About our hearing aid rating algorithm

The problem with hearing aid reviews

Unlike most consumer electronic products, it is a challenge for a consumer to draw any useful conclusions from online hearing aid reviews.

In fact, many hearing healthcare providers and experienced hearing aid wearers would even say it is a mistake to rely on hearing aid reviews in any way.

Why?

Hearing aid reviews are subjective.

Two users of the same hearing aid may have drastically different opinions of the quality and efficacy of that hearing aid.

The nature of hearing loss, and the varying ways in which hearing aids are sold and fitted to consumers, introduces numerous factors that make it difficult to reach any meaningful conclusions from reading hearing aid reviews.

These factors include, but are not limited to:

  • The type of hearing aid programming service that was included (face-to-face vs. remote care)
  • The skill level of the hearing care provider programming the hearing aid
  • Whether or not the hearing aid was fit using best practices
  • Whether or not the hearing aid was appropriate for the user’s hearing loss and lifestyle
  • The wearer’s expectations of the hearing aid. A user with unrealistic expectations is likely to be less satisfied with their hearing aid as compared to a user with realistic expectations.

The list goes on and on.

As a result of the subjective nature of hearing aid reviews, the internet is awash with “experts” creating lists such as “Top 5 Hearing Aids.”

At best, hearing aids that end up on these “Best of” lists are placed there by well-intentioned and informed industry professionals who are trying to provide helpful content for consumers.

At worst, products on those lists are merely advertisements for hearing aids that are sold (often exclusively), by the same company who published the list.

In either case, products usually end up on these “Best of” lists in a mostly arbitrary or self-serving way.

We thought there should be a better way to rate hearing aids.

Our solution: a comprehensive hearing aid rating algorithm

We sell thousands of hearing aids each year.

Each transaction provides us with valuable data, such as user reviews and return rates.

We figured it was time to put this data to use.

So, we built a hearing aid rating algorithm that takes all these data points into consideration (and we hope), tells us a story about what it all means.

The algorithm is unbiased and there are no paid placements on the resulting “best of” lists.

Our solution is far from perfect; we clearly point out the inherent flaws, and we invite your feedback.

In spite of these flaws, we believe a bit of objective data has the potential to play a valuable role in helping consumers choose a hearing aid.

Built by experts

We’re a hearing aid retailer, not a data company.

So to create this algorithm, we enlisted the help of professionals to advise us on the construction of the algorithm, its logic, and the best practices we should follow.

Our sincere thanks to the following experts who helped with the creation of this algorithm:

Jeff Hall - president of ZipHearing

Jason Katz

MS, Machine Learning Engineer

Jeff Hall - president of ZipHearing

Jеffrеу Сhrаbаszсz

PhD, Data Scientist

Methodology

Every hearing aid model listed on our site is subjected to the same rating criteria.

Below, you’ll see a brief overview of the logic behind our algorithm.

Technical summary:
  • We use Bayesian Statistics to adjust review scores and return rates based on sample size.
  • We use Normalization to score models in comparison to each other.
  • Each category’s weight is based on its relative importance and ability to accurately and precisely measure the metric in question.
Hearing Aids prices range 25%
Hearing Aids prices range 25%
CATEGORIES USED PERCENT
User Reviews 25%
Return Rates 25%
Model Age 25%
Sales Popularity 25%

User Reviews (25%)

This category looks at user-submitted reviews (verified and unverified) for individual models.

Potential flaws:

  • With very few reviews, a model’s score can be easily influenced by an outlier review.
  • With very few reviews, the category is still worth a weight of 25%.
  • A negative review may not be indicative of poor product quality.
  • Potential for fake reviews. All reviews are manually-reviewed before being published, and we have high confidence that our review system is not being spammed, but the potential exists. In addition, more than 80% of the user reviews our site receives are from verified purchasers, so if fake reviews are received, we estimate they make up a small percentage of overall reviews and will not have a significant impact on this category.
  • On some sites, customers may be more likely to leave reviews if their experience was good; on others, only if it was very good or very bad. In either case the resulting ratings can suffer from selection bias. As a result, any reviews we receive might not accurately represent the full range of customers’ experiences of the product.

Return Rate (25%)

This category looks at return rates for individual models.

Potential flaws:

  • A returned item may not be indicative of poor product quality.
  • If a model has very few sales, its return rates can be easily influenced by a single return.
  • If a single line of hearing aids from a manufacturer contains contains both in-the-ear models and receiver-in-canal models, this line's overall return score is going to be negatively impacted by the fact that custom products have a higher return rate across the board than receiver-in-canal hearing aids. This unfairly penalizes a hearing aid line that offers multiple styles within the same line, as opposed to a hearing aid line that names a separate line of hearing aids for their custom products. We are currently working on a solution to this problem.

Model Age (25%)

This category looks at which generation a hearing aid model is (relative to other competing models of the same brand). A newer generation model will earn a higher score in this category.

Potential flaws:

  • Newer does not always mean better. However, in most cases, when a newer generation product is released, it has small iterative improvements over the previous generation, and it stands to reason that a newer model will be a better overall product.

Sales Popularity (25%)

This category looks at our internal sales data over the previous 60 days and assigns a higher score to products with a greater volume of sales.

Potential flaws:

  • If a product is new, it will take a little time for it to gain attention and generate sales.
  • A new product might get a high volume of sales initially because of a marketing campaign.
  • Higher sales volume doesn’t mean better. However, because we work with 600+ hearing aid offices nationwide, we do think that if a model is seeing high sales volume, that reflects a strong vote of confidence in that model by hearing providers.
  • A product with a high volume of sales will earn a high score in this category, which leads to an overall higher score for the product, a better position on our ranked lists, and then more sales for the product. This gives a perhaps unfair advantage to products that appear at the top of our ranked lists. However, because all products start with a score of 0/25 in this category, a product cannot earn a top position in one of our ranked lists if it hasn't previously generated a high volume of sales (without having the benefit of a top placement and increased visibility).

Results

The algorithm takes the sum of a model’s score in each of the categories, which then becomes the model’s total numerical score, ranging from 0-100, with 100 being a perfect score.

This score then determines a model’s position (if any) on one of our “Best of” lists, such as:

Hearing aid model scores are updated nightly at the close of business.

Things you should know

Please do not use our “Best of” lists exclusively to form any conclusive opinions.

Instead, use our lists as one point of reference as you research hearing aids. There is no perfect hearing aid rating system. Because of this, we feel consumers should reference multiple reputable sources.

Here is another reputable source we recommend:

HEARINGTRACKER.COM - BEST HEARING AIDS

HearingTracker’s “best of” list is curated by Doctor of Audiology and industry expert, Abram Bailey, AuD. HearingTracker also surveys hearing providers and publishes those results, providing additional insight and perspectives.

Learn all you can from our website and the source above. Once you’ve done that, discuss your findings with your hearing care provider. Since your hearing care provider should have the best understanding of your requirements for hearing aids, ultimately, you should put the strongest weight in their recommendations.

It’s not feasible for us to rate every hearing aid that is commercially available via this algorithm.

The data used in this algorithm is our internal data, and we only have this data on hearing aids we offer for sale. However, we do sell hearing aids from all 6 of the biggest hearing aid brands, so the most popular hearing aids on the market are represented in our results.

Customers who have purchased hearing aids through ZipHearing are solicited to write hearing aid reviews 3 weeks after their purchase.

In exchange for writing a review, a customer receives a small gift card to the retailer of their choice. However, this gift is not dependent upon the customer writing a positive or negative review- we simply request that the review be honest.

Finally, we know our rating system is not perfect, and we invite your feedback.

We’re always looking to improve our algorithm and methodology, and as we make improvements to it, we’ll post updates here.

If you have any feedback, we’d love to hear from you!