All figures below are in Colombian pesos (COP).
This is piece 02 of our traffic series. The pillar guide covers the full menu of channels. This one goes deep on Google Ads only.
At PymeWebPro we build pages from $390.000 COP. A fast, well-structured page with a Pixel and proper conversion tracking is 50% of the job. The other 50% is getting the right traffic there without burning money. Google Ads, done right, is the most measurable tool in marketing. Done wrong, it's a faucet that empties your bank account in 72 hours.
If your page doesn't convert with high-intent paid traffic, no other channel will save it. Google Ads is the fastest reality check you can run.
Acronyms used in this guide
01 When Google Ads is the right channel
It's not for everyone. Three criteria decide whether to turn on the account:
Makes sense if
- You sell a service or product people are already searching for (lawyers, clinics, plumbers, hotels, software).
- Your average order value is $200.000 COP or higher, or you have strong SaaS LTV.
- You have a landing page that already converts above 2% on warm traffic.
- You can commit at least $2,5 million COP over the first 3 weeks.
Doesn't make sense if
- You're creating demand for something nobody knows exists yet.
- You sell products under $80.000 COP with thin margins.
- Your page takes longer than 3 seconds to load or has no clear CTA.
- You've got $300.000 COP in your pocket to test. That doesn't buy you data.
Left column, keep going. Right column, fix that first. Google Ads isn't magic. It amplifies what you already have.
02 The 5 campaign types, translated
Google offers five campaign types on the creation screen. Most founders pick wrong here. The honest translation:
| Type | What it does | When to use |
|---|---|---|
| Search | Text ads in search results. High intent. | Local services, B2B with clear keywords. Start here. |
| Performance Max | Black box that splits spend across Search, Display, YouTube, Gmail, Maps. | E-commerce with a product feed. Avoid if you're just starting. |
| Display | Banners on Display Network sites. | Remarketing only. As prospecting it burns money. |
| YouTube | Pre-roll and in-feed video. | When you have a good video and want awareness. |
| Shopping | Products with photo and price in results. | E-commerce with catalog and Merchant Center feed. |
Common trap
Performance Max promises one-click results. What it actually does in a new account with no conversion data is dump 60% of your budget into low-intent Display and YouTube. A new SMB account should start with pure Search, manual keywords, until you have 50+ tracked conversions. Then Performance Max starts making sense.
03 Account structure that works
The typical SMB account looks like: one campaign, one ad group, 200 mixed keywords. Result: Google averages everything, the good keywords subsidize the bad ones, and you can't see what's working.
The right structure is hierarchical. Think floors of a building:
Account
One Google Ads account per business. Don't open multiple.
Campaign
One campaign per objective + type + location. Example: "Leads · Search · Medellín". Another: "Leads · Search · Bogotá". Budget gets set at this level.
Ad group
One group per intent theme. "Dental implants", "invisible braces", "teeth whitening" are three groups, not one "dental" group.
Keywords
5 to 15 per group, all related to the theme. Phrase or exact match.
Ads
2 to 3 responsive ads per group. Each with headlines and descriptions specific to that group's theme.
This separation lets you see which service group converts best, shift budget there, and kill what doesn't work without accidentally killing the good stuff.
04 Keywords: the intent hierarchy
Not all searches are worth the same. Someone typing "what is a dental implant" is learning, not buying. Someone typing "dental implant Medellín price" is one click from paying.
The four intents
Start with the first two. Transactional and commercial searches convert 5 to 15 times better than informational ones and cost less per conversion, even when CPC is higher.
How many keywords to use
Practical rule: 5 to 15 keywords per ad group, all variations on the same intent. If your group has 80 keywords, there are three groups hiding in there.
Match type: which to use
| Match type | Coverage | Recommendation |
|---|---|---|
| Broad | Any variation Google considers related. Very wide. | Only with big budget and mature Smart Bidding. Not for new accounts. |
| Phrase | Searches that include your phrase, with words before or after. | The default to start. |
| Exact | Only identical searches or very close variants. | For your best keywords, once identified with data. |
05 Negatives that save your budget
If you read one section of this guide, read this one. Negative keywords tell Google which searches shouldn't trigger your ads. In the first 2 weeks, a good negatives list saves 30 to 50% of your budget.
Universal negatives (paste into every new account)
Those are the baseline. After that, check the Search Terms report every Monday and add niche-specific negatives. You'll see actual searches that triggered your ads, some brilliant, some absurd.
Real case
A PymeWebPro client, a traffic lawyer in Medellín, spent $480.000 COP in a week before discovering that 40% of his clicks came from "abogado tránsito empleo" (people looking for a job, not services). One negative fixed it and dropped cost per lead from $42.000 COP to $18.500 COP.
Shared negative lists
Create a negatives list at account level and apply it to every new campaign. That way you don't start from zero each time. It's under Tools → Shared Library → Negative keyword lists.
06 Minimum viable budget
The most common question: "can I test Google Ads with $300.000 COP?" The honest answer is no. You need enough click volume to reach statistically useful conclusions.
The minimum math
To make an informed decision, you need at least 30 conversions in your initial test. If your conversion rate is 3% and your average CPC is $4.000 COP:
- 30 conversions ÷ 3% = 1,000 clicks needed
- 1,000 clicks × $4.000 COP = $4 million COP for a clean answer
If your economics can't sustain that, test with a smaller budget but accept the answer will be ambiguous. PymeWebPro's practical rule:
The most expensive mistake isn't spending too much. It's spending a little for months without ever getting enough data to decide.
07 Writing ads that convert
Google Ads uses responsive search ads. You provide up to 15 headlines (30 characters each) and 4 descriptions (90 characters each), Google combines them dynamically and surfaces the combinations that perform best.
The headline formula
3 headlines with the main keyword
"Dental Implants Medellín", "Dental Clinic Medellín", "Dental Implant El Poblado".
3 headlines with a benefit
"Painless, in 1 session", "10-year guarantee", "Implant in 48 hours".
3 headlines with proof or authority
"15 years in Medellín", "+500 happy patients", "German technology".
3 headlines with CTA or urgency
"Book your consult today", "Limited May slots", "Quote in 24 hours".
3 headlines with a unique differentiator
What only you offer. No marketing-speak.
Extensions that double your CTR
Ad extensions are free and almost nobody uses them well. Turn them all on:
- Sitelinks (4 minimum): "Services", "Pricing", "About us", "Book"
- Call: WhatsApp number or landline
- Location: connect your Google Business Profile
- Structured snippets: list of services or brands you carry
- Promotion: if there's a current offer
Ads with all extensions configured typically see 40 to 80% higher CTR than bare-bones ads.
08 Conversion tracking (without this, the rest is blind)
If you're not tracking conversions properly, you're not doing Google Ads. You're making donations to Google.
What to track
Define primary conversions (the ones that actually matter) and secondary (signal-only, not the goal).
| Business type | Primary conversion | Secondary |
|---|---|---|
| Services (lawyer, dentist, plumber) | Form submitted, 60s+ call | WhatsApp click, phone click |
| E-commerce | Completed purchase | Add-to-cart, checkout started |
| B2B SaaS | Demo booked, trial started | PDF download, newsletter signup |
| Real estate | Viewing booked | WhatsApp click, brochure download |
How to set it up
- Google Tag installed across the site (included on every PymeWebPro Pro build).
- Google Analytics 4 connected to Google Ads as a conversion source.
- Enhanced conversions enabled (sends hashed email/phone when forms fire).
- Calls tracked: use Google Ads call forwarding or a service like CallRail.
Done right
An SMB with conversion tracking dialed in can know, within 30 days, exactly which keyword, ad, location, and time of day generates the most leads. Without it, you're flying blind and gambling.
09 The 8 mistakes that kill 70% of accounts
We've seen the same 8 mistakes in dozens of SMB accounts. Any one of them can burn 40% of monthly budget. Combined, they kill the account entirely.
Broad match without negatives
Google pulls clicks from searches like "[your service] free pdf course job". Fix: phrase match + universal negatives list.
No conversion tracking
Or worse, tracking "viewed more than 30 seconds" as a conversion. If you're not tracking actual purchases or form fills, you know nothing.
Sending every click to the homepage
The homepage talks about everything. The landing page for "dental implants" has to talk about implants only. One landing per ad group.
Performance Max from day 1
Without conversion data, PMax is a black hole. Start with pure Search.
Smart Bidding without data
Automated strategies (Maximize Conversions, Target CPA) need 30+ conversions to work. Before that, use Manual CPC.
Turning campaigns off on Fridays
Google Ads learns from continuity. Pausing and resuming resets learning and sends you back to the expensive phase.
Not reviewing search terms
Every Monday, 15 minutes. Search terms report. Add negatives. Identify new keywords.
Ignoring devices and dayparts
If 80% of your conversions come from mobile between 6pm and 10pm, adjust bids to favor that. It's free money.
10 Scripts and automation for SMBs
Google Ads supports scripts (JavaScript) that automate repetitive tasks. You don't need to be a programmer, pre-made scripts are available.
3 scripts worth installing right away
- Account Anomaly Detector: Emails you if spend, clicks, or CPC move more than X% versus average. Wakes you up if something goes haywire on a Saturday at 3am.
- Search Query Mining Script: Finds search terms with spend but no conversion. Candidates for negatives.
- Bid by Weather (optional): If your product depends on weather (umbrellas, rain gear, ice cream, roof cleaning), auto-adjusts bids based on the forecast.
Find them on Optmyzr, FreeAdwordsScripts.com, or Google's public scripts repo. Pasting them in takes 5 minutes.
11 When to cut and when to scale
The hardest part. After 3 to 4 weeks, you have data. These are the decision points:
| Signal | Action |
|---|---|
| CPA under your profitability ceiling and volume rising | Scale 20% per week. Faster derails learning. |
| CPA at ceiling but stable, volume flat | Hold. Optimize creative and landing before spending more. |
| CPA 30% over ceiling after 4 weeks | Pause bad campaigns, reinvest in good ones. If all are bad, shut down. |
| Zero conversions after 1,000 clicks | Page or offer problem, not an Ads problem. Fix that before spending more. |
| High conversions but junk leads | Qualification problem. Tighten keywords (more exact), add qualifiers to the ad ("from $X"). |
Reality
70% of new Google Ads accounts don't work on the first try. The difference between accounts that end up profitable and those that get abandoned is: did you iterate 2 to 3 times? Did you change the landing? Did you try different keyword types? Did you adjust the offer? People who quit after try #1 never find out whether Google Ads could have worked for them.
90-second executive summary
- Use Google Ads if you sell something people already search for, with ticket above $200.000 COP.
- Start with pure Search campaigns, manual CPC, phrase match.
- Paste in the universal negatives before the first click.
- Structure: 1 campaign per objective + location. 1 group per intent theme. 5 to 15 keywords per group.
- Budget for $2,5 to $4 million COP over 21 days as minimum validation.
- Conversions tracked properly, with Google Tag + GA4 + enhanced conversions.
- Check the search terms report every Monday. Add negatives. Bid up what converts.
- After 30 to 50 conversions, test Performance Max with a separate budget.
Your Google Ads lives and dies by your landing page.
It doesn't matter how well-tuned your campaigns are if clicks land on a slow, confusing, or non-converting page. We build sales pages for Colombian SMBs that load in under a second, ship with Google Tag and Meta Pixel pre-installed, and are designed specifically to convert paid traffic. From $390.000 COP. Request a free mockup and see what yours could look like.
Mike (English-preferring clients in Colombia) and Santiago (Spanish, Colombian market) personally review every request. Reply within 24 hours, often the same day.
Continue the series
This is piece 02 of the series. The pillar and the rest:
- 01How to drive traffic (pillar)
- 02Google Ads for SMBs · you're reading
- 03Meta Ads creative playbook
- 04Local SEO and Google Business Profile
- 05WhatsApp Business and catalog
- 06Word-of-mouth and formal referrals
- 07Email marketing for SMBs
- 08Affiliate programs from scratch
- 09Trade shows and local events
- 10Local PR without an agency
- 11AI search visibility