Digital Ads 101: Goals, Budgets, Channels and Measurement

Updated September 2026 · Written and maintained by the Progression Agency strategy team

Most guides to digital advertising describe the platforms. The more useful thing to understand first is the arithmetic: what you can afford to pay for a customer, how many leads that implies, and therefore what budget the goal actually requires. Get that right and the platform choice becomes straightforward. Get it wrong and no amount of campaign optimization rescues it. This guide covers goals, audiences, channels, budgets, measurement and the mistakes that quietly consume most beginner budgets.

Digital ads 101 usually starts with the platforms. This guide starts with the arithmetic instead, because the platform choice follows from it rather than the other way round.

What is digital advertising, in plain terms?

Paying a platform to put a message in front of people it can identify, and paying either for the impression, the click or the resulting action.

That is the whole of it. Everything else — targeting, bidding, creative formats, attribution — is detail about who sees it, what you pay for, and how you know whether it worked.

The two families of digital advertising

Demand capture reaches people already looking for what you sell — search ads, shopping ads, marketplace listings. Demand creation reaches people who were not looking — paid social, display, video. They behave completely differently and are frequently judged by the same standard, which is where most disappointment comes from.

What you actually pay for

Cost per thousand impressions for awareness formats, cost per click for most search and social, and cost per lead or action in some products. The pricing model shapes the incentive: paying per click means the platform is rewarded for clicks, not for customers.

Auctions, not price lists

Almost all of it is auction-based, so what you pay depends on who else wants the same audience at the same moment. This is why costs rise in competitive seasons and why no fixed price can be quoted for a click.

Why relevance lowers cost

Platforms reward ads people engage with, because engagement makes them more money per impression. A more relevant ad and a better landing page genuinely reduce what you pay for the same position.

What are your goals, and can advertising serve them?

Advertising is good at demand capture, lead generation and direct sales, moderate at awareness, and poor at fixing an offer nobody wants.

The first question is not which platform but what has to be true for the business at the end of the quarter. ‘More inquiries from people we can serve’ leads to a different plan from ‘more people in our town know we exist’, and both are legitimate.

Write the goal as a number

Not ‘grow awareness’ but ‘thirty qualified inquiries a month at under two hundred dollars each’. A goal you cannot fail is not a goal, and it makes every later decision arguable rather than checkable.

Work out your break-even first

Gross margin per customer multiplied by the share of leads that close gives the most you can pay for a lead. Every judgment about whether advertising is working follows from that one number, and most businesses have never calculated it.

Be honest about awareness goals

Brand advertising can be worthwhile and it is rarely cleanly measurable. Anyone promising precise attribution for awareness spend is overstating what the available tools do, and treating it as an investment with fuzzy measurement is more honest than pretending otherwise.

Some goals are not advertising problems

If inquiries arrive and nobody follows up quickly, if the price is wrong, or if the product does not do what customers need, advertising will make the problem more expensive rather than solving it.

Who is your target market, and how much of that can platforms actually use?

Enough to matter and less than the interfaces imply. Platform targeting has narrowed considerably, and your own customer data is now the most valuable input you have.

The useful exercise is describing who buys from you in terms a platform can act on: where they are, what they search, what they already bought from you, and what they visited on your site. Demographic guesses are usually the least useful part.

The second and third rows matter more each year as platform-inferred targeting weakens and automated bidding takes over. Your own data is the input a competitor cannot replicate.

Describe the buyer, not the demographic

‘Homeowners aged 35-55’ describes a census category. ‘People whose water heater just failed’ describes a buyer, and the second is what search advertising can actually find.

Consent and data handling are not optional

Uploading a customer list to an advertising platform involves personal data, and rules differ by jurisdiction. Check what you are permitted to do with the contacts you hold before uploading anything.

Exclusions are targeting too

Excluding existing customers from acquisition campaigns, excluding areas you do not serve, and excluding job seekers costs nothing and improves every metric that matters.

Which channels should you use?

Start where the demand already exists. For most small businesses that means search first, then the free local channels, then paid social once something is working.

The instinct to start with paid social because it is cheap per click is understandable and usually wrong for a business that needs inquiries this quarter. Cheap clicks from people who were not looking are not cheaper customers.

Search advertising

Reaches people actively looking. Highest intent, highest cost per click, fastest to produce inquiries, and the easiest to waste without negative keywords.

Shopping and product ads

For retail, product data rather than keywords decides what you match. The feed becomes the main lever, and it is the part most often neglected.

Paid social

Reaches people who were not looking. Creative-led, fatigues quickly, better at demand creation and retargeting than at capturing urgent need.

Video

Effective for demonstration and for brand building, expensive to produce well, and hardest to attribute. Worth it when the product benefits from being seen in use.

Display and programmatic

Cheap impressions, low engagement, and easily wasted. It has a role in retargeting and a poor record as a primary channel for small budgets.

Retargeting

Advertising to people who already visited. Reports excellent returns because it reaches existing intent, and the honest question is how much of that revenue would have arrived anyway.

Marketplace and platform ads

Where people already shop — retail marketplaces, app stores, review platforms. High intent within a closed environment, and you build no owned asset.

Local service and directory ads

Pay-per-lead products for service businesses. Fast, shared with competitors, and won on response speed rather than on cleverness.

How do you set an advertising budget?

Backwards from the goal: how many customers you want, divided by your close rate, multiplied by what a lead realistically costs in your category.

That gives a required spend rather than an affordable one, and the two frequently differ. When they do, the answer is to narrow the ambition — fewer terms, tighter geography, one strong offer — rather than to spread a small budget thinly across everything.

These figures are an illustrative worked example rather than a claim about any particular business. The point is the method: a budget derived this way can be sanity-checked against margin, and a budget chosen because it felt affordable cannot.

Minimum viable budgets are real

Automated bidding needs conversion volume to learn from. A campaign producing four conversions a month cannot distinguish a good week from luck, and neither can you.

Budget for creative and landing pages

Media spend is not the whole cost. Paid social in particular consumes creative continuously, and a campaign pointed at a weak page wastes whatever it buys.

Keep a reserve rather than spending evenly

Most businesses have periods when demand spikes. Holding some budget for those beats spreading it equally across twelve months of unequal opportunity.

How do you know whether it worked?

By cost per acquisition against your break-even figure. Impressions, clicks and even cost per lead can all improve while the business gets worse.

The most common measurement failure is stopping at leads. Broadening targeting reliably produces more leads and worse ones, so an account reporting only lead volume can look like it is improving for months.

Verify conversion tracking by using it

Submit the form yourself and confirm it appears where it should. Ten minutes, and it catches the failure that invalidates every other number, including double-counting that makes results look twice as good.

Define conversions carefully

Automated bidding buys whatever you tell it to value. If a newsletter signup counts the same as a sales inquiry, the system will efficiently buy newsletter signups and be doing exactly what it was asked.

Reconcile platform numbers against reality

Every platform claims the same conversion. If three platforms report forty sales between them and you took twenty-five orders, all three are counting some of the same customers.

Track lead quality, not only volume

Recording which inquiries were viable takes minutes a week and changes what the whole program optimizes toward.

What does a landing page have to do?

Match the promise in the ad, load fast on a phone, answer the obvious objections, and make the next step obvious.

A large share of advertising outcomes is decided after the click, and the page is usually the cheapest thing to improve. Sending expensive traffic to a homepage is the fastest route to concluding that advertising does not work.

What are the most common mistakes?

No negative keywords, homepage landing pages, untested tracking, no exclusions, and judging campaigns before the learning period ends.

All five are cheap to fix and all five are present in a large share of accounts. They persist because none of them produces an error message; the campaign runs perfectly and spends the money on the wrong people.

Should you run advertising yourself or hire someone?

Yourself for a simple offer, one landing page and a small budget. Hire once spend, complexity or competition rise beyond what a few hours a month can handle.

Agency fees are commonly ten to twenty percent of spend or a flat retainer of one to five thousand dollars a month, which below roughly five thousand dollars of spend is too small to buy real attention and too large a share of the budget.

How long before you can judge a campaign?

Two to four weeks for early signal, sixty to ninety days for a fair judgment, longer where conversion volume is low or the sales cycle is long.

Judging in the first fortnight judges the learning period. Judging at six months with no interim checkpoints is expensive patience. Agree in advance what gets reviewed at thirty, sixty and ninety days.

What context should you establish before blaming the ads?

What else changed, what the competition is doing, what season it is, and whether anything downstream of the click has moved.

Campaign performance is frequently affected by things outside the ad account entirely. Diagnosing inside the platform alone leads to endless bid tinkering while the actual cause sits somewhere else.

The fourth and fifth rows are the ones most often missed, because both look like an advertising problem from inside the platform and neither is fixable there.

What should you do first, this week?

Calculate break-even, verify tracking, write down the goal as a number, and check what your landing page actually says.

None of those costs anything and all of them change what happens next. Starting a campaign without them is buying traffic and hoping, which is the default and the reason so many businesses conclude digital advertising does not work.

Product awareness

The stage before anyone is comparing anything.

Product awareness is the point at which a buyer knows a product exists and roughly what it does, and it is where most marketing budgets are misallocated in one direction or the other. Businesses with existing demand overspend on awareness when their problem is capture; businesses launching something nobody is searching for underspend on it and wonder why performance channels do not work.

The diagnostic is search volume for the category. If people are already searching for what you sell, awareness is not the constraint and capture is. If nobody is searching — because the category is new or the problem is not recognized — there is nothing to capture and awareness is the only available first move.

Frequently asked questions