What PPC is, when it actually works, and when it doesn't.

PPC, or pay-per-click, is any form of online advertising where you pay the platform only when someone clicks your ad. Google Search Ads is the original and still the largest example. Meta, LinkedIn, Microsoft, TikTok, and Amazon all run variations of the same model. You write the ad, you set a budget and a target, the platform shows it to people, and your credit card is charged per click rather than per view.

The model sounds straightforward because the model is straightforward. What gets complicated is the auction sitting underneath it, which decides whether your click costs a dollar or twenty, and whether it ever shows up at all.

Every time someone types a query into Google, an auction runs in roughly a quarter of a second. Every advertiser bidding on that query (or a close variant) gets evaluated. The platform picks the winners based on two things: how much you bid and how relevant the platform thinks your ad and landing page are to the searcher.

That second factor is called Quality Score on Google. A well-built ad with a tight landing page can win the auction at a lower bid than a sloppier competitor willing to pay more. Google explains the mechanics in their own help documentation on Quality Score, which is one of the few platform-published resources that doesn't try to upsell you something while explaining it.

The practical takeaway: bid is not the whole story. A serious PPC operator spends as much time on ad copy, keyword grouping, and landing page relevance as on bid strategy.

• Cost per click (CPC). What the platform charges per click. Varies wildly by category. A click on "personal injury lawyer" can cost over $200. A click on "vegan cookbook" can cost a dollar.
• Click-through rate (CTR). The percentage of people shown your ad who click it. Good CTRs are 5 to 8 percent on search, well under 2 percent on display. Low CTR drags your Quality Score down and your costs up.
• Conversion rate. The percentage of clicks that turn into a defined action: a form fill, a phone call, a purchase. This is where most PPC programs die quietly. You can run great ads to a bad landing page forever and lose money the whole time.
• Cost per acquisition (CPA) or return on ad spend (ROAS). CPA for lead-gen, ROAS for ecommerce. CPA = total ad spend divided by total conversions. ROAS = revenue generated divided by ad spend. These are the only numbers your accountant cares about.
• Lifetime value (LTV). The total revenue a single customer generates over the relationship. If your LTV is $500 and your CPA is $200, you have a working channel. If your LTV is $80 and your CPA is $200, you do not, regardless of how good the ad copy is.

PPC earns its place in a budget in a few specific situations. The first is when there is real, measurable buying intent in the search bar. Someone typing "emergency plumber Phoenix" at 11 p.m. is not browsing. The conversion rate on that click can be ten times what a social ad ever produces, which is why the CPC on those keywords is also five times higher. The math still works because the buyer is ready.

The second is time-sensitive launches: a new product, a seasonal push, a geographic expansion where you have weeks rather than years to show up. SEO compounds over time. PPC turns on in an afternoon.

The third is hyper-local or hyper-targeted situations where you need to appear only to a specific zip code, a specific industry, or a specific job title. The targeting infrastructure on the modern ad platforms is good enough that a well-built campaign can find a thousand people who match your buyer profile and ignore the other 300 million on the internet.

The clearest no is when there is no search demand to capture. If you are launching a category nobody has heard of, there is nothing to bid on. The closest keyword either does not exist or has fifty searches a month, which means paid search cannot scale you. You need content, PR, or social to manufacture demand first.

The second no is when the unit economics simply do not work. If your average order value is $30 and your gross margin is 40 percent, you have $12 per sale to spend on acquisition. The reality is that in most paid channels, a CPA under $25 is rare. The math is the math.

The third no is when the same query could rank organically inside six months. Paying for a click on your own brand name when you already rank first organically is a tax most companies pay because their agency told them to. WordStream has run the numbers on this for years and the answer is nuanced: brand bidding is sometimes worth it (when a competitor is bidding on your name) and often is not (when you already dominate the SERP).

Both your agency and Google have an interest in you spending more. Google rewards spend with better support and richer data access. Most agencies charge a percentage of spend, which means every dollar you cut from your budget is a dollar cut from their revenue. Neither party is the villain here. The alignment just isn't quite where you want it.

The way to insulate yourself is to insist on CPA and ROAS as the headline numbers in every report, not click volume or impression share. A good operator will be the first to tell you to pull back when efficiency drops. A bad one will keep finding reasons to scale. Search Engine Land's paid search archive is the best free running record of how the channel has actually changed year over year, written by people who have to live with the consequences of what they publish.

If you are starting from zero on PPC and want to see whether it can be a channel, here is what an honest first quarter looks like.

• Week one: install conversion tracking properly. Almost every account we audit has tracking that is wrong, double-counting, or missing the actual money events.
• Weeks two through four: launch a small set of tightly-themed campaigns on your highest-intent keywords. Cap daily spend so a runaway bid doesn't burn the budget on day three.
• Weeks four through eight: optimize ruthlessly. Pause keywords that aren't converting. Improve the landing page for the ones that are. Add negative keywords to filter out junk traffic.
• Weeks eight through twelve: decide whether the channel works at your LTV. If CPA is under your LTV target, scale carefully. If not, kill the channel without sentiment and put the money somewhere that pays.

The honest answer about PPC: it is a real channel for businesses with high-intent search demand and durable margins. For everyone else, it is a meeting people have because they think they are supposed to.