.

Co-founder · CTO, AI & operations

Prasad Iyer

Runs engineering and the AI division. Builds the systems the marketing runs on: multi-agent content pipelines, outbound booking agents and operations dashboards.

What happens next

Answer a few quick questions about your business. About two minutes.
Pick a time that suits you for a free 30-minute call with a founder.
We study your site before the call, so you leave with a plan, not a pitch.

Book your free strategy call

Free, no obligation. Your answers save as you go, so you can stop and pick up where you left off.

Ten hours an article became about 35 minutes.

Where it started. Where it is now.

The writing was not the bottleneck. The handoffs were.

Give each handoff to an agent. Keep the judgement human.

The writing was not the bottleneck.

A research agent reads the whole results page

A strategy agent picks the angle that can win

A writing agent drafts in the site’s own voice

A person approves or edits. Nothing skips this.

One click from approved to live

What we did, move by move.

A research agent reads the whole results page

A strategy agent picks the angle that can win

A writing agent drafts in the site’s own voice

Choosing a CRM for a small team

A person approves or edits. Nothing skips this.

Choosing a CRM for a small team

One click from approved to live

The engagement, in order.

Map the workflow

Build the agents

Install the gate

Run it here

Where the hours went, before and after.

Three things to take into your own business.

Automate the handoffs first

One agent per job beats one big prompt

Keep a person at the gate

AI content engine

An outbound system that finds ideal prospects and books them onto your calendar

Prasad runs this kind of work.

Prasad Iyer