Google Ads Recommendations: Which to Accept and Which to Skip

Open almost any Google Ads account and a percentage sits at the top of the Recommendations page: the optimization score. Under it is a list of suggestions, each promising to lift that number by a few points. Raise your budget. Try broad match. Add these new keywords. For a busy owner running ads for a travel desk or a clinic, accepting everything can look like the responsible thing to do.

It usually isn't. Some recommendations fix real problems in a minute. Others mainly help you spend more. The skill is telling the two apart, and it takes about fifteen minutes a week.

What the optimization score measures

Google describes optimization score as an estimate of how well your account is set to perform, on a scale from 0 to 100 percent. It is calculated in real time from your account's statistics, settings and status, the likely impact of the recommendations available to you, and your recent recommendation history. It changes as you apply or dismiss recommendations, as your settings change and as new suggestions appear.

Look at what that definition leaves out. The score does not measure leads, calls, cost per enquiry or profit. An account at 100 percent can lose money, and an account at 60 percent can be profitable and well run. Google also says the score is not used in your Quality Score.

The score matters more to agencies than to you. The Google Partner badge that many agencies display requires the manager account they use for clients to keep an optimization score of at least 70 percent. Google says both applying and dismissing suggestions raise that score, so a high number does not tell you which ones were accepted. It is fair to ask your agency how they decide.

Recommendations that are usually worth accepting

These fix something broken, or fill a gap without raising what you pay per click.

• Conversion tracking problems. If Google flags a tracking problem, such as a sudden drop in recorded conversions or a bid strategy with no conversion action to learn from, fix it first. Every bidding decision rests on that data, as the guide to conversion tracking explained.
• Missing ad assets. Sitelinks, callouts and call assets are free to add and can make your ad larger and more useful. You still pay per click, and a click on a sitelink or a tap on a call asset costs the same as a click on your headline.
• Conflicting negative keywords. When a negative keyword blocks one of your own keywords, your ads quietly stop showing for it. Fix these.
• Redundant or non-serving keywords. Clean-up suggestions like these tidy the account without moving spend.

Even here, read before clicking. A suggested sitelink might point to a page you retired last month, and a "redundant" keyword might be one you kept on purpose in a separate ad group.

Recommendations to treat with suspicion

These are the ones that move money, and each deserves a proper look first.

Raise your budget

Google shows this when a campaign missed part of its potential traffic because of its budget, along with an estimate of the extra results a higher budget might bring. The estimate can be reasonable. But a campaign that runs out of budget may also be spending part of it on poor searches. Before adding money, open the search terms report and look for clicks that were never going to become customers. Blocking those often frees the budget you were about to add. Raise it only when the current spend already brings leads at a cost you are happy with.

Switch keywords to broad match

Broad match lets Google show your ad for searches it judges related to your keyword, and some of those can be far from what you sell. Google pairs it with Smart Bidding, and it tends to work better in accounts with plenty of conversion data. On a small account spending around 500 rupees a day with a handful of conversions a month, it can spread the budget thin across loosely related searches. The earlier piece on match types covers when the switch makes sense.

Change your bid strategy

Moving to Maximize conversions or a target CPA can work well once Google has enough conversion data to learn from. Accept this only when your tracking is accurate and conversions arrive steadily every week. Without that, automated bidding is learning from noise. The bidding strategies article explains what to have in place before you switch.

Add new keywords

Google suggests new keywords for searches it thinks are relevant to your business that you do not yet target. Some are useful. Others are broad terms such as "holiday packages" for a travel desk that only sells Char Dham yatras. Add the specific ones by hand and leave the rest.

Expand where your ads show

Suggestions to create a Performance Max campaign, turn on AI Max, or add Google search partners or Display Expansion change where your ads appear and who sees them. Clicks from new placements can cost less and convert worse. If you want to test one, run it as a separate campaign with its own budget, so its results stay visible on their own instead of blending into what already works.

Check your auto-apply settings

Google Ads can apply some recommendation types automatically. You choose which ones under Auto-apply settings on the Recommendations page, and it may already be on if someone who managed the account before opted in. Change history shows who turned it on. For a small account, keep the types that change spend or reach off: bidding, broad match keywords, new keywords and reach expansion such as search partners. Anything applied automatically still shows up in your change history, so you can see what changed and when.

Read your change history

Change history is the account's logbook. It records who changed what, and when: an edited budget, a paused keyword, a new ad, a recommendation applied for you. When leads suddenly drop or spend jumps, open it before anything else. Set the dates to the week the trouble started and you will often find the cause quickly.

It is just as useful when several people have access. If a freelancer, an agency and the owner all log in, change history shows the email address behind most changes, and automated changes appear under a system name. A recommendation applied by mistake can be undone from change history within 30 days. The ownership article covered why admin access should stay with the business, and this is one of the reasons.

The question to put to every suggestion is simple: will this bring more customers at a cost I can afford, or only more clicks? The optimization score cannot answer that. Your leads and your cost per lead can.

A fifteen-minute weekly routine

1. Open Recommendations and clear the repairs first, such as tracking problems and conflicting negatives.
2. Accept missing assets after checking the text and the pages they link to.
3. For anything that changes budgets, bids, match types or targeting, compare it with your weekly numbers before deciding. If cost per lead is healthy, a budget increase may be fine. If it is not, fix that first.
4. Dismiss what does not fit your business. Nothing in your campaigns changes when you do.
5. Glance at the week's change history to confirm nothing changed that you did not expect.

FAQ

Will a low optimization score hurt my ads?

Not directly. Google's help page says the score is not used in your Quality Score. The number estimates how well your account is set up, based largely on the recommendations still open in it, so judge your ads by leads and cost per lead instead.

Is it safe to dismiss recommendations?

Yes. Dismissing removes a suggestion from the list without changing your campaigns, though it does move the optimization score. Google may show it again after 28 or 60 days if the campaign is still eligible.

My agency's report shows a 95 percent score. Is that good?

It shows they have worked through Google's suggestions, by applying or dismissing them, and little more. Ask instead for leads or calls, cost per lead, and what they changed this month and why. Change history will confirm the last part.