Underestimating the problem of misspelling
Unbxd's data indicates that about 25% of eCommerce site search queries need to be corrected.
Our research found that shoppers tend to misspell the simplest of words. For example, on a major fashion retailer's site, we found that misspelled products were among the most frequently searched.
The results are shocking:
"Jumpsuit" was misspelled in 216 ways, and "bodysuit" was misspelled in 223 ways.
Misspelled words above are all simple, single words. The complexity exponentially increases for long-tail queries.
Manual mapping cannot fix large corpora of misspellings
Usually, most eCommerce sites manually map misspelled words to the product information. But this means hours of manual effort and intervention. Moreover, while this might be a viable solution to solving small-scale problems, it is highly ineffective if you have an extensive catalog with thousands of products that people search for using complex compositions of words and symbols.
Even with a large team to manually match incorrect queries to the right search results, businesses can only fix 30%-40% of misspelled queries. The rest are easily lost, leading to high customer attrition and sales losses.
An innovative, automated system that handles site search and common misspellings solve this at scale.
But care has to be taken in choosing the system. The majority of spellcheck solutions on the market only match text patterns. First, they check the misspellings against the fundamental English corpus and product catalog. They then determine how many letters are misspelled and match them to the closest possible alternatives.
Despite being a quick fix to manual mapping, it often leads to incorrect solutions in the long run. The most popular spellcheck solutions provide an accuracy of less than 60%.
With an intelligent site search and common misspellings solution, you can improve accuracy by 30%
Besides the basic English corpus and product catalog, Unbxd searches for the closest possible alternatives to common misspellings using a context-aware, industry-specific corpus. As a result, it is 30% more accurate than most spellcheck solutions.
For example, for a fashion retailer, the system understands that "skort" is an apparel category, not a misspelled "skirt" variation. Therefore, it returns relevant results for skorts instead of incorrectly accounting for the search query as erroneous.
While most spellcheck systems cannot effectively identify more than two incorrect letters in the same word, the Unbxd solution offers 90% accuracy. Unbxd can correct up to four misspelled letters in the same word. While correcting one misspelled letter in a word is relatively easy, the complexity increases as the number of misspelled words increases.
For example: Consider the search for "jumpsuit."
The most commonly misspelled version is "jumpsuite," which has one misspelled letter. Most spellcheck solutions can correct this mistake and show relevant results.
Consider an unlikely spelling mistake in the same search query; say "jmpsuts." It has three misspelled letters. Most spellcheck solutions cannot correct this mistake, much less show relevant results.
Unbxd's spell check solution analyzes a shopper's vertical and expected keystroke errors to determine the right user intent behind a misspelled query.
Similarly, the system considers the surrounding keys of a misspelled letter in a word and improves upon the suggestions for correct alternatives.
Most users misspell a search query by pressing an adjacent key to a letter by mistake. This is particularly true for mobile shoppers due to the "fat finger" syndrome; therefore, spelling mistakes due to QWERTY key positions are more likely to occur.
Although spellcheck is only one of your site's search features, it dramatically impacts your revenue.
Don't let poor spelling ruin your bottom line
Site searchers are 2-3 times more likely to convert. Human errors account for 25% of site search errors. Fixing common misspellings fixes those errors. Unbxd data shows that these conversions could add a potential $3M to your revenue.
On the other hand, if the spell check feature is not fixed or improved, it could cost businesses more than $3M in a potential loss of revenue, not discounting high customer attrition and low brand loyalty.
To know how much additional revenue Unbxd's spell check feature can add to your topline, book a demo.