Freeze one real use case before comparing candidates
This method begins after two or three Hacoo candidates have already been found. Do not reopen discovery or ask which product looks most exciting. Write one sentence describing the job the purchase must perform: “a light outer layer for a cool, wet commute that fits over one sweatshirt,” “a compact bag for a tablet, bottle and daily essentials,” or “shoes for several hours of city walking.” Include the user, setting, frequency and constraint that would make the product useful. A candidate suitable for a festival outfit may not satisfy a daily commute, even when both are jackets. Hacoo currently describes itself as a content-sharing community built around ideas, creators and personal experience. That material can inspire a shortlist, but it does not define your requirement. Lock the use case before reading more captions or reviews so a persuasive image cannot quietly change the task. If the use case contains two unrelated jobs, split it into two decisions rather than forcing one product to do everything.
Translate the use case into observable must-haves
Turn the sentence into four to seven conditions that can be checked on the exact candidate. Use nouns, numbers or visible functions instead of praise words. “Fits over my reference sweatshirt” is testable when you record the sweatshirt width and the candidate garment measurement; “comfortable” is not testable from a listing. “Holds a 28-centimetre tablet and closes fully” is clearer than “roomy.” A useful list may include minimum internal dimensions, required closure, exact colour family, selected quantity, compatible model, maximum garment length or a hard budget ceiling for the item. Keep preference out of this first pass. Decorative details, trend appeal and a favourite shade belong later unless the use case truly fails without them. A must-have has a consequence: if the condition is not met, the product cannot perform the named job. If you cannot explain the consequence, rewrite the item as a preference or remove it.
Define the evidence that can satisfy each criterion
A criterion is only useful when you state what evidence counts. For a measurement, require a current chart or visible specification tied to the exact size. For capacity, require dimensions or a clear internal layout rather than a styled photograph. For a phone case, require the precise device model and cut-out arrangement. For a set, require a written contents list rather than assuming every object in a promotional image is included. Hacoo’s public intellectual-property guidance says content and links can come from independent creators and users; that means a creator statement is not automatically an official product specification. Record evidence as direct, indirect or missing. Direct evidence answers the criterion for the chosen variant. Indirect evidence suggests an answer but leaves an important assumption. Missing evidence supplies none. This is not a quality score and does not prove authenticity or performance. It simply stops an unsupported statement from being treated as a passed requirement.
Apply the criteria to exact variants, not broad listings
A Hacoo listing can contain several colours, sizes, bundles or versions. Compare the option you could actually choose, not an average impression of the whole page. Copy the exact option label beside each candidate and note whether selecting it changes the image, displayed specification, contents or price. A size chart shown elsewhere on the page may not clearly apply to every version. A bag photograph may show a long strap although the selected bundle does not list one. A low starting price may refer to a smaller or incomplete option. When the evidence cannot be tied to the variant, mark the criterion unknown. Do not borrow an answer from another candidate or a visually similar product. The comparison remains fair only when every row asks the same question of the same level of product identity. If a candidate has disappeared or changed since the shortlist was made, remove it or rebuild its record before scoring anything else.
Use pass, unknown and fail before any weighted score
Create a table with one row per must-have and one column per candidate. Allow only three results: pass, unknown or fail. Pass means the required evidence supports the condition for the exact variant. Fail means the evidence clearly contradicts it. Unknown means the necessary fact is absent, ambiguous or detached from the selected option. Do not convert unknown into half a pass merely because you prefer the design. Add a short evidence note and the date checked to every result. Then apply the rule consistently: a true fail eliminates the candidate, while a high-consequence unknown pauses the decision until the fact is resolved. This gate comes before price points, style scores or weighted preferences. It prevents a product from winning with attractive secondary details while missing the reason for the purchase. The table may eliminate every candidate. That is a valid result and evidence that the shortlist, not the criteria, needs to change.
Keep preferences separate and deliberately lower priority
Only after a candidate passes the must-have gate should you compare nice-to-haves. These may include a preferred shade within the acceptable colour family, a pocket arrangement, a decorative finish or a lower current item price. Write them in a separate section and rank them rather than quietly promoting them to requirements. This protects the original use case from choice overload. It also makes trade-offs readable: Candidate A and Candidate B both meet the commute requirement, but A has clearer care information while B has the preferred colour. The reader can now choose consciously instead of pretending the colour determined function. Do not award extra points for vague terms such as “premium,” “viral” or “best seller” unless the term supplies evidence for a stated need—which it usually does not. Popularity, creator enthusiasm and polished photography can influence preference, but they cannot repair a failed dimension, incompatible model or missing included part.
Finish with a short decision statement and a stop rule
End the comparison with one sentence that names the use case, the selected candidate and the evidence supporting every must-have. For example: “Candidate B remains because its recorded width supports the planned layer, the selected option lists the required closure, and the current dimensions fit the carry limit.” If the sentence needs an unsupported claim about material performance, comfort, quality, authenticity, delivery or future availability, the decision is not ready. Write the unresolved fact instead. Set a stop rule as well: choose a passing candidate only when every high-consequence criterion has direct evidence; otherwise request clarification, wait for better information or choose none. Reopen the exact live option immediately before acting because pages, variants and prices can change. This final check does not guarantee an outcome. It shows that the choice still matches the requirement you fixed before comparison and that uncertainty has not been disguised as confidence.