What "Best Match" Actually Weighs: Cassini Beyond the Title

Why a strong title is necessary, and not sufficient

The title decides which searches a listing can be matched against at all - covered in full in writing titles that get found. Best Match, the default sort buyers see for almost every search, then decides where a listing lands among everyone else who also matched that search, and that ranking draws on a genuinely wider set of signals than the words in the title alone. Two listings can both be a perfect keyword match for the same search and still land in very different positions, because Best Match is also weighing who is selling the item and how well that specific listing tends to convert - not just whether it was found in the first place. That distinction is worth sitting with, because it changes what "optimizing a listing" actually means in practice - it is not a single edit made once and left alone, it is an ongoing set of account-level habits that happen to also improve where a listing sits in a search, rather than a search-specific trick applied to the listing on its own.

The three broad categories Best Match actually weighs

Beyond keyword relevance, eBay's own ranking behavior is generally understood to weigh two further categories alongside it: seller performance and listing quality. Seller performance covers the account-level record this blog has already gone into in depth elsewhere - feedback score, dispatch speed, and defect rate all feed directly into how a seller's listings rank, not only into the fees and limits covered in how seller levels actually work. A seller with a clean, fast, low-defect track record is not just protected from Below Standard penalties - their listings are placed better in search results than an identical listing from a seller with a shakier record, which is a separate and additional reward for the same underlying behavior.

Listing quality is the second category, and it goes beyond simply having a title that matches the search: a fully completed set of item specifics, a genuinely complete set of photos, and accurate categorization all feed relevance matching in their own right, not just the presentation a buyer sees once they click through. A bare-bones listing with the right title but empty item specifics and three photos is giving Best Match less to work with than a fully completed one, even before a single sale has happened on either.

Why price has to be read together with postage

Best Match does not judge an item's price in isolation - it weighs the total a buyer actually pays, price and shipping combined, against comparable listings for the same search. This matters specifically for a dropshipper who might be tempted to advertise a lower headline price while recovering the real cost through postage: two listings with an identical item price but different shipping costs are not equivalent in the buyer's total cost, and Best Match is understood to account for that combined figure rather than the item price alone. The practical implication is straightforward - a genuinely competitive total cost, not just a competitive-looking item price, is what actually helps a listing's ranking, and pricing strategy built around this distinction should already be informed by working from every real fee, not just the headline number on the listing. Building postage into the item price and offering "free delivery" is not simply a presentation choice for a dropshipper, then - it is worth testing genuinely, since a lower combined total at the same real margin can be a direct ranking advantage over a competitor splitting the same total cost across item price and a separate postage charge.

Sales history: the signal a brand new listing simply does not have yet

A listing's own sales performance over time - how consistently it actually converts a view into a sale, sometimes referred to as sell-through rate - is a genuine, separate ranking input, and it is one a completely new listing cannot compete on directly, having no history yet to show. This is worth understanding honestly rather than as a source of frustration: a brand new listing being outranked by an older one with a proven, steady conversion record is not a sign anything is wrong with the new listing, it is Best Match doing exactly what it is built to do - favouring a demonstrated track record over an unproven one, all else being similar. The practical response is patience paired with the fundamentals actually being right (title, price, listing quality) rather than repeatedly ending and relisting a slow-starting listing in the hope a fresh start ranks better - that resets the very sales history a listing needs time to build, rather than fixing the underlying reason it was slow to begin with.

Top Rated status: a ranking reward, not just a fee discount

Top Rated Plus already carries a real fee discount, covered in detail in how eBay seller levels work - but the same underlying status is also understood to carry its own separate visibility benefit in Best Match itself, on top of the fee saving and the badge a buyer sees on the listing. This means the standards behind Top Rated status - fast, reliable dispatch, a low defect rate, generous returns - are doing double duty: protecting an account from the penalties covered elsewhere on this blog, and independently helping every listing on that account rank better against otherwise-similar competition. Chasing Top Rated status purely for the fee discount already undersells what it is actually worth. For a new seller who has not yet reached it, this is also a genuine reason not to be discouraged by a slow start in search results specifically - the gap between a Top Rated competitor and a newer account is not purely about who wrote a better listing, it is partly a status that simply takes consistent, honest trading over time to reach, the same way it does for the fee discount attached to it.

Best Match is the default, not the only option

It is worth remembering that Best Match is the sort buyers see by default, not the only one available - a buyer can switch to sorting by price, or by newest listed, at any point, and neither of those alternative sorts uses Best Match's weighting at all. A newly listed item with no sales history yet can still be found by a buyer specifically sorting for the newest listings, entirely independent of how it would rank under Best Match - which is one honest reason not to panic about a slow start under the default sort specifically, since it is genuinely not the only route a buyer has to a new listing, just the one most buyers happen to use without changing it.

A worked example: two identical listings, two different outcomes

Take two sellers listing the same oak bookcase, sourced from the same manufacturer, at the same £75.99 item price. Seller A has a clean, fast-dispatch record, ships with free delivery built into that price, has sold four of this exact listing steadily over recent weeks, and holds Top Rated status. Seller B is a newer account with no defects yet but also no track record on this specific listing, charges the same item price plus a separate £4.99 postage charge, and has not yet reached Top Rated. Both listings can have an identically well-written title. Under Best Match, Seller A is very likely to rank above Seller B on this search - not because of anything in the title or the description, but because of the combined weight of a lower total buyer cost, a proven sales record, and an account-level standard neither listing's own wording ever mentions. This is the actual argument for treating the fundamentals covered throughout this blog - honest pricing, reliable dispatch, a supplier that does not let you down - as the real ranking strategy, with the title doing its own separate, narrower job alongside them.

What is actually worth checking, week to week

Given how many of these signals are genuinely outside a seller's direct, one-time control - they build or erode gradually rather than being set once and forgotten - the practical habit worth building is a short, regular check rather than a one-off listing audit. Watching how a specific listing's own interest figures move - how often it is shown in search, how often that turns into someone actually opening it - is a real, direct read on whether the fundamentals covered above are working, not a vanity number. A listing whose "shown in search" count is healthy but whose "opened" rate lags behind comparable listings is pointing at a title or price problem specifically; one where both figures are healthy but sales still lag is pointing somewhere else entirely - the listing itself, once opened, not whether it was found or clicked at all. Treating these as three genuinely separate questions - am I being shown, am I being opened, am I converting - is a far more useful diagnostic than a single vague sense that "sales are slow," because each of the three points at a different fix.

Where Sellhelm fits into a ranking conversation it does not control

None of Best Match's actual ranking logic is something any third-party tool can see inside or influence directly - eBay does not publish the exact weighting, and treating any claim to have "cracked" it with real precision is worth genuine scepticism. What Sellhelm does keep visible is the inputs a seller genuinely does control: pricing worked out from every real fee rather than a guess, a supplier's stock and price checked regularly enough that a listing does not go stale, and the order-side standards - dispatch time, defect rate - that feed both seller levels and, on the evidence above, ranking itself. Getting those right consistently is the actual lever available, not a shortcut around the algorithm itself. Chasing a rumoured trick - a magic keyword, a specific relisting time, a particular category workaround - is time spent on something eBay's own ranking signals were never actually built to reward, where the same time spent on genuinely improving dispatch reliability or supplier pricing moves a real, documented input instead.

Questions people ask