How do you actually know if your marketing is working?

Most business owners answer this question with a feeling, not a number. Leads feel slower than last quarter, or the phone feels quieter, so the assumption is that marketing stopped working. Sometimes that's true. More often the real problem is that nobody set up a way to know either way, so every conversation about marketing performance turns into an argument about impressions and gut instinct instead of data. Here is the checklist we walk every new client through before we touch a single ad or article: the four numbers that actually answer the question, the tracking stack that makes them visible in about five minutes, and the specific red flags that mean something is genuinely broken, whether that's the strategy or the agency running it.

Feeling Isn't A MetricMeasuring What Works

The metrics that lie to you first.

Every marketing dashboard defaults to the numbers that are easiest to report, not the ones that prove anything. Impressions, reach, follower growth, engagement rate, and even raw website traffic all climb easily and all feel like progress. None of them reliably correlate with revenue on their own. A Facebook post can reach ten thousand people and produce zero calls. A website can double its traffic because of a viral unrelated post and never move the lead count. These numbers aren't useless, they're just answering a different question than the one you're actually asking, which is whether the money going into marketing is coming back out as customers.

This is the exact gap agencies exploit, intentionally or not. A monthly report full of green upward arrows on reach and impressions feels like proof of a job well done, and most business owners don't have a competing number in front of them to argue with. We've inherited client accounts where the previous agency's reporting was airtight on vanity metrics and completely silent on lead volume or cost per lead, because those numbers would have exposed a campaign that had been flat for six months. That isn't a rare story. It's the most common failure mode in local business marketing.

Four questions, in this order.

Strip away the dashboard noise and there are only four questions worth asking, and they need to be asked in this sequence because each one only means something in light of the one before it. Is lead volume actually growing. Is the cost to generate each lead flat, falling, or quietly climbing. Of the leads coming in, how many turn into paying customers. And can you actually trace those leads back to the channel that produced them, or are you guessing. Skip any one of these and the answer to "is marketing working" becomes a coin flip dressed up as an analysis.

Signal one: real lead volume, not traffic.

Lead volume means a phone call, a form fill, a text, or a booked appointment, something a real person did that shows intent to buy. It does not mean a session in Google Analytics. A local service business with a healthy landing page typically converts somewhere between 3 and 8 percent of visitors into a lead, so a doubling of traffic that doesn't move the lead count usually means the traffic is the wrong audience, the page isn't asking clearly enough for the lead, or the form is too long. Chasing traffic as a proxy for leads is one of the most common ways businesses convince themselves marketing is working when it isn't.

The fix is simple to describe and easy to skip: track leads, not sessions, as the primary weekly number. Pull it from your CRM or call log, not from a platform's own dashboard, which will always show you the metric that makes the platform look good. If you run Meta ads, the platform's "leads" count includes people who clicked a lead form and abandoned it halfway through. Your CRM only counts the ones who actually finished. Those two numbers can differ by 30 percent or more, and the CRM number is the only one that matters.

Signal two: cost per lead, trending which way.

Lead volume without cost is only half the picture. Ten more leads a month sounds like growth until you find out you tripled ad spend to get them. Cost per lead is the number that tells you whether growth is efficient or just expensive. Ranges vary a lot by category, in home services and auto repair we typically see $20 to $55 per lead on Meta and $35 to $90 on Google Search, with competitive B2B and legal categories running two to four times higher. The specific number matters less than the direction it's moving.

A cost per lead that climbs steadily for two or three months without anyone flagging it is the single most common silent failure we find when auditing a new account. It usually means the campaign has gone stale, the audience has saturated, or a competitor entered the auction and nobody adjusted the strategy in response. This is exactly the kind of drift that a monthly report built around reach and impressions will never surface, because those numbers can keep climbing right alongside a cost per lead that's quietly eating the account's profitability.

Signal three: close rate and show rate.

This is the number almost nobody tracks, and it's often the one that actually explains why marketing "feels" like it isn't working. A campaign can hand off plenty of leads at a reasonable cost, and if the sales process is losing half of them to no-shows or the front desk is slow to follow up, the business owner experiences that as "marketing isn't working" when the real gap is on the other side of the handoff. We've seen accounts where lead volume and cost per lead both looked strong, and the actual problem was a 48-hour average response time to new leads, which tanks close rate regardless of how good the marketing was.

Track two numbers here: show rate, the percentage of booked leads who actually show up or answer, and close rate, the percentage of leads that become paying customers. If close rate is falling while lead volume is rising, that's rarely a marketing problem. It's usually a lead quality issue worth diagnosing separately, or a sales process issue that no amount of better targeting will fix.

Signal four: what customers say vs what actually happened.

Ask a new customer how they found you and you'll get an answer, but it won't always be the accurate one. Customers regularly credit word of mouth when a Google search actually reminded them of the referral, or say "I saw you online" without remembering which platform. Treat the CRM's source field as directional, not gospel, and back it up with systems that record what actually happened instead of what a customer recalls three days later.

Two tools do most of this work. Call tracking numbers, a different phone number for each channel that forwards to the same line, tell you definitively which channel drove each call. UTM parameters on every link in every ad, email, and post tell your analytics exactly which campaign a form fill came from. Without these, a business is left reconciling a CRM full of "not sure" and "saw an ad somewhere" entries, which makes any real attribution conversation impossible. Our lead generation channel breakdown goes deeper on ranking channels by real speed to first lead once attribution is actually working.

The right timeframe for each channel.

Judging every channel on the same calendar is the fastest way to draw the wrong conclusion. Meta and Google direct response ads should produce measurable results, real leads at a readable cost per lead, within 24 to 48 hours of launch, and you should have a confident read on performance by day 14. If a paid campaign hasn't shown any lead activity after two weeks, that's a real problem worth escalating immediately, not something to wait out.

SEO runs on a completely different clock. Meaningful ranking movement takes three to four months at minimum, with real lead volume typically showing up between month six and month twelve. Judging an SEO investment on a 30-day window will always look like failure, because the channel simply doesn't move that fast, and no honest agency will tell you otherwise. Referral and review-driven leads are slower still to measure cleanly, since they compound over months as review volume and word of mouth build.

Old-school channels, direct mail, coupon mailers, radio spots, print ads, billboards, don't belong in this conversation at all, because they can't be measured with any of the tools above. There's no call tracking number, no UTM, no attribution path that survives someone seeing a billboard and calling three days later from a different device. That's the old playbook: untrackable, and it should be skipped in favor of channels you can actually measure, not because it never generates any awareness, but because you'll never know if it did.

The five-minute tracking stack.

You don't need an expensive analytics suite to answer this question. You need four pieces working together, and most small businesses already have access to at least two of them for free. Google Analytics 4 and Search Console cover traffic and organic visibility. A call tracking service, priced roughly $30 to $75 a month for a single-location business, assigns trackable numbers per channel and records every call. A CRM or even a well-maintained spreadsheet with a mandatory source field captures what happens after the lead comes in, all the way through close. And UTM parameters, free and built into every major ad platform, tag every link so a form fill or call can be traced to the exact campaign that produced it.

Once those four pieces are in place, a weekly check takes about five minutes: pull lead count, cost per lead, and close rate for the week, glance at source attribution, done. This is the same setup we run across the client accounts we manage, including the tracking stack behind Network Automotive, a multi-location Arizona auto repair group where we run content and local SEO, and Network Collision, a Gilbert, Arizona collision center where we run Google Search, Local Services Ads, and content together. Neither business guesses whether marketing is working. They look at four numbers on a Monday morning.

Red flags your marketing isn't working.

Some of these point at the strategy. Some point at the agency running it. Either way, they're worth taking seriously the moment you notice one.

• The monthly report only covers reach, impressions, and engagement. None of those connect to revenue on their own, and an agency that leans on them exclusively is usually protecting a number they don't want to show you.
• Cost per lead has climbed for three straight months and nobody flagged it or proposed a fix before you asked.
• Your CRM's source field is mostly blank or "unknown." Without it, no one, including your own team, can actually answer whether a specific channel is working.
• You're being asked to judge SEO on a 30-day window, or asked to keep funding a paid campaign with zero leads past the two-week mark. Both are timeline mismatches, not automatically bad performance, but both deserve a real conversation.
• Nobody can tell you the close rate. A business that knows its lead count but not its close rate only has half the picture and is flying blind on whether those leads are worth anything.
• An agency won't give you direct access to your own ad accounts or analytics. That's a control tactic, not a data limitation, and it should end the relationship on its own.

None of these mean the sky is falling. Cost per lead drifts sometimes for reasons everyone can explain, and a slow month happens. The pattern that actually matters is whether anyone is watching these numbers at all, and whether the answer to "is this working" is a specific number or a shrug. If your current setup, in-house or agency-run, can't produce the four signals above on demand, that's the real problem, not any single bad month.

Frequently asked
questions.

What is the single best metric for knowing if marketing is working?

There isn't one. Lead volume tells you if interest is growing, cost per lead tells you if that growth is efficient, and close rate tells you if the leads are worth anything once your team gets them on the phone. Any one of these on its own can mislead you. Together they answer the question honestly.

How soon can I tell if a new marketing campaign is working?

For Meta and Google direct response ads, you should see real signal in lead volume and cost per lead within 24 to 48 hours of launch, and a confident read by day 14. For SEO, meaningful movement takes three to four months minimum, so judging it on a 30-day window will always look like it isn't working even when it is.

Why do my website traffic numbers look great but leads haven't grown?

Traffic is a vanity metric until it converts. Common causes are traffic from the wrong audience, a landing page that doesn't ask for the lead clearly, or a form that's too long. Check your visitor-to-lead conversion rate specifically, not just session count. A healthy local service page typically converts 3 to 8 percent of visitors into a lead or call.

Can I trust the source field in my CRM?

Partially. Customers misremember how they found you more often than you'd expect, especially with word of mouth versus a Google search that reminded them of a referral. Treat the CRM source field as directional, and cross-check it against call tracking numbers and UTM-tagged links, which record what actually happened instead of what the customer recalls.

My agency's reports always look good. How do I know if that's real?

Ask for the same four numbers this article covers: lead volume, cost per lead, close rate, and source-level attribution, in your own CRM and analytics, not just the platform's dashboard. If an agency's reporting leans on reach, impressions, and engagement rate without connecting to leads or revenue, that's the report of an agency protecting itself, not one proving results.

Mitchell Wolfert Founder, Provision Digital. Running paid media, SEO, and funnels for local and national brands from Queen Creek, Arizona.