When the portal isn’t the marketplace: overused first-party data and the function of the media agency in a vertical market

It is a truth universally acknowledged that a media owner in possession of a targeted, opted in and responsive audience must be in want of your cash. And if that audience is finite, as most are, and if ambitions for revenue growth are not, as most are not, then it follows that the same audience will be leveraged in several different ways. Understandably, this will suit the commercial interests of the media owner but may not necessarily be of similar benefit either to the consumer or to the advertiser.

For the media owner it makes perfect sense: if you could repackage the same raw material three different ways and sell it over and over again to the same consumer, why would you do anything else? For the advertiser, choice and the potential to achieve frequency across channels are extremely worthwhile and can give flexibility and additional scope to a media schedule. However, the premium usually charged for access to this audience means it also pays to have an eye on the impact that a layered approach may have on a campaign’s overall cost per goal.

This happens most notably where listings platforms operate within a well-defined vertical marketplace. Prominent examples would be travel websites, recruitment portals, property portals, and motors websites. Within each sector the competing portals are in a race to make best use of the latest advancements in tech and data so as to leverage their audience in new ways and grow revenue. The phenomenon is particularly concentrated with larger ticket purchases or key lifestage decisions like changing job, buying a car or moving home, since both the audience and the client base are limited by external factors that the portals can’t easily influence. Property portals cannot create estate agents or property developers, and they are unable to significantly alter the macro vicissitudes of the property market, so revenue growth has to come from within. Similarly, recruitment websites cannot create job growth, and while they might try to plant the idea that moving jobs is per se A Good Thing, ultimately most people aren’t quite so capricious as to career hop purely because a TV advert suggests that they should. So for these media owners, consistent increases in turnover must again come from sweating existing consumers and advertisers harder.

For the media agency, the real value of a portal’s audience is the fact that we know these people are in market and we understand their needs in considerable detail: we know what they want to buy; what they can afford to spend on it; where they would like it to be. This makes communicating with that audience an issue of handing them the right message at the right time. We know what it is they want to buy and by fantastic coincidence we have a client who is selling some of those.

Contrast this with cheaper, high-frequency channels that reach a horizontal audience: radio, Facebook, TV, programmatic display, outdoor. We can make broad assumptions about the demographics or the interests of the people we reach through these other channels, but we definitively know far less about them than portals know about their consumers. The lower price paid for these channels reflects the fact that we don’t know for sure that their audiences want to go on holiday, or more specifically that they want to fly from Luton to somewhere in the Med and stay in a four-bed apartment for six nights. So rather than tailoring a message specific to their situation and needs and handing it to them at considerable cost per pair of eyes, we’re instead cheaply bombarding consumers with several messages in hope that one of them hits home or that their cumulative effect is enough to plant a thought, turn a head or put the advertiser front of mind when the consumer’s situation changes.

To give some context, a well-bought radio campaign ought to cost maybe £X.XX per thousand impacts. A highly targeted emailer with a property portal might cost you £X,XXX per thousand recipients. Cost notwithstanding, such email distributions do still have their place, because the available targeting, the quality of the portal consumer and their propensity to be interested in the right message are all so strong. They are also able to deliver an in-market audience at considerable scale over a short period, which can be very useful to a tactical campaign and can’t be said of every channel.

The problem with escalating costs per goal will come when we start introducing additional frequency to the schedule through these precision-targeted channels; effectively, treating a tool which is ideal for reaching a vertical audience as though it needed the frequency and broad-brush approach of a cheaper horizontal medium. Rather than delivering a right-time, right-place message to the right audience, it’s easy to decide that this group of well-targeted people represents your potential audience in its entirety, and opt to pay over the odds to speak with them with frequency and at length rather than with precision and concision. It is in the commercial interests of the media owner to encourage this activity, and they are developing an increasing array of ways to help the advertiser pursue it.

With commercial needs that do not necessarily overlap with those of the advertiser, vertical platforms will be keen to sell emailers, Facebook campaigns, programmatic bursts, postal DM, SMS and more. Advertisers will be encouraged to take all options possible, with incentivised rates and the lurking suggestion that this audience represents every possible purchaser and consequently should be wrung dry at every opportunity. However as well as being extremely costly, the precision targeting applied can actually make the application of frequency redundant: if someone is genuinely in market for a product or offering like yours, but has already ignored an email, a “remail”, an SMS and a Facebook advert, then the issue is likely to be a lack of suitability rather than one of awareness. Having already been made aware of your product several times, they have probably already decided it isn’t for them.

Another consequence of this proliferation of routes to the same limited audience is that a sales rep is no longer selling just one opportunity, rather she is keen to discuss half a dozen media and how best to use them to build out an entire campaign. Crucially though, whilst this moves the media owner into the bailiwick of the media agency, the activity here differs from the function of the media agency in several ways. It probably doesn’t need saying, but I ought to declare an interest here: it’s my intention not just to suggest that over investment into one media platform might be detrimental to a campaign’s performance, but to argue that any one media owner isn’t best placed to offer cross-channel media activity or all-round marketing advocacy: the agency is. Here’s why…


The media agency trades in media but is ultimately selling expertise. The media owner can leverage expertise, but ultimately they’re selling media. This has a massive impact on what each body is trying to achieve for the advertiser.


The rapid expansion of media platforms from one area into another can mean that the front-line sales staff, and even the back-office team implementing the campaigns, aren’t as versed in their product as they might be.


A media agency doesn’t have a horse in the race. Their horse is too busy checking the going and advising on odds. Reporting will always be impartial and a failure will always be described as such.

This isn’t just a question of independence or integrity. Because reporting is an essential part of the service an agency provides, it will have an array of tools that can help to deliver it. They will also have visibility across a range of channels, rather than just reporting on one product. This provides context and reporting on a range of investments on a like-for-like basis, according to a set of pre-agreed KPIs.


Another key function of the media agency is keeping online activity safe. A media owner might be able to talk up the benefits of following their audience away from their platform and on to the wider internet, but can they explain what safeguards are in place to prevent an advert from appearing on unsavoury websites or to protect against click fraud?


Purveyors of opted-in, properly targeted first-party data can reasonably expect to charge a premium for their audience. But there are often some substantial margins being made here. The increased cost of a premium audience should be offset by the strength of the results it yields: it’s worth paying more per click if a greater proportion of those clicks result in goals or sales, or if their average basket value is commensurately higher. The overall ROI will still be higher than that achieved from cheaper, lower-quality traffic.

But if the mark up on the quality data is too high it can end up being far cheaper per goal to buy more of the lower-quality traffic and let the quantity make up for the lack of quality. This is something that Facebook does very well. Bounce rates are high and conversion rates are on the low side, but if you’re only paying a few pence per click then it’s still all to the good, because you’re getting enough cheap traffic to offset the lesser conversion rate.

So what?

The proliferation of media channels has generally muddied the waters and made what were hitherto distinct roles somewhat interchangeable between agencies of all trades and clients alike. Conversations about where social should sit or who’s best placed to run SEO abound. But as the owners of vertical audiences begin to make use of the increased availability of media channels to make best use of their audiences, it’s vital that the role of the media agency and its provision of impartial advocacy are preserved. As with many things that matter, media planning and reporting are only worth doing if they’re done properly. Professional cynicism and a consideration of each party’s own commercial agenda suggest that as disciplines and the borders between them blur, the importance of impartiality cannot be underestimated.