Case study · Our own publication · Closers Weekly

We launched a brand-new website with $0 in ads. Then we threw out 68% of the traffic.

No domain history. No email list. No ad budget. In 18 days, our new sales publication reached 178 real readers who stayed close to two minutes each, and Google and ChatGPT were already sending people every day. Google Analytics said 558. We report 178, because that's how many were real people.

$0spent on ads. The only cost was an $11.25 domain.
18 daysfrom launch day to these numbers
178real readers, out of 558 visits Google Analytics reported
17readers sent by ChatGPT, averaging over 2 minutes each

Why we're showing you our own website

Before we use a method on a client's business, we test it on our own. Closers Weekly is a publication we built and run for salespeople. It isn't a client, so we can show you everything: the plan, the build, the traffic and the numbers we threw away.

The question it answers is one every owner asks: if you started from zero today, with no ad budget, could you get real people to find you? Here's exactly what happened when we tried.

The idea

A publication that sounds like the sales floor

Salespeople learn from other salespeople. Most sales content online is written by software companies selling software. We wanted a publication for working closers that helps them get better, close faster and earn more for their families.

Before writing a word, we mapped 10 topics salespeople search for, five real search terms in each, and five free tools they'd actually use. Every guide on the site traces back to that plan.

  1. Oct 17

    Launch day

    The site goes live with 20 articles, full schema, a sitemap, an RSS feed and an llms.txt file for AI tools. Analytics and Search Console are connected. A 12-question Closer Score quiz goes live and is placed inside every article. The first profiles of real, working salespeople go up.

  2. Oct 18

    Lead capture

    An application page opens where salespeople pitch their own stories, with a private dashboard that tracks every quiz taker and applicant. A 10-article series is scheduled to publish itself one morning at a time.

  3. Oct 21

    Free tools

    12 free tools go live: commission and tax calculators, a cold email grader, an objection drill and more. Each one can email people their results.

  4. Oct 31

    Search tune-up

    Nine pages sitting just off page one of Google get sharper titles, descriptions and internal links. The site is verified in Bing.

  5. Nov 3

    Job board

    A free board of 34 open closer, setter and field sales roles from 20 companies, each checked against the employer's own hiring page. No placement fees.

  6. Nov 4

    Repositioned

    Using what the first 18 days showed, the homepage now leads with the job board, plus a free weekly job alerts email.

How people found a site that didn't exist three weeks ago

No paid traffic and no boosted posts. Four things did the work.

1

Answer what people already search

Profiles of real, working salespeople answer a question people already type into Google: "who is this person?" Searches for those names put the profiles on page one, and they became the biggest source of search clicks.

2

Make it easy for AI to cite you

Every page has a clear question-and-answer layout, schema markup and an llms.txt file, so AI tools can read and quote it. ChatGPT sent 17 readers who stayed over 2 minutes on average, longer than any other source.

3

Tell Google and Bing on day one

Every new page was pushed to search engines the day it went live. Google News was already listing 29 of the site's stories two days after launch.

4

Give people something to do

The quiz, the tools and the job board give visitors a reason to act, not just read. That's where the sign-ups came from.

Where the 178 real readers came from

SourceReadersAvg time
Google search88111 sec
Social media37118 sec
Direct (typed in, bookmarks, shared links)3569 sec
ChatGPT17129 sec
Referral148 sec

Google, day by day

Shown 5,589 times in Google results and clicked 88 times in the first 17 days. From 40 showings on day two to 300 to 500 a day.

Times shown in Google0200400600Clicks0612Oct 17: shown 0 timesOct 17: 0 clicksOct 17Oct 18: shown 40 timesOct 18: 1 clicksOct 19: shown 284 timesOct 19: 8 clicksOct 19Oct 20: shown 421 timesOct 20: 5 clicksOct 21: shown 528 timesOct 21: 10 clicksOct 21Oct 22: shown 520 timesOct 22: 5 clicksOct 23: shown 477 timesOct 23: 8 clicksOct 23Oct 24: shown 383 timesOct 24: 6 clicksOct 25: shown 342 timesOct 25: 5 clicksOct 25Oct 26: shown 370 timesOct 26: 8 clicksOct 27: shown 326 timesOct 27: 3 clicksOct 27Oct 28: shown 292 timesOct 28: 7 clicksOct 29: shown 289 timesOct 29: 3 clicksOct 29Oct 30: shown 277 timesOct 30: 4 clicksOct 31: shown 250 timesOct 31: 4 clicksOct 31Nov 1: shown 390 timesNov 1: 9 clicksNov 2: shown 400 timesNov 2: 2 clicksNov 2

The part most agencies skip

Google Analytics said 558. We report 178.

Across the sites we run, a big share of "visits" come from bots, click farms and scripts. Most reports count them anyway. So we built our own filter. Every day it pulls each visit out of Google Analytics, runs it through a set of rules, and keeps only people who stayed at least 10 seconds, read two or more pages, or signed up.

178Real readers
88Quick exits
38Our own city
254Bots
  • Real readers: stayed 10+ seconds, read 2+ pages or signed up
  • Quick exits: passed the bot rules, left within 10 seconds
  • Our own city: every visit from Torrance, where we're based, so our own checks can't count
  • Bots: removed by the rules below
RuleWhat it catchesRemoved
Bot countriesCountries that send click-farm and scripted traffic across all our sites, plus visits with no known country117
Direct visits under 2 seconds"Direct" visits that leave almost instantly are nearly always bots or page checkers79
Data-center citiesCloud server towns (Amazon, Google, Microsoft) where most hits are scripts29
Visits that never loaded a pageScripts firing analytics events without opening a page24
Bot citiesCities that hit almost every site we run in the same small numbers5

Ask whoever runs your marketing one question: does your traffic report remove bots? If the answer is no, a lot of what you're paying to "grow" may not be people.

Real readers stayed and acted

The pages that held people longest, and what they did next.

Landing pageReadersAvg time
Featured closer profile 173132 sec
Homepage18121 sec
Featured closer profile 212138 sec
Featured closer profile 37137 sec
Sales job board7109 sec

97 read on a phone, 79 on a computer and 2 on a tablet.

Sign-ups in the first 18 days

6salespeople finished the full quiz and gave their contact details for a personal report
4companies submitted open sales roles within a day of the job board opening
1salesperson applied to be featured

Our own test entries are removed. Nobody was paid or asked to sign up.

Day 19

Then we used the data to reposition the whole site

The first 18 days showed what salespeople came for: who's winning, what they could earn, and where to go next. So we moved the front door. Closers Weekly now opens as a free board of remote sales jobs, with the publication behind it and a weekly email that builds a list we own.

That puts the site in front of a far bigger search market than profiles alone, and four companies had already submitted roles before the homepage even pointed to the board. Real numbers first, then the decision. That's the order we work in for clients too.

Monthly U.S. searches the new position targets

remote sales jobs18,100
remote closer jobs3,600
appointment setter jobs3,600
high ticket closer jobs1,300
door to door sales jobs720

What this means for your business

You don't run a sales publication. But the same five moves work for a painter, a law firm or a clinic, because customers find every business the same way.

Start with what customers already search

We didn't guess at topics. We built pages for questions people were already typing. For you, that's your services, your cities and the questions customers ask on the phone.

Give visitors one thing to do

A quiz, a calculator, a booking form. People who would never fill in "Contact us" will happily get a score or a price.

Be ready for ChatGPT

AI tools now send real customers, and they sent this site its most engaged readers. Clear answers, schema and an llms.txt file are how you get quoted.

Build the foundation before you buy traffic

Ads multiply whatever your website already does. When the site ranks, converts and tracks on its own, every ad dollar works harder.

Count people, not visits

Most reports would have said 558. The real number was 178. If your marketing report doesn't remove bots, you're making decisions on fake traffic.

We've used the same foundation-first method for clients. See how a painting company turned ads into signed jobs and how we rebuilt its website in 16 days.

Want to know what's real on your website?

Rob will read your site the way your customers do, check it against what Google and ChatGPT see, and show you what's costing you calls, with the fix written out.

Rob does a limited number of audits each week, so he can read every page himself.

Rob Pene
Mission Driven Brand
562-444-5620

Where the numbers come from: Google Analytics 4 for closersweekly.com, Oct 7 to Nov 3, 2026, after our bot filter and with every Torrance visit removed (the site launched Oct 17). Google Search Console, Oct 17 to Nov 2, 2026. Google News count checked Oct 19, 2026. Sign-ups from the site's lead database on Nov 4, 2026. Search volumes: Ubersuggest, U.S., July 2026. Ad spend: none.

Book a callFree audit