MGID tracking setup: passing widget and teaser IDs into your tracker

A working MGID tracking setup comes down to two jobs: get MGID's widget and teaser IDs into your tracker as sub-IDs on every click, and send conversions back to MGID by postback so its algorithm learns which placements convert. Do both and you can rank widgets by revenue per click instead of by CTR. Skip either and you're optimising on MGID's dashboard alone, which shows you clicks and spend but not what happened after the click on your offer.

This guide walks through the exact token mapping, the postback wiring, and the reporting view you need to make cut decisions on MGID widgets and teasers. It assumes you already have a campaign running and want proper attribution, not another generic "set up a campaign" walkthrough.

Which MGID macros do you need for tracking?

You need four things from MGID on every click: the widget ID (the placement), the teaser ID (the creative), the campaign ID, and the click ID for postback matching. MGID exposes these through its tracking macros, which you append to your landing page or offer URL as query parameters. Your tracker then captures them and stores them against the click.

The two that matter most for optimisation are the widget and teaser macros. The widget tells you where your ad appeared on MGID's network — which publisher site and which widget slot. The teaser tells you which creative image and headline combination earned the click. You want both as separate sub-IDs so you can slice by placement and creative independently, and cross-reference them when a widget only works with one teaser.

MGID's macro syntax follows the industry-standard {macro} pattern, but the exact token names change and MGID documents them in its own help centre. Always pull the current token list from MGID's tracking documentation before you build the URL — do not copy a token string from a blog post, including this one, because networks rename macros without fanfare. The ad network tracking macros explained guide covers the general pattern of what tokens are and how they map to sub-IDs if you're new to this.

How do you map MGID widget and teaser IDs to sub-IDs?

You map them by building your tracking URL so each MGID macro lands in a dedicated sub-ID slot. Adtraxo passes sub1 through sub5, which is enough for widget, teaser, campaign, and a spare for source or device. Here's the configuration that works:

1. Create the tracking link in Adtraxo and set your offer or landing page as the destination. On the Pro plan you get sub-IDs sub1–sub5 and S2S postback conversions; the Free plan doesn't include postbacks or sub-ID passing, so this setup needs Pro.
2. Append MGID's macros to your Adtraxo tracking link as query parameters, one per sub-ID. The pattern looks like this:
   https://yourdomain.com/click?sub1={widget_id}&sub2={teaser_id}&sub3={campaign_id}&sub4={click_id}
   Replace each {token} with MGID's current macro name from its documentation. The structure stays the same even if the token names differ.
3. Paste that full URL into MGID's campaign tracking field as the destination URL. MGID replaces the macros with real values at click time.
4. Verify the capture. Send a test click through MGID's preview or a live low-budget campaign, then open the click log in Adtraxo. You should see sub1 populated with a widget ID and sub2 with a teaser ID. If a slot is empty or shows the raw {macro} text, the token name is wrong — fix it before spending.

One decision to make early: do you want the widget ID in sub1 and teaser in sub2, or the reverse? Pick one order and use it across every MGID campaign. Consistency is what lets you pull a report across campaigns later and have sub1 always mean "widget." If you run Taboola or RichAds alongside MGID, keep the same slot convention so your cross-network reporting stays readable. The network-by-network setup guide covers how the other networks map their tokens.

How do you send MGID conversions back by postback?

Postback is how MGID learns which widgets and teasers produced conversions, and it's the half of MGID tracking setup most affiliates skip. Without it, MGID optimises toward clicks because clicks are all it can see. With it, MGID's algorithm can bid up the placements that actually convert.

The flow works like this: a user clicks your MGID ad, lands on your offer, converts, and your affiliate network or offer platform fires a postback to Adtraxo with the click ID. Adtraxo matches that conversion to the original click, records it against the right sub-IDs, and — if you've configured it — fires a second postback out to MGID with MGID's own click ID so the network attributes the conversion.

1. In Adtraxo, set up the conversion postback URL from your affiliate network. The network appends its click ID (often as {clickid} or similar) so Adtraxo can match the conversion to the click.
2. In Adtraxo's postback settings for the MGID campaign, add MGID's conversion postback URL. This is where the click ID you captured in sub4 gets passed back to MGID using MGID's expected parameter. Pull MGID's postback template from its documentation — it specifies the endpoint and the click-ID parameter name.
3. Test the chain end to end. Fire a test conversion through your affiliate network's test tool or a real low-value conversion, then check both Adtraxo's conversion log and MGID's conversion report. The conversion should appear in both within the postback window. If it shows in Adtraxo but not MGID, the outgoing postback URL or click-ID parameter is wrong.
4. Confirm the conversion lands on the right sub-IDs. In Adtraxo, open the conversion and check that sub1 (widget) and sub2 (teaser) are attached. If they're blank, the click didn't capture the macros correctly — go back to step 4 of the mapping section.

If your offer is on a platform that doesn't support postbacks, you're stuck with pixel-based conversion tracking, which is less reliable across browsers. For native traffic where you control the offer page, S2S postback is the standard and worth the setup time. The Taboola conversion tracking guide walks through the same postback logic if you want a second example of the pattern.

What does a good MGID widget report look like?

Once clicks carry widget and teaser IDs and conversions flow back, you can build the report that actually drives decisions: widget ID, clicks, conversions, cost, and revenue per widget. Here's the shape of it in Adtraxo's per-campaign analytics, grouped by sub1:

Widget (sub1)	Clicks	Conversions	Cost	Revenue	ROI
widget_A	420	9	$84	$162	+93%
widget_B	310	1	$62	$18	−71%
widget_C	180	0	$36	$0	−100%

Those numbers are illustrative, not benchmarks — your own data will look different. The point is the shape of the decision. Widget A is carrying the campaign. Widget B has enough clicks to judge and is losing money. Widget C has no conversions and modest spend. The natural move is to cut C now, watch B for another burst of clicks before cutting, and leave A alone.

Now cross-reference by teaser (sub2). If widget_A converts only with one teaser and dies with the others, you've found a placement-creative pairing worth isolating. That's the kind of insight MGID's own dashboard can't give you, because it doesn't see your offer's conversions unless you postback to it — and even then, it won't show you the teaser-by-widget cross-tab the way your tracker can.

Two reporting habits make this sharper:

• Segment by country and device. Adtraxo's country analytics and device breakdown let you spot a widget that converts on mobile but not desktop, or in one geo but not another. A widget that's ROI-negative overall might be profitable in a single geo, which changes your cut decision.
• Use fraud detection before you blame the widget. A widget with high clicks and zero conversions might be a bad placement — or it might be bot traffic. Adtraxo's fraud detection flags IP velocity, datacenter IPs, bot user agents, and invalid referers. Check the fraud flags on a dead widget before you cut it; if the traffic is fraudulent, you have a refund or block case with MGID, not just a cut.

How do you decide which MGID widgets to cut?

Cut on conversions and revenue, not CTR. A widget with a 4% CTR and zero conversions is worse than a widget with a 1% CTR and a 6% conversion rate. The decision framework:

1. Set a minimum click threshold before you judge any widget. A widget with 30 clicks and no conversions tells you almost nothing. Wait for enough clicks that a zero-conversion result is meaningful — the threshold depends on your offer's typical conversion rate, but the principle is: don't judge on noise.
2. Rank by revenue per click (revenue ÷ clicks) rather than raw revenue. This normalises for spend and lets you compare a high-traffic widget against a low-traffic one fairly.
3. Cut the bottom of the list first. Widgets with enough clicks, zero conversions, and meaningful spend go first. Then widgets with conversions but negative ROI after a reasonable sample.
4. Blacklist, don't just pause. Add the widget ID to MGID's blacklist so it can't reappear in a new campaign. The MGID widget blacklist guide covers the cut-vs-watch decision in more depth, including when to give a widget a second chance after a creative change.
5. Re-test cut widgets occasionally. A widget you cut six months ago may perform differently with a new offer or a seasonal shift. Keep your blacklist exportable so you can selectively re-enable.

The same logic applies to teasers. If a teaser has clicks but no conversions across multiple widgets, it's the creative that's failing, not the placement. The push ad creative testing guide covers how to compare creatives by conversions rather than clicks — the same principle applies to MGID teasers.

Why do MGID and your tracker show different click counts?

They usually will, and it's rarely a bug. MGID counts a click when its server records the redirect; your tracker counts a click when the request reaches your tracking domain. Between those two events, a user can close the tab, lose connectivity, or hit a browser that blocks the redirect. A small gap is normal. A large gap — say MGID showing significantly more clicks than your tracker — points to something worth investigating: a slow tracking domain, a broken redirect, or bot traffic that MGID counts and your fraud detection filters.

The click discrepancy guide covers the causes and how to diagnose them. For MGID specifically, the most common culprit is a tracking domain that resolves slowly in the geos you're buying. A custom tracking domain on Adtraxo (one is included on Free; unlimited on Pro) lets you use a domain that's fast in your target markets rather than a shared one.

What about landing page arrival tracking?

MGID clicks that reach your tracker but never load your landing page are a silent budget leak — you pay for them and the user never sees the offer. Adtraxo lets you add a one-line arrival snippet to your landing page on any plan, which records whether the page actually loaded in a real browser. If your click count is healthy but arrival count is low, the problem is between your tracker and your landing page: slow hosting, a redirect chain, or a geo where your page doesn't render. Fix that before you cut widgets, because you'd be cutting placements that were fine and blaming them for a page-speed problem.

Frequently asked questions

Ready to run MGID with real per-widget attribution? Set up your tracking link, map the widget and teaser macros to sub-IDs, wire the postback, and start cutting on revenue instead of CTR. Start your Adtraxo Pro trial at adtraxo.com — $27/month at the founding rate, 30-day money-back guarantee, cancel any time.