You spend eleven hundred dollars on Facebook ads over six weeks and get four phone calls. Two of them are wrong numbers. The obvious conclusion is that Facebook doesn’t work for your industry, so the budget moves to Google, and three months later you reach the same conclusion about Google.
Before you start marketing on any channel, five things have to be true, and four of them have nothing to do with marketing. A channel has exactly one job: put a person in front of your offer. Everything that happens after that arrival is yours, and when any of the five is missing, the channel still runs, still spends, and still hands you a number that looks like the channel’s fault.
Four of the five can be fixed whenever you get to them. One of them quietly stops being fixable the day you skip it, and it happens to be the cheapest thing on the list.
The Five Things That Have to Be True
- You can say what you sell in one sentence a stranger understands.
- Your site works on a phone, held in one hand, on a normal connection.
- Something is counting before the first visitor arrives.
- Google knows you exist and has confirmed you’re really you.
- A real person answers the calls and forms, quickly.
That’s the whole list. None of it is clever, and none of it is what most owners worry about when they start.
These five sit in sequence, and each one only matters if the one before it holds. A brilliant ad for a confusing offer sells nothing. A clear offer on a site that won’t load sells nothing either. It’s the same reason a break in any one part of the marketing machine wastes the other four: the parts don’t add up, they multiply, and anything multiplied by zero is zero.
Each of the five below comes with a test you can run yourself, roughly what it takes to fix, and the specific fingerprint it leaves in your numbers when it’s the thing that’s broken.
That fingerprint is the useful part. A missing prerequisite doesn’t announce itself. It shows up as a disappointing channel.
1. You Can Say What You Sell in One Sentence
The test. Ask someone who doesn’t know your business to look at your homepage for ten seconds, then close the laptop and tell you three things: what you sell, roughly what it costs, and what happens if they want it.
If they can’t answer all three, your offer isn’t clear yet, no matter how good the design is.
Most owners fail the second one. Prices scare people, so the number gets left off entirely, and a visitor with no price anchor assumes the worst and leaves. You don’t need a price list. You need a range, a starting point, or an honest sentence about what drives the number.
How long it takes to fix. An afternoon with a notepad. This costs nothing, which is why it’s so often skipped, because free work never gets scheduled.
The fingerprint. Every channel converts badly at roughly the same rate. Say your Google traffic converts at 0.4 percent, your Facebook traffic at 0.4 percent, and your referral traffic at 0.5 percent.
When the numbers are uniformly poor, the fault sits downstream of all of them, which means it’s yours, not theirs. Channels behave very differently from each other when the problem really is the channel.
Fixing this one costs you nothing but an afternoon. One of the other four costs you two weeks of waiting whether you hurry or not.
2. Your Site Works on a Phone, Held in One Hand
Every competing checklist says “make sure your website is mobile friendly”, which sounds like a matter of taste. It isn’t. Google publishes actual numbers.
The test. Run your homepage and your main service page through PageSpeed Insights, which is free. Then check the three Core Web Vitals against Google’s published thresholds:
| Metric | What it measures | Good |
|---|---|---|
| Largest Contentful Paint | How long until the main content appears | Under 2.5 seconds |
| Interaction to Next Paint | How fast the page reacts when tapped | Under 200 milliseconds |
| Cumulative Layout Shift | How much the page jumps around while loading | Under 0.1 |
Those thresholds get judged at the 75th percentile of real Chrome visits over a rolling 28 day window, so you pass a metric only when three quarters of actual visits hit the bar. Lab scores don’t count. Google’s own report card sits in Search Console, in the Core Web Vitals section, and it’s free too.
Then run the human test. Stand outside, in daylight, holding your phone in one hand, and try to find your phone number in five seconds using nothing but your thumb.
How long it takes to fix. A day to a few weeks, depending on how the site was built. This is the one that most often costs real money.
The fingerprint. Your ad platform reports far more clicks than your analytics reports sessions. That gap is people leaving before the page finished loading, and you paid for every one of them. Alongside it, mobile converts at a fraction of desktop while mobile carries most of the traffic.
3. Something Is Counting Before the First Visitor Arrives
The test. Fill in your own contact form, from your phone, on cellular data. Open Google Analytics, go to the Realtime report, and watch for the event.
If nothing appears within a couple of minutes, nothing has been recorded for anyone else either. Then check that the event is marked as a key event, because an event that isn’t marked can’t be used as a conversion.
Do the same for the phone number. A tap to call link on mobile can be tracked as an event. Calls placed from a number somebody reads off the screen and dials by hand cannot, which is why a call tracking number matters for any business that gets most of its leads by phone.
How long it takes to fix. Thirty minutes for a basic setup. Half a day if you want call tracking and clean event names.
The one you can’t fix later. Analytics starts counting the moment the tag fires, and it never fills in what came before. There’s no backfill, no reconstruction, no support ticket that recovers it.
The eight weeks you ran ads without tracking are blank permanently. So when you finally sit down to decide whether the ads worked, the only evidence you have is your memory of how busy the phone felt.
That’s what makes it different from the other four. A slow site fixed in March is just as good as a slow site fixed in January. Tracking installed in March gives you March onward and nothing else, and the decision you needed it for was the one you made in February.
The fingerprint. You hold strong opinions about which channel works and no evidence for any of them. Every conversation about the budget turns into a debate about impressions, because impressions are the only number anybody can actually see.
4. Google Knows You Exist and Believes You
The test. Open a private browsing window on your phone and search your business name plus your town. Look at the panel on the right, or the card at the top on mobile.
If it says “Own this business?” or “Claim this business”, you don’t control it. If there’s no panel at all, you don’t have one.
Verified ownership is what lets you set your hours, your service area, your phone number, and your categories. Those fields are what decide whether you appear when somebody searches for what you do rather than who you are.
Worth knowing what you’re actually holding: the profile lives on Google’s platform, under Google’s rules, and it can be suspended without warning. It’s a strong asset and a borrowed one at the same time, which is the difference between attention you rent and attention you keep.
How long it takes to fix. This is the item with someone else’s clock on it. Phone and email verification finish in minutes when Google offers them. A video recording review runs up to five business days. Most postcard codes arrive within 14 days.
You don’t get to choose the method either. Google assigns it automatically based on your business type, your public information, and your region, and a rejected video means starting over.
The fingerprint. You rank fine in the regular search results, your phone doesn’t ring, and businesses with worse websites than yours sit above you in the map results. Local visibility and organic visibility are two separate systems, and only one of them cares about your blog.
5. A Real Person Answers, Fast
Response speed reads like an operations detail. It’s a marketing input, because it changes what an identical lead is worth.
The MIT and InsideSales Lead Response Management research, led by James Oldroyd, looked at six companies, more than 15,000 leads, and more than 100,000 call attempts. Contacting a web lead within five minutes rather than thirty made that lead 21 times more likely to qualify. Same lead, same ad, same spend, and the only variable was the twenty five minutes.
A separate study by Oldroyd with Kristina McElheran and David Elkington, published in Harvard Business Review in March 2011, planted test leads with 2,241 US companies. The average first response took 42 hours, and 23 percent of the companies never responded at all.
That 21 times figure gets credited to the HBR article almost everywhere online, including by people selling software to fix it. It belongs to the MIT and InsideSales work, and the two studies are worth keeping straight because they measure different things.
The test. Have a friend submit your form and call your main number at 2pm on a Tuesday, then again at 6pm on a Saturday. Write down the minutes. Do it again next week without warning yourself.
How long it takes to fix. A week to build the habit. Longer if it needs a second person, and the honest version of that conversation is that some businesses should stop advertising until they can staff the phone.
The fingerprint. Your cost per lead looks perfectly reasonable and your cost per customer looks terrible. The ads did their job, the leads arrived, and they went cold in the inbox. This is the failure most often blamed on lead quality, and lead quality is rarely the cause.
The Order to Fix Them Before You Start Marketing
Sort by importance and you’ll start with the offer, which is correct and slow. Sort by who controls the clock and the order changes, because two of the five cost you something every day they wait.
| Order | Prerequisite | Time to fix | Who holds the clock |
|---|---|---|---|
| Start today | Google Business Profile | Minutes to 14 days | |
| Start today | Tracking installed | 30 minutes | You, but the meter is running |
| This week | Your offer in one sentence | An afternoon | You |
| This month | The phone experience | A day to a few weeks | Your developer |
| Before you spend | A person who answers | A week of habit | You |
Google Business Profile verification goes first because the waiting isn’t yours to shorten. Submit it, then go do something else while the postcard travels.
Tracking goes second because every day without it is a day you can never audit, and thirty minutes now protects months of decisions later.
The offer, the site, and the answering habit are all things you can work on while those two run in the background. None of them requires a budget, which matters, because none of the five is something an ad account can buy for you.
What It Means If All Five Are True and Nothing Works
Passing all five is real information, not just tidiness. It removes the explanations that account for most disappointing campaigns, and it turns the next question into a channel question you can answer honestly.
Three explanations are left standing:
- Wrong channel for the moment. Somebody scrolling Instagram at 9pm isn’t shopping for an emergency plumber, and somebody typing “emergency plumber near me” at 9pm is. The channel has to match the buying moment, not the audience demographics.
- Not enough time in market. Search and content take months to compound. Paid takes weeks to leave the learning phase. Judging either one at week three measures your patience, not the channel.
- Priced or positioned against somebody stronger. If three competitors offer the same thing at the same price with more reviews, the marketing is working fine and the offer is losing a fair fight.
Those three are genuinely different problems from a broken prerequisite, and telling them apart is most of the skill. It’s the same division that separates a traffic problem from the three faults more traffic makes worse.
Run the five tests this week. They take about an hour together, and an hour is a cheap price for finding out that Facebook was never the problem.
Frequently Asked Questions
What should I do before starting digital marketing?
Get five things in place: a clear offer a stranger can describe after ten seconds on your homepage, a site that loads and works on a phone, analytics and conversion tracking installed and verified, a claimed and verified Google Business Profile, and a person who answers leads quickly. Start the Google Business Profile verification and the tracking first, because one has a waiting period Google controls and the other loses data permanently every day it isn’t installed.
Do I need a website before I run ads?
For most businesses, yes. Google Ads and Meta both allow lead forms that collect a name and number without a website, and they work reasonably well for simple, low priced services. For anything a buyer wants to research first, sending paid traffic to a Facebook page or a phone number alone wastes most of the spend, because there’s nowhere for a cautious buyer to satisfy themselves that you’re real.
Why do I get clicks but no calls?
Three causes account for most of it. The page loads too slowly on mobile and people leave before it appears, so check whether your ad platform’s click count is much higher than your analytics session count. The offer doesn’t answer what it costs or what happens next, so visitors leave to compare. Or calls are arriving and nobody answers fast enough, which shows up as a reasonable cost per lead and a terrible cost per customer.
How long does Google Business Profile verification take?
It depends on the method Google assigns you, and you don’t get to pick. Phone and email verification finish in minutes. A video recording review takes up to five business days. Most postcard codes arrive within 14 days, and if one never shows up you request another and wait again. Start the process before you need the profile, not after.
Do I need Google Analytics if I already have the Facebook Pixel?
Yes, because they answer different questions. The pixel tells Facebook what happened to the traffic Facebook sent, which is what its algorithm needs to optimize. Analytics tells you what happened to all your traffic from every source in one place, which is what you need to decide where the next dollar goes. Relying only on ad platform reporting means every platform grades its own homework.
Should I fix my website first or start marketing first?
Fix the offer and the tracking first, because both are fast and cheap and the tracking can’t be recovered later. The website itself only needs to clear a functional bar before you start: it loads in a couple of seconds on a phone, the phone number is tappable without scrolling, and the contact form works. A redesign can wait, and it’ll be a better redesign once you have three months of data telling you which pages people actually use.
Previous 
