1. Counting a visit as a business outcome
A page view, contact-button click, delivered message and qualified enquiry describe different stages. If a lead campaign counts an ordinary homepage view as its success action, its conversion column will not answer the sales question. Start with a written definition of the result and follow that specific route through the website and the account.
Google supports separate conversion actions for different activities. Test the actual action on the real destination, then inspect the outgoing event and the destination that receives it. A manually triggered test event can prove the tag responds, but it does not prove a customer completed the whole form or call journey.
2. Counting the same enquiry journey several times
One interaction can produce several website events: a form may prepare a message, open WhatsApp and record a quote request. Those signals can be useful for understanding the journey, but treating every signal as a separate lead inflates the commercial picture. Keep the journey events and the business outcome distinct.
Check the counting setting for each conversion action. Google offers One and Every; the appropriate choice depends on what the action represents. For a lead action, One can be useful when repeated occurrences after the same ad click should not count as multiple leads. Different conversion actions still have separate counting rules.
3. Reading only the keyword list
The keyword you add is a targeting choice; the search term is the query a person actually used. Reviewing only the keyword list can hide searches with a different intent from the service being offered. Read the search terms report alongside the ad, landing page and recorded outcome.
Group the available terms into useful demand, ambiguous demand and clearly irrelevant demand before changing targeting. The report does not expose every low-volume query because Google applies privacy thresholds. A missing term is therefore not proof that the search never happened. Keep the date range and reporting scope with each observation.
4. Adding negatives without checking intent
Negative keywords can exclude unsuitable searches, but an indiscriminate list can also block useful demand. A training provider may want a course-related search that a service agency would exclude. The decision should follow the actual offer and query, rather than a generic list copied from another account.
Review the match behaviour before saving an exclusion. Negative keywords do not use the same expansion behaviour as positive keywords, and singular, plural and related terms may need separate consideration. Keep a record of the reason for each material exclusion so the team can revisit it when the offer or customer changes.
5. Treating a selected city as proof of physical presence
Location selection and location options are separate controls. A campaign aimed at Dubai may include people who have shown interest in Dubai, depending on the chosen setting. That can be appropriate for some offers and inappropriate for others. Match the setting to who can actually buy and where the service can be delivered.
Review the campaign location option and the geographic reports together. Google distinguishes Presence from Presence or Interest, and location signals are estimates. For an agency serving UAE companies, a useful brief defines whether international decision-makers researching the UAE are relevant before changing that setting.
6. Sending a specific promise to a generic destination
An ad for a particular service should lead to a page that explains that service and carries the same next action. A visitor should not need to rediscover the offer in a large catalogue. Check the complete mobile route: headline, service detail, trust material, contact control and the handover after contact.
Google recommends aligning the landing page with the ad and keywords. A useful review also checks whether the offer can be fulfilled and whether the form, phone or WhatsApp route is available. A visually attractive page is only one part of that experience; the destination needs to help the person finish the intended task.
7. Chasing a diagnostic score instead of useful enquiries
Quality Score is a diagnostic aid, not the business result. Its component readings can help the team investigate relevance and landing-page experience, but reporting a higher score as though it were a booked customer skips the most important part of the story.
Use the diagnostic to decide which ad, keyword or destination deserves attention. Evaluate that work alongside the defined conversion action and the quality of the resulting enquiries. Google states that the displayed Quality Score is not itself an input to the ad auction. Reorganising labels alone is not a substitute for a better customer experience.
8. Changing several things and claiming a single cause
If the offer, landing page, targeting and budget all change together, a later increase cannot be assigned confidently to only one change. Record what changed, when it changed and which result will be reviewed. A period containing an outage is also a weak comparison for a stable period after a relaunch.
Where an eligible campaign and suitable traffic make it useful, a Google Ads experiment can compare a defined variation with the original. Keep the question narrow and allow for conversion lag and account conditions. The result of a small test is evidence about that test, not a universal performance promise for every business.
A practical pre-scale check
- Define the exact customer action and confirm it is the action being counted.
- Test the real website-to-contact route and record which destination receives the event.
- Review available search terms, exclusions and location options against the offer.
- Check that the ad promise and mobile landing page match.
- Keep enquiry quality and bookings separate from clicks and diagnostic scores.
- Record one clear question for the next comparison or experiment.
Questions businesses ask
Does a low cost per click mean the campaign is working?
It describes the cost of a click, not whether the person became a useful enquiry. Read it alongside the defined outcome, the destination and the quality of contact received.
Should every campaign use the same exclusions and location setting?
No. Those decisions depend on the offer, service area and customer intent. A rule that helps one account can remove relevant demand from another.
Can this checklist guarantee more sales?
No. It is a method for finding gaps and making the evidence clearer. Sales depend on the offer, market, customer journey and follow-through as well as advertising.