I almost wasted $4,000 on an SEO agency. Here's the checklist that saved me.

Two years into building my site, my traffic flatlined. Not a slow decline, just a flat line that wouldn't move no matter what I published. I'd been doing SEO myself in the cracks between shipping features, and I'd clearly hit the ceiling of what I knew how to do.

So I did what most founders do when they hit that wall. I decided to pay someone smarter than me.

I posted in two communities asking for agency recommendations, and within a week I had three proposals in my inbox. All three looked sharp. All three used the exact same vocabulary: technical audit, content strategy, topical authority, link acquisition. Two of them quoted me between $3,000 and $4,000 a month. The third was cheaper, around $1,500. I almost signed with the expensive one purely because its deck was the prettiest.

Then a friend who runs a B2B SaaS company asked me a question that stopped me cold.

"Can any of them show you a client in your niche whose traffic actually went up, that you can verify yourself?"

I didn't have an answer. So I went back and asked all three.

The test that exposed two out of three

Two of them couldn't do it. One sent me a slide of anonymous logos and a "+312% organic growth" chart with no client name attached. The other said their case studies were "confidential," which is a real thing sometimes, but felt convenient given the question.

The third agency sent me a case study with an actual client name. So I pulled that client's site up in Ahrefs and looked at their traffic graph myself. The growth was real and it lined up with the timeline they claimed. That was the first proposal that survived contact with a basic fact-check.

That afternoon I realized something uncomfortable. I'd been about to spend $48,000 a year based on who had the nicest PowerPoint. Not on evidence. On vibes.

So I didn't sign with anyone that month. Instead I sat down and built a checklist, and I've run every agency and freelancer through it since. It has saved me money twice and saved me from a bad hire at least once. Here it is.

1. Make them prove revenue, not rankings

A ranking screenshot is the easiest thing in the world to cherry-pick. A keyword can sit at position one and send you zero customers if the intent is wrong or the volume is tiny.

What I ask now: "How do you tie organic traffic to actual leads or sales?" The good agencies talk about conversions, pipeline, and revenue. The weak ones get visibly uncomfortable and steer back to "rankings and impressions." That discomfort tells you everything.

2. Demand one verifiable client

Anonymous case studies are marketing, not proof. I always ask for a single client in my industry whose results I can look up myself.
When an agency gives you a real name, you can check the traffic trend in any standard SEO tool and confirm the story. When they dodge, you've learned something important for free. I'd rather lose a "confidential" agency than gain an unverifiable one.

3. Test their technical depth on the spot

This one is fun because it catches people off guard. On a call I'll ask something specific, like how they'd diagnose slow Core Web Vitals, or what they'd do about a site wasting crawl budget on thin pages.

One agency immediately pivoted to "well, the real win is content." The agency I eventually trusted gave me a dull, specific, technical answer about render-blocking scripts and indexation. Boring answers from people who actually do the work beat exciting answers from people who sell it.

4. Find out exactly how they build links

This is where the cheapest agencies hide their worst habits. Bulk guest posts on junk sites, private blog networks, and "1,000 backlinks for $99" packages can get your site penalized, and recovery takes months.

I make them name the type of sites they pitch. Real editorial links from publications people actually read will help you. If an agency is cagey about their link sources, assume the worst, because the worst is usually cheap and that's why they're cagey.

5. Ask about AI search, and watch their face

It's 2026. AI Overviews sit above normal results for a huge share of searches now, and tools like ChatGPT and Perplexity send real referral traffic. An agency still selling a pure "rank for ten blue links" playbook is solving a problem from five years ago.

I ask directly: "How do you optimize for AI Overviews and getting cited in AI answers?" The current ones explain how they structure content to get quoted. The outdated ones give me a buzzword and change the subject.

6. Read the contract before you fall for the proposal

I got this backwards the first time. I got excited about the pitch and only later noticed the 12-month lock-in with no performance benchmarks and vague monthly reporting.

A confident agency earns your business month to month. Long contracts with no exit and no clear metrics exist to protect the agency from its own results. The deliverables, reporting cadence, and an exit clause should all be in writing before any money moves.

7. Be deeply suspicious of cheap

While I was shopping, I saw "full-service SEO" offers for $200 a month. The math doesn't work. A decent SEO specialist earns more than that in a single day, so at $200 a month you're buying a couple of hours of automated tooling, not a strategy.

Cheap SEO isn't a discount. It's spending money to make nothing happen, and sometimes spending money to actively hurt your rankings with spam. I'd rather do nothing than pay for that.
What actually happened

Here's the part that surprised me. After all that research, I still didn't hire an agency right away.

The process of building the checklist taught me so much about what good SEO actually looks like that I went back and did a few obvious things myself first. I fixed the technical mess slowing my site down. I consolidated a pile of thin, overlapping posts into a few strong ones. I stopped chasing keywords nobody was searching for and started writing for real questions people asked.

My traffic recovered over the next few months. Not because of any agency, but because I finally understood the game well enough to play it. And when I did eventually bring in outside help for the parts I couldn't scale, I picked the right partner in about fifteen minutes, because I knew which questions exposed the difference between an expert and an expensive logo.

I ended up writing the whole thing up as a full guide for ventsmagazine.it.com, including the real 2026 pricing benchmarks I found and the agencies that keep showing up across independent third-party rankings. If you're staring at a few proposals right now, read the full breakdown of the best SEO companies before you sign anything.

If you're about to hire someone

Slow down for one afternoon. Ask the seven questions. Make them show you one verifiable client, test their technical depth, and read the contract before you read the pitch.

The founder who signs first and checks later usually pays twice. The one who asks hard questions first sometimes finds out they didn't need to hire anyone at all. Either way, fifteen minutes of skepticism is the cheapest SEO investment you'll ever make.