Managing monetization yourself or delegating it: how to decide

Monetizing a site is not an installation, it is maintenance. Before deciding whether to run it in house or delegate it, it helps to see the work involved, what it takes to sustain it, and what to demand from a partner. With one warning up front: some sites are better off alone.

Two routes, and neither is right for everyone

There are two ways to monetize a site with programmatic advertising. One is to run it in house: set up header bidding, onboard demand partners, keep ads.txt current and fix whatever breaks. The other is to delegate it to someone already doing that work for several publishers at once. Both work.

The decision comes down to three things: the volume you have, the time you can give it, and how much control you are willing to let go of. This is published by a company that sells the second model, so it is worth saying early: some sites are better off with a simple setup, and pushing them into anything else costs them money.

The real work behind monetizing a site

«Putting ads on the site» sounds like a one-off task. It is not: it is ongoing maintenance, and that is exactly the part almost nobody prices in when deciding.

• Maintaining header bidding. Every partner has its own adapter, configuration and timeouts; a setup that worked six months ago may be leaving bids on the table.
• Negotiating with SSPs: onboarding, terms, floors by country and format. And renegotiating when the market moves.
• Watching fill rate and viewability. Not as a monthly figure, but by slot, device and country, which is where you see what is actually failing.
• Keeping ads.txt current. A mistyped line, or a reseller still listed after the relationship ended, translates into bids that never arrive.
• Privacy compliance: consent, a management platform, signals that travel with every request. When it breaks nothing throws an error, revenue just falls.
• Fixing demand outages. A partner that stops bidding on a Monday morning does not send a notice: you have to spot it in the data.
• Testing placements and formats. Move one slot, remove another, measure for two weeks, decide. Then repeat.

None of this is hard on its own. What is hard is that it all happens at once, every week, competing with the time to write and publish.

What it takes to run it in house

Running it yourself is viable, and for plenty of publishers it is the better option. It takes three things, all three at the same time.

• Recurring time, not a project. The setup happens once; the maintenance never ends. If nobody has those hours blocked out, the system degrades on its own.
• Someone who owns it. Not necessarily full time, but one specific person who reads the dashboards and knows when to email a partner. Split across five people, nobody does it.
• Enough volume for partners to pay attention. Below a certain size the problem is not worse terms: it is that there is nobody to talk to, and your issues sit in a web form.

And the warning in the opposite direction, which is the most honest one: if your site is small, running header bidding with eight partners probably will not pay for itself. A simple setup, fast and well placed, leaves more than complex infrastructure nobody maintains.

What you gain by delegating

Delegating is not someone else pressing the same buttons. What you buy is, above all, access and scale.

• Demand you would not reach alone. Whoever aggregates inventory from many publishers reaches deals and premium demand that a single site is not in a position to negotiate.
• Direct campaigns. Beyond the open auction there are direct advertisers buying specific audiences at a fixed CPM, and that almost always arrives through someone who aggregates supply.
• Less technical load. The integration is handled with a single Header Script rather than a list of tags to maintain, and going live is measured in days: some close it with a 48-hour implementation.
• Someone watching when you are not. The day-to-day value is in someone noticing on a Monday in August that an SSP stopped bidding.

What you lose by delegating

And now the other half.

• Less direct control. You do not set every floor or pick every partner yourself: you propose and it gets agreed.
• One more layer between you and the money. Someone keeps a share for doing the work; that is legitimate, but how much and in exchange for what has to be clear.
• Dependence on someone else's reporting. You start reading what the partner reports, so how they report matters as much as what they achieve.
• The demand relationship may not be yours. If you switch partners tomorrow, do you take the deals with you or start from zero?

Delegating is not giving up understanding monetization. It is giving up executing it.

Supply Path Optimization: how many hands sit between your slot and the advertiser

Supply Path Optimization (SPO) is the industry name for something simple: buyers want to reach your inventory by the shortest route.

Between the slot on your page and the advertiser there can be a long chain: your wrapper, an SSP, another reselling into the first, a DSP and a further layer. Every link charges, so the longer the chain, the less arrives at the end. And every hop adds latency: a bid that arrives late does not compete.

There is something less obvious. When the same inventory shows up to a buyer through four different paths, the buyer sees duplicates and cannot tell which route is legitimate, and many DSPs drop the poorly identified ones. A clean ads.txt and a short chain are the difference between your inventory being bought or ignored.

When assessing a partner, that is the underlying question: does it shorten the path or add a step? One bringing direct advertisers and demand you could not reach shortens it. One reselling into the same SSP you already worked with lengthens it.

What to ask before signing with anyone

Whoever the partner is, these questions get asked before signing and the answers should come in writing.

• What is the revenue share and what is it calculated on? A percentage of gross and the same percentage of net are not the same deal. Ask what gets deducted before the split.
• Are there costs outside that percentage? Technology, ad serving, data. Make the final number the actual number.
• Is there a lock-in, and how do you exit? A partner with good terms does not need to tie you down: some work with no lock-in and propose a 30-day trial.
• Whose relationship with the demand is it? If you leave, do the accounts and deals stay with them?
• How is what you are paid audited? Is there a dashboard with data by day, by slot and by source, and can you cross-check it against your own analytics?
• Who appears in your ads.txt and why? Every line is a route into your inventory.
• What happens if something breaks on a Sunday? Who do you write to, how fast do they answer, and who fixes it.

If any of them makes the person across the table uncomfortable, you already have information. You do not need to like every answer; you need the answers to exist.

How to decide without overthinking it

Boiled down to the essentials:

• If you have volume, a dedicated person and the appetite to manage it, keep it in house.
• If you have volume but nobody with time, delegating usually pays even with a share taken out: a maintained system yields more than an abandoned one.
• If you are starting out, do not build infrastructure you cannot sustain. Prioritise speed, few well-seen slots and growing your traffic.

There is a third route that often gets forgotten: delegate the programmatic execution and keep the direct campaigns you can sell yourself.

Whatever you decide, decide it with numbers: what you earn today, how many hours a month go into maintaining it, and how many you would rather spend publishing.