How to Track Return Links in Link Exchange Deals

Start with the simplest thing that works. One row per partner, two counts, one difference. Links they gave you, links you gave them, and the balance between them. Positive means they are ahead of their commitments, negative means you are.

Set a threshold for when negative becomes a problem. Two links is a sensible default, and one link of drift is normal in an active partnership rather than a grievance. Past the threshold, the rule is that you stop publishing in their direction until something comes back.

The critical detail, and the one that quietly ruins most homemade versions: count only verified live links. If you count agreed placements, your balance flatters whichever side has made promises rather than published, which is exactly backwards from what you need to know. A partner who agreed to five and published one has given you one.

Same discipline on your own side, including links you gave from other properties or other client sites. Undercounting your own giving produces a comfortable number and a bad decision.

A straight count assumes every link is equivalent, which is fine when you and your partner are similar in authority and misleading when you are not. Five links from a strong publisher is not the same trade as five from a new blog.

So add a value calculation alongside the count, or run it in the return link balance calculator. Value received is their link count multiplied by their domain rating. Value given is your link count multiplied by your domain rating. The net is received minus given.

That asymmetry is the part people get wrong, and it is worth being explicit about. A link sitting on your site carries your authority to your partner, not theirs. So your side of the ledger uses your rating. Multiplying both sides by the partner's rating produces a number that looks rigorous and means nothing at all, and it is the single most common error we see in spreadsheet balance formulas.

Use count mode when you and your partners are broadly comparable, because it is simpler and simpler is more likely to be maintained. Switch to value mode when there is a real authority gap, because that is where count mode tells you a comfortable lie.

Understanding the mechanisms is useful because each one has a different fix, and most of them are process fixes rather than tracking ones.

The publish first problem is the most common. You agree a mutual exchange, your content calendar has a slot this week, so you go first. Their slot is next month. Then their editor changes or the piece gets deprioritized and the reciprocal link never happens. Nobody decided to shortchange you.

The multiple teammate problem scales with team size, and it is the case agencies hit first. Two people each agree exchanges with the same partner, neither aware of the other. Every individual deal looks fine and the aggregate does not. This one is invisible without a shared record, and no amount of individual diligence prevents it.

Time decay is the reason thresholds need to fire early. A return link promised in March is genuinely awkward to raise in September: the context is gone, the person you agreed with may have left, and bringing it up feels like an accusation. So most teams write it off silently, which is why reminders at 14 and 30 days matter more than any escalation process.

Then the aggregate blind spot, which is the expensive one for agencies. The same publisher exchanging with three of your clients, each relationship looking balanced, the total heavily negative. Per-partner tracking cannot see it by construction.

The point of a cadence is to make the follow up easy while it is still recent, which is the same reason a removed link is chased in week one, because a specific, timely nudge works and a vague late one does not.

At day zero, record the obligation on both sides when the deal is agreed. This is the step that makes everything else possible, and it takes thirty seconds at the moment you have the context.

At day fourteen, a first polite nudge. Two weeks is about right: sooner reads as impatient to anyone with an editorial calendar, and it is early enough that the original conversation is fresh. Reference the specific agreement and the specific link you published.

At day thirty, treat it as a problem. Pause further giving to that partner and say so plainly rather than doing it silently, because a partner who does not know you have paused cannot fix the situation. At day sixty, treat it as unlikely and flag the partner for review.

Then actually enforce the pause. This is the step teams skip, usually because the next content slot is already filled and publishing is easier than having the conversation. A threshold you never enforce is not a threshold.

A-B-C exchanges exist to avoid a direct reciprocal footprint. You link to site B, B links to site A, and no straightforward reciprocal pair appears anywhere. They complicate tracking in a specific way that is worth handling carefully.

Record each leg as its own row with its actual direction, and note the chain. The failure mode is double counting: logging the obligation once against B and once against A, then concluding that a partner owes you links they never agreed to provide. Getting this wrong produces exactly the confident, wrong number that damages a relationship.

Balance for a three way chain has to be computed per leg rather than per relationship. Your position with B is about what you gave B and what B gave you or arranged for you. Collapsing the whole chain into one balance figure loses the information you need to chase the right person.

Our honest view is that three way chains are worth the tracking overhead only if you are doing them at some volume. For a handful of exchanges the added complexity outweighs the footprint benefit, and a small number of relevant direct exchanges between real sites is not the thing that gets anyone in trouble.

The number is the easy part, and the balance calculator produces it. Telling somebody they are behind without ending a relationship you want is the part worth thinking about.

Lead with specifics. We have published four links to you since January, most recently the piece on your pricing guide, and one has come back so far, just checking where the other three sit in your calendar. That is factual and low temperature, and it is far harder to ignore than a general reminder because it gives them something concrete to respond to.

Offer an exit. Editorial priorities change, and a partner who has decided not to proceed will find it easier to tell you if you make that acceptable. If it is no longer a fit, tell me and I will stop counting on it costs you nothing and converts a slow silent failure into information.

Do not threaten removal, in a first message or ideally at all. It is hostile, it ends the relationship permanently, and removing a link you already published does nothing for your own site. Pausing is the proportionate response.

One thing worth saying about your own obligations too. If a partner removed their link because the one you promised never arrived, that is not a recovery case and no email will fix it. Publish what you owe, then ask.