A single traffic figure can make a publisher look attractive without explaining how that figure developed. A sharp increase may deserve a closer look; a fall may have a narrow explanation. A trend review helps an agency ask better questions before booking, especially when a sports, betting or casino campaign depends on a particular audience.
This is a suggested research process, not a scoring formula. It does not turn estimated search traffic into verified readership, and it does not establish that a publisher can deliver a particular number of visits to your article.
Write down the comparison scope first
Record the tool, report, country or database, device scope where shown, target domain or subfolder, selected months and checked date. Compare the same scope throughout. Keep any worldwide view separate from a country-specific view; label a report’s actual scope rather than relying on a chart’s visual similarity.
Semrush’s historical-data documentation describes month-by-month estimated SEO traffic in its relevant reports and notes that historical availability depends on the database. Confirm what the account and report actually provide. If a month is unavailable, leave it missing instead of filling the gap with an invented estimate.
Describe the shape before proposing a cause
Use plain observations: “the estimate rose over three comparable months” or “the selected month is below the previous range.” Then list the questions that might explain the pattern. Do not write “a penalty caused the fall” unless there is evidence supporting that explanation.
Ask whether changes are concentrated in a few pages, topics or countries. Check whether a newer section accounts for much of the observed growth and whether that section relates to your campaign. A publisher can be relevant to a brand without every popular page being relevant to the planned article.
Illustrative example: identical totals, different histories
Illustrative data only: Publisher A shows monthly estimates of 20,000, 21,000 and 22,000. Publisher B shows 5,000, 35,000 and 22,000 under the same hypothetical report scope. Both end at 22,000, but the second series gives the researcher a different question to investigate. Neither series is a real publisher measurement or a forecast of referral visits.
A useful note for Publisher B would request the relevant page or topic breakdown and the reason for the unusual month. It would not automatically reject the site or promise that its latest estimate will continue. Keep the research conclusion provisional until the supporting evidence is checked.
Keep search estimates separate from placement terms
Alongside the trend note, record the proposed article category, content suitability, current price and the terms still needing confirmation. A chart does not confirm acceptance of casino, crypto or poker content. It also does not confirm a homepage position, a link format or a publication date.
For a returning publisher, keep a dated research record rather than overwriting the old figure with a new one. That preserves what the agency knew when the campaign was approved. If a team member changes the report scope, start a separate series so a later reviewer can understand the difference.
Turn the review into a useful shortlist note
Summarise the evidence in three lines: the observed pattern, the question resolved or still open, and the consequence for this campaign. For example, the consequence might be “obtain a category breakdown before approval” rather than an unsupported quality verdict.
Combine this review with our country-mix guide when comparing the Route77Media inventory. Send your niche and target audience through our placement service, and plan the included article promo video as a separate content asset rather than adding assumed traffic to the publisher estimate.
Sources and further reading
Sources checked 24 October 2026.