Future plc, which owns Tom's Guide and TechRadar, saw its affiliate revenue fall 24% in its most recent half-year and blamed lower search audiences, according to Ahrefs' affiliate marketing guide. The same guide notes that the publishers still growing tend to be brands people visit directly.
Two affiliate sites can earn the same commissions this month and face very different next quarters. The difference is where their visitors come from. You can't see the next Google update coming, but you can measure how badly your site would hurt if search traffic dropped.
This audit takes about an hour and uses three exports. It ends with a scorecard, a list of your riskiest pages and merchants, and a 90-day plan.
One caveat up front: every threshold below is a rule of thumb and a judgment call. None comes from research, and none predicts what Google will do. The audit measures exposure. It does not promise that any fix will recover rankings or traffic.
The one-hour plan
- Export Search Console data and split branded from non-branded clicks (15 minutes).
- Export analytics sessions by channel (15 minutes).
- Export affiliate revenue by page and by merchant (15 minutes).
- Reconcile the three sources so you can trust the numbers (10 minutes).
- Score each metric and rank your fixes (5 minutes).
Use the same date range for every export. Ninety days works for most sites. If your niche is seasonal, use 12 months so one holiday spike doesn't distort the picture. If you have only a few months of data, use everything available and treat the results as a starting baseline, not a verdict.
Step 1: Split branded from non-branded Google clicks
In Search Console, open the Performance report for search results and set your date range. Add a query filter, choose a custom regex, and enter your brand name with its common variants, such as "examplegear|example gear|examplegear.com". Record the clicks. Then note total clicks and subtract to get non-branded clicks.
Branded clicks come from people who already know you and typed your name. Non-branded clicks come from people who searched a topic and found you among competitors. Google can reroute non-branded traffic, and AI answers can intercept it.
Ahrefs reports that an AI Overview now correlates with a 58% lower average clickthrough rate for the top-ranking page (Ahrefs). That is a correlation, not a guarantee for your queries. It is also an inference, not a finding, that people searching your name are less likely to be diverted than people searching a generic topic.
One gap to know about: Search Console hides some rare queries. Your filtered branded clicks may undercount slightly, so use the report's total clicks as the denominator.
Step 2: Measure non-search traffic share
In your analytics tool, pull sessions by default channel for the same period. In GA4 this lives under the traffic acquisition report, though menu names change, so check your version. You want sessions from organic search, direct, email, YouTube, other social, and referral.
YouTube and social often land in the wrong bucket. Tag your video descriptions, newsletter links and social posts with UTM parameters so they show up under their own names. Also remember that "direct" includes some traffic analytics can't attribute, such as links opened from certain apps. Treat it as a rough proxy for people who know you, not a clean count.
For the scorecard, add direct, email, YouTube and social sessions and divide by total sessions. Call that your owned-channel share. Then estimate non-branded search share of sessions: organic sessions multiplied by your non-branded click share, divided by total sessions.
Step 3: Calculate revenue concentration
Export commissions by merchant from each affiliate dashboard or network. For page-level revenue, you need a way to tie commissions to pages. Many programs support sub-IDs or tracking tags per link. If yours don't, use analytics outbound-click events by page as an approximation, and label it as one. For more on this, see related: how to choose b2b marketing channels: a scoring model.
Calculate four numbers:
- your top merchant's share of revenue
- your top two merchants' combined share
- your top page's share
- your top-10 pages' combined share
If your site has fewer than 30 pages, use the top 3 pages instead. A top-10 share means little when 10 pages are most of the site.
Concentration can be steep. Ahrefs' guide describes the craft blog Random Olive, which earned $2,568 in August 2026, mostly from TikTok Shop ($1,274) and Amazon ($1,218), then fell to $1,915 in September (Ahrefs). By simple arithmetic, those two programs made up about 97% of the August total. Ahrefs does not state that figure; it is my calculation from Ahrefs' numbers. It still shows how one program change could move a month.
Step 4: Check that your numbers are reliable
The three tools will never agree exactly. Run these checks so you know how much to trust each figure.
Search Console vs. analytics. Compare Search Console clicks with analytics organic sessions. The two differ because analytics counts sessions, includes other search engines, and loses visitors who decline tracking. As a rule of thumb, a gap within about 20% is normal. A larger gap suggests broken tags or filters, and you should fix those first.
Affiliate dashboard vs. analytics. Compare affiliate dashboard clicks with analytics outbound-click events. Merchants filter bot and duplicate clicks, so the dashboard should come in somewhat lower. A far bigger gap points to link or tagging problems.
Commission status and timing. Use approved or paid commissions where possible, and expect recent weeks to be understated. Pending sales, returns and reversals shift totals. Cookie windows add lag too. Amazon's cookie lasts only 24 hours unless the item goes into a cart, which can extend credit to as long as 90 days (Ahrefs), so a sale can post well after the click.
Worked example (hypothetical numbers)
Everything below is invented to show the math. "Example Gear Reviews" is not a real site, and these figures are not benchmarks or case-study results.
| Input (90 days) | Value | |---|---| | Search Console total clicks | 18,000 | | Branded clicks | 900 | | Total sessions | 24,000 | | Organic search sessions | 19,200 | | Direct sessions | 1,440 | | Email sessions | 480 | | YouTube sessions | 600 | | Other social sessions | 720 | | Affiliate dashboard clicks | 2,100 | | Analytics outbound-click events | 2,400 | | Affiliate revenue | $6,000 | | Top-10 pages revenue | $4,500 | | Top page revenue | $1,200 | | Top merchant revenue | $3,300 |
Start with reconciliation. Analytics organic sessions (19,200) run about 7% above Search Console clicks (18,000), inside the 20% rule of thumb. The affiliate dashboard shows 2,100 clicks against 2,400 outbound events in analytics, or 87.5%, which is also plausible. The data is usable. We cover related ground in go deeper on search share vs page views: is your traffic really down?.
Now the metrics:
| Metric | Calculation | Result | Rating | |---|---|---|---| | Branded share of Google clicks | 900 / 18,000 | 5% | High risk | | Non-branded search share of sessions | 19,200 x 95% / 24,000 | 76% | High risk | | Owned-channel share | (1,440 + 480 + 600 + 720) / 24,000 | 13.5% | High risk | | Top-10 page revenue share | 4,500 / 6,000 | 75% | High risk | | Top page revenue share | 1,200 / 6,000 | 20% | Medium risk | | Top merchant revenue share | 3,300 / 6,000 | 55% | Medium risk |
The rule-of-thumb thresholds
These are the cutoffs used above. They are judgment calls, so adjust them to your niche and site size.
| Metric | Low risk | Medium risk | High risk | |---|---|---|---| | Branded share of Google clicks | Above 30% | 10% to 30% | Below 10% | | Non-branded share of sessions | Below 40% | 40% to 65% | Above 65% | | Owned-channel share | Above 30% | 15% to 30% | Below 15% | | Top-10 page revenue share | Below 50% | 50% to 70% | Above 70% | | Top page revenue share | Below 15% | 15% to 25% | Above 25% | | Top merchant revenue share | Below 40% | 40% to 60% | Above 60% |
Score low as 0, medium as 1 and high as 2. A total of 0 to 4 is low overall risk, 5 to 8 is medium, and 9 to 12 is high.
The example site scores 10, so it rates high. That does not mean it will lose traffic. It means a bad quarter in search would cost it most of its income, with little to cushion the fall.
Step 5: Turn weak results into a prioritized action list
Fix the problems you can control fastest first. Email and merchant mix move quicker than video or brand building.
Also read: see how to automate content refreshes without publishing slop
| Weak result | Fix | First move | |---|---|---| | Low owned-channel share | Email list | Add a signup offer to your top 10 revenue pages, such as a buyer's checklist for that topic | | Low branded share | YouTube or community presence | Turn your best-earning review into a video that shows the product in use, and link back with UTM tags | | Page concentration | Original testing content | Add your own photos, measurements and test notes to the top pages | | Merchant concentration | Wider merchant mix | Find two alternative programs for your top-earning category and compare payout terms | | High non-branded share | All of the above | Track weekly signups and branded searches as your leading indicators |
The testing fix follows Google's own guidance. Ahrefs notes that if your site adds nothing beyond the merchant's product page, Google has little reason to send people to you instead. It also quotes one of Google's self-assessment questions: whether your content would be useful to an existing or intended audience if they came directly to you (Ahrefs). A site that can't answer yes has the most to gain from this work.
A 90-day sequence
- Days 1 to 30: Install email capture on your top pages, fix any tagging gaps found in Step 4, and add UTM tags to every channel you own.
- Days 31 to 60: Send your first regular email, publish two or three videos or community posts based on your top pages, and apply to one or two additional merchant programs.
- Days 61 to 90: Add original testing material to your top pages, then re-run this audit.
HouseFresh shows why a second channel matters. Ahrefs reports that the air purifier site lost about 95% of its Google traffic after the 2023 update and was back to pre-update levels by October 2025, helped in part by a YouTube channel (Ahrefs). Ahrefs also says not every site recovers. Treat HouseFresh as an example of resilience, not a forecast.
Re-run the audit every quarter
Save a dated copy of your worksheet, such as "2026-Q4". Next quarter, use the same date range length, the same brand regex and the same threshold table. Record each metric and the score. Add a notes column for anything that changed, such as an algorithm update, a new video, or a merchant cutting its rates.
Compare direction, not decimals. As a rule of thumb, a move of a few percentage points is likely noise, especially for small sites. A steady rise in owned-channel share and a falling top-merchant share over two or three quarters is a stronger signal. If a score gets worse, check your reconciliation numbers before blaming Google.
What to expect next
Expect the first re-run to look similar to this one. Email lists and branded searches build slowly, and a quarter is a short time. The most useful result after 90 days is a clearer picture of which pages and merchants carry your income, and which channels are starting to grow independently of Google.
