How to use negative keywords to stop junk enquiries

Build negative keyword lists that keep the wrong buyer out of your inbox and recognise the junk that no keyword list can ever filter.

Good lead volume and a furious sales team

The reports look healthy. Cost per lead is down. Meanwhile the person answering the phone is talking to students writing projects and job seekers and a man who wants to know how to do the job himself. Cost per sale is climbing while cost per lead falls and those two facts have the same cause.

Junk enquiries are not free. Each one consumes a reply and a follow-up and a slot in the pipeline. In a small business the cost is paid in the attention of the one or two people who can actually close.

What negative keywords cannot do

Negative keywords exclude queries. They do not exclude people. Nothing in the list can tell you a searcher has no budget or no authority or no intention of buying this year. Someone searching your exact commercial term with no money behind it will reach your ad and there is no keyword that stops them.

They also do not help where the junk is not coming from search terms. If the waste arrives through lead form extensions or display placements then the search terms report is empty and the fix sits elsewhere. Check where the enquiries originate before writing lists.

In money the work costs nothing. The cost is judgement. Every negative you add removes real demand alongside the waste and a careless list does more damage than the junk did. If you would rather the pruning sat with someone who reads these reports weekly then search advertising management is published with a fixed scope and rate. Service fees raised in India attract GST at 18% on top.

When filtering hard is the wrong move

If you sell at the value end of your market do not block the words that describe it. Cheap and low price and budget and affordable are the vocabulary of your actual buyer. Businesses copy a generic list off the internet and remove their own market.

If your account is young hold off. With only a few weeks of data you cannot yet tell which odd query converts and a theme that looks like noise is sometimes a segment. Let the search terms report accumulate and block only the obvious.

And if your sales capacity is idle then a slightly loose filter is cheaper than an empty afternoon. The calculation only inverts once enquiry volume exceeds what your team can work properly. A single salesperson handling thirty real conversations a week should filter ruthlessly. Four people with gaps in their day should not.

Building the list properly

Work in four families. Informational queries where the searcher wants to learn rather than buy. How to and DIY and tutorial and meaning and full form. Commercial mismatches where the buyer wants a different thing. Free and used and second hand and rental and wholesale if you sell retail. Audience mismatches where the searcher is not a customer at all. Jobs and salary and vacancy and internship and course and syllabus and certification. And geography. The names of cities and states you do not serve.

Mine them rather than invent them. Export the search terms report and count how often each word appears across queries that produced spend without enquiries. Words that recur across many wasted queries are the ones worth blocking as phrase negatives. A single offending query is better handled as an exact negative.

Hold the main families in shared negative lists applied at account level and keep campaign specific exclusions separate. Negative keywords do not match close variants so add plurals and the common Indian spellings and the obvious typos yourself. Then qualify on the page as well. A stated minimum order value or a budget band on the form removes the enquiries that no keyword could have caught.

How to tell it is working

Ask whoever answers the enquiries to mark each one as workable or not. That accept rate is the metric that matters and cost per workable enquiry is the one that follows from it. Both will look worse than your old cost per lead and both are real.

The metrics that flatter are total leads and cost per lead and click-through rate. All three were better when you were accepting everyone. A business that reports only on cost per lead is structurally unable to see this problem.

The common mistake

The expensive error is blocking price and cost and rate and quotation. They read like tyre kicking and in India they are among the strongest commercial signals available. Someone searching for the price of a product has moved past curiosity. Blocking that vocabulary removes the buyers nearest to a decision and the damage is invisible because you never see the queries you excluded.

The second error is adding a broad phrase negative at account level without checking what it catches. A negative on free silently kills a campaign built around free delivery. Review new account level negatives against the previous month of converting queries before you apply them.

Where the filter ends

Negative keywords are a coarse instrument and they always will be. They stop you paying for the wrong question. They cannot qualify a buyer. The last stage of filtering belongs to the landing page and the first reply and those two carry more of the load than any list.

This guide explains how something works. It is not legal, tax or financial advice, and rules change. Confirm anything regulatory with a qualified professional before you act on it.