B2B: the second search engine is often a market's best first test

Microsoft Advertising is the only advertising platform outside LinkedIn that can target on LinkedIn profile information: company, industry and job function. Microsoft documents this itself. For a B2B business that wants to know whether a new market contains its buyers, that is an instrument Google does not offer.

On this page

Why start with the second engine

The objection comes immediately: search volume is lower there. That is accurate, and it is precisely what makes it a good testing ground rather than a poor one.

The distinction to draw is between a sales channel and a decision instrument. A channel is judged on the revenue it returns per dollar invested. A decision instrument is judged on the price of a reliable answer. A management team that confuses the two evaluates its market test with the channel's yardstick, concludes that the return is weak, and cuts before getting what it was paying for.

A market test does not need volume, it needs a reliable answer at the lowest cost. Two things favour the second engine on that point. Advertising competition is less dense, so the cost of a click is generally lower for the same query. And on this engine alone, you can learn whether the people clicking actually hold the job function you are aiming at.

That second capability changes the nature of the information you get. On an engine without that reading, a test tells you how many people searched your category in a territory. Here, it can tell you whether they were buyers, engineers, students or competitors. For a management team deciding whether to open a market, that is the difference between a volume and a signal.

There is a third argument, less often named. In a market where your brand means nothing, category queries are dominated by established competitors paying dearly to defend them. Arriving on an engine where the auction is less contested lets you get data before going to fight where bids cost more.

What the platform actually allows

Microsoft's documentation describes three targeting dimensions drawn from the professional profile: company, meaning a specific organization name, industry, such as finance or media, and job function, such as sales, accounting or purchasing.

This targeting is available on search campaigns, dynamic search campaigns, shopping campaigns, audience campaigns and performance max campaigns. It is set at the campaign or ad group level, with a bid adjustment tied to each target.

The mechanism is worth understanding, because it determines how to read the results. You are not excluding other people, you are adjusting your bid when the platform recognizes the professional attribute you are aiming at. Reach stays wide, and you pay more where the professional signal matches your buyer. That difference explains why professional targeting does not shrink your volume: it redirects your budget inside the same volume.

The limits to know before building on it

Three documented constraints, and they land badly if you learn them along the way.

A fourth limit is not documented but it shows up in practice: targeting works on the job function and not on the exact title. You reach purchasing, not the head of purchasing. For a business whose buyer is defined by a precise title, the precision will be missing and keywords will have to make up for it.

None of these limits disqualifies the approach, but each one moves the budget. An incomplete account list forces you to widen to the industry, which raises the cost per useful enquiry. A job function that is too broad makes you pay for engineers' clicks when you are looking for buyers. The price of the test is therefore set after checking inside the account, not before.

The test protocol

Four steps, over six to eight weeks, and the objective is a decision, not sales.

First, pull the industries and job functions of your last twenty won deals. That is your buyer definition, it already exists, and it is more reliable than any profile written in a workshop. Without that step, you target what you believe rather than what buys.

Next, open a search campaign on the target territory, with geographic targeting set to strict presence, otherwise your test will also measure people who are not in the market. Set a single professional dimension to start, job function or industry, not both, to keep the reading clear.

Then let it run long enough to accumulate a number of enquiries that supports a conclusion. That number is calculated before you start, from your real close rate, and it determines the budget far more than the other way round.

Finally, read the results by dimension rather than in aggregate. An average cost per enquiry says nothing. The same campaign can be excellent on one job function and disastrous on another, and it is that breakdown that tells you whether the market holds your buyers.

Before launching the test

The geographic setting that makes a test readable is detailed in your test market campaign is measuring people who are not in the market, and the full sequence in your export plan does not say where the customers will come from. The comparison with the native professional platform is in when LinkedIn's high cost is justified, and how long a test needs in how long it takes to test an ad. The minimum budget for a test is calculated in the campaign that learns before the campaign that sells.

What those inquiries become once they arrive is covered in generating leads in a market that does not know you, and the callback delay in what an unanswered inquiry costs you.

Frequently asked questions about testing a B2B market