4 Ways to Navigate Direct Buys in Today's Age

In the past, direct buys were by and large how promotional deals got done. You or your media buying agency would call up the marketing department of a TV station, radio station, or newspaper and negotiate an ad spot with them for a price. This might end up looking like your brand showing up in a TV commercial after the main segment of the 6 o’clock news, or it could mean your brand getting promoted on the most popular local radio station for the next week.?

This process of buying negotiated, pre-packaged ad deals with websites was common when the internet was in its infancy and a few main websites had most of the clout, but it became somewhat less popular as new forms of advertising such as search ads, social ads, and programmatic advertising grew in popularity.

Increasingly with our clients, we’re seeing a push to explore negotiated, pre-packaged deals, but in new and more modern formats, such as Newsletters and Podcasts, as well as the traditional direct deals with individual publishers. Unsurprisingly, the partners trying to sell these deals do their best to make them as appealing as possible by dialing up the hype to 11. Statements such as, “We get 100k+ views every hour, and that’s just for our US Newsletter!” or “We’ve got millions of devoted listeners who are in prime buying mode and your perfect demographic profile!” get thrown around with shocking regularity.

It’s certainly easy to fall prey to the mania unless you have a consistent framework to analyze these deals. Here are 4 suggestions to increase the chances of making the right decision:

1 - Crunch the Numbers

Whenever possible, we try to use both quantitative and qualitative data to drive our decisions, and this is no exception. While the deal owner is very likely to share the qualitative data with you (e.g. “Our audience is 23-50 year old males in the finance space with household incomes between $50k-$300k”), you’ll often need to do your own homework to make sense of all the numbers. My suggestion in these scenarios is to build a flexible performance model, an example of which is below:

The orange cells are dynamic cells where you can change the values. The white cells are formula based and should be left alone. When putting together your own model, the key inputs you’ll need to consider from the get go are:

• How many people do I expect to see or hear my ad?
• What % of the people who engage with my ad do I expect to visit my website?
• What % of the people who visit my website do I expect to take the desired action (e.g. buy a product, purchase a ticket, etc.)?
• Of the people who take the desired action, how much revenue am I expecting to receive on average?

For most of these questions, the partner you’re working with should hopefully have some benchmarks (e.g. “our newsletter has a readership of 50,000 total and about 60% of them open the newsletter each week”). Once you’ve built out a model with some numbers and base assumptions, you’re ready for step #2.

2 - Verify Your Assumptions

Before spending a dollar, you should get as much detail as possible from the partner you are considering a deal with and do everything you can to verify your forecasts. You’ll likely want to ask specifics of your partner and drill down into individual parts of your model.

For example, if your ad is going to run in the client’s newsletter on a Sunday, and they tell you upfront, “We average 10,000 readers per day”, a reasonable follow-up to ask them is “How many readers do you average on Sundays?” The same question should be asked about hourly readers or listeners if you’re running in a specific time slot for a newsletter or podcast.

Another area that’s often misunderstood by advertisers is the difference between impressions and viewable impressions, and this is especially important for website buys. A website may say, “We get 25,000 visitors to our site per day.” Upon hearing this, your first question should be, “Will my ad show up on every page of your site and thus have the potential to be visible to those 25,000 visitors, or will it only be on select pages?” From there, you can ask, “Is my ad placement going to be above the fold (i.e. visible when a user first arrives on the site/page) or below the fold?” and ultimately, “what is the average viewability of impressions on your site by placement?”

An impression on a webpage just means that someone visited the page. A viewable impression is defined by the IAB as one that “appears at least 50% on screen for more than one second.” If your ad is near the bottom of the page, and only 30% of users scroll that far, that means you only have 30% viewability. As a result, you’re missing out on 70% of the readership and potential impressions that are being promoted to you.

Other key components to understand are the quality of the audience you’ll be exposed to, how relevant they are to your brand, and how likely they are to buy. Ideally, you can get case studies or data from your partner on conversion rates for similar clients they’ve worked with in the past who have had successful buys with them. A reference from someone in your network can be even better to fill in the blanks for you based on their experience and performance. Realistically, the rate at which the audience will convert (i.e. take your desired action) is one of the hardest variables to predict, but taking the above precautions is the best way to remove ambiguity and reduce potential risk.

3 - Run the Test

The next step is to run the test, collect data, and see how the data aligns with your validation framework. I always recommend clients to run the smallest tests possible upfront, and ideally to test out a few different deals or partners. Since there are plenty of unknowns involved with these tests, it’s best not to be overconfident with your forecasts.

For example, running a $2k deal and discovering it’s a flop is far better than committing to a $20k buy and getting few if any results to show for it. In addition, spreading your budget around can help you to answer the question, “Is this deal or category of deals performant for my business?” You might run a deal with one podcast and see few results, but had you run deals with five podcasts, you might have found that two of them were massively successful for you and ones you should commit to further deals with.

4 - Analyze the Results

Once you’ve taken your results and plugged them into your model, you’ll need to determine whether the test was a success or failure. There are two ways to do this:

First, you could analyze the performance of the deal versus alternative channels or options. For example, if you receive a 2x ROAS from your direct buy on a client’s site, but a 4x ROAS on Google Ads, it’s not wrong to say that this direct buy isn’t as effective as Google Ads, and as a result you shouldn’t be taking focus away from your Google Ads approach.

A second option is for you to ask, “Was this profitable for me?” Using the example above, if your break-even ROAS is 1.5x and you’re getting 4x ROAS on Google Ads and 2x ROAS on your Direct Deal, it could still make sense to invest in both options, since you’ll make some money on both buys. This option becomes even more viable if based on your analysis, the Direct Deal unlocks a net new audience to drive awareness and conversions against.

–

If you’re seeking a partner to test, analyze, and scale new advertising channels, Stackmatix could be your solution. From pre-seed to Series C, we aim to generate qualified sales conversations through bespoke, automated sales and marketing solutions to maximize revenue and marketing return. Kick off an email thread at [email protected] for a free growth consultation to explore how we can help you to expand your business.