New product testing is a critical part of any successful business, B2B or B2C, which requires new products and niches for its business model to function. Whether you are an ecommerce looking for winning new products to stock, promote and profit from, or a solopreneur wanting to widen your product offering within your category or searching for a new niche, this is the guide for you.
Testing product niches is even more important if you stock in volume or your time to market is long (e.g., you are developing a super comprehensive video course on your own which is time-consuming or manufacture your newly invented product which is capital-intensive).
In these cases, when you decide to go with a product or niche, or line of products for a given niche, you are making a significant commitment. Needless to say, if you miss on a product, it can be a very expensive miss.
To avoid this situation, you should test some of your product ideas and research them thoroughly. This article summarizes some of the tests and techniques I use with my consulting clients to screen merchandise, niches, or ideas, and tests we do to improve forecasting accuracy. Before reviewing these techniques, I would like to set the stage by briefly reviewing the basic merchandise planning procedure.
While what you will read below may sound as if it was designed only for medium and large organizations, this type of thinking is equally good for the smaller firms and even solopreneurs who realize it is the planning and systems that make you successful.
Merchandise Planning
As you approach a planning period, you must describe your specific objectives. These objectives should refer to:
- Number of products you are going to actively promote.
- Number of promotional campaigns.
- Number of new versus continuing products
- Promotional or design accents
- The strength of the campaigns – what are you doing that is new?
- Point of immediacy – what will make customers decide to buy?
Once you have established objectives, you begin development of your new products. You should make key decisions regarding types of products, make vs. buy, develop outside buy product with your designs vs. pick-up of vendor’s design, acceptable margins, and pricing, etc.
To help in executing a product promotion plan, you can employ two techniques to expedite the process: implementation teams and dual planning schedules.
Implementation Teams
These teams are established when there is a product or group of products that need special attention or are on a shorter schedule. An implementation team usually consists of five to eight people drawn from different departments of the company. Their task is to resolve product development issues in a shorter period than would normally be allowed.
The team develops expertise regarding the designated product, develops the product and prepares it for market.
Once the mission is accomplished, the team is dissolved. You may find this technique useful in expediting, as well as creating, a sense of expertise and pride in your employees.
Dual Planning Schedule
The lead times for product put merchandisers or product managers in a difficult position. It is hard to anticipate what products will be popular many months or even a year in advance. The solution to this problem is to develop your product promotion plan in two major phases.
The first phase begins a 3-, 6-, or 12-months in advance (depending on the type of your business). The second phase is maintained on a much shorter schedule. This enables a development group, art department and graphics group to spread execution over a longer period. It also allows the product management and merchandising group to wait and watch the market until closer to the product launch time.
Product Testing and Screening Niches – “Laboratory” Research
There are two basic types of research I use to do preliminary screening of product and niche ideas as well as actual products. They are niche and product concept testing and prototype testing.
Niche and Product Concept Testing
At the beginning of any given promotional period, you should develop literally dozens of new product niches and hundreds of specific product ideas. Concept testing is used to aid you in deciding which ones to develop and produce.
Objective
The objective of this research is to determine consumers’ overall inclination to buy new products and/or their inclination to buy from a new product niche or category.
Format
In this research you test written descriptions of new product concepts which consumers rank on a scale to indicate their willingness to buy a product. Typical formats follow for a specific product test and product niche test:
| Definitely would buy (1) | Probably would buy (2) | Maybe (3) | Probably would not buy (4) | Definitely would not buy (5) | |
| A kitchen craft short video course and a print book with 20-25 ideas for Christmas gifts you can make. All would be made from common household ingredients and could include stovetop potpourri, fingerpaint set, sachets, ornaments, etc. | |||||
| Non-caloric chocolate – paper products with chocolates featured as the tempting design in numerous forms. Could include photos of rich, creamy chocolates to tempt you or puns to make you laugh. |
Test Design
I generally test 30 to 50 new product and niche ideas in each research effort. You can execute this test through email to 4,000 randomly selected active customers in your database (you could try direct mail or phone calls as it will get you as much as 10 times higher response rate, meaning your customer sample could be in the low 100s).
You should mix a variety of product types in any given concept test, i.e., a cookbook idea might be followed by a birthday card design idea, followed by a craft product idea, etc.
You should be able to get about ten ideas to a page (whether online or print), and it is important to rotate the pages so that each idea has equal exposure on page 1, page 2, etc.
In every test, include at least two benchmarks of existing products with proven sales history. Select one good seller and one average seller as benchmarks.
More on concept testing and idea screening here:
How To Choose The Winning Product Ideas for Your B2C or B2B Business (And It’s Not About the Obvious Wet Testing or Lean Method).
Some Warnings about Concept Testing
A promising idea can test poorly if it is written poorly. A capable copywriter with excellent product sensitivity should write the concept statements. The statements should be reviewed by your top product managers before being exposed to customers. Some of your products just cannot be well translated into a simple statement – especially those heavily dependent upon specific use or its complex nature (e.g., B2B solutions).
Showing designs or prices can save or kill a product idea so you may elect to do this. Concept testing for niches and products is a directional guide so it should not be adhered to slavishly. If you feel confident that you can execute a product smashingly, then even if it has performed poorly in research, follow your instincts, and give it a try.
Prototype Testing
This research takes place at the time you have an actual prototype of a new product (i.e., a newly invented gardening equipment), but before you have done a production run or committed to any inventory or serious development.
Objective
This research is designed to determine the customers’ inclination to buy a specific product at a specific price. It allows you to drop or revise products which have low customer appeal.
Format
This is most often a central location (on site) study done with a group of your own customers who are recruited in advance and have specific appointments. Some products may be suitable for one-on-one online interactions or for another, this time more in-depth, online survey.
This depends on your budget, time, and type of products.
In the case of new inventions, you put the prototype (a 3D, real-life prototype, possibly functional) on a table with copy and pricing as it would appear in your promotional materials. Each product is numbered, and customers move around the research room carrying a clipboard with a scoresheet.
They review each product, use it, and score it on appeal, price, and inclination to buy. A typical scoresheet would have this information:
| Very Appealing | Not At All Appealing | ||
| 1. How would you the appeal of the product? | 1 2 | 3 | 4 5 |
| A Very Good Value | A Very Poor Value | ||
| 2. What do you consider the price of this product to be? | 1 2 | 3 | 4 5 |
| Definitely Would Buy | Definitely Would Not Buy | ||
| 3. How inclined would you be to buy this product? | 1 2 | 3 | 4 5 |
| Comments: |
In the case of B2B complex products or services, which normally require a sales representative to make a presentation or conduct a consultative sale, this would have to be provided as well in a similar setting.
Test Design
As in the concept test, you could test from 30 to 80 items in prototype testing. This type of research allows you to test a full line, e.g., you might test all of your Christmas gift wrap designs, or all the products in a new design look.
These products should be grouped together just as they would be in your online or offline promotional pieces.
As with concept testing, benchmarks should be included in each study.
Generally, this research correlates with actual sales at a 70% level, if done properly. The research is especially valuable in pinpointing big winners and big losers. You can react to both situations: winners can get better positioning, more advertising dollars, etc., while losers can be dropped, revised, etc.
It is in this category of losers that the research will be most valuable for you. Often you can find price resistance and reconfigure a product or agree to take reduced margins. On occasion, when a whole new product line has scored poorly, you could follow up with focus groups to determine why.
This could lead you to redesign a product line and turn it from a loser to a big winner. And, of course, toy will often just drop losing niches and products, happy at not having discovered they were losers in the actual campaigns.
Cost for a prototype test may be as low as $100,00 to $1,000 including photography, recruiting, data processing of results, etc. Obviously, replacing one or two losers with better products could easily cover this expense.
To increase the number of respondents you may give participants, for instance, $20 in gift cards and a $40 coupon good on their next purchase. (Although this may skew the results so proceed with caution.)
Prototype tests are a great way to get your customers’ reactions to a variety of new concepts beyond just products. My clients have successfully used them to test new packaging, product ingredients, quantity of product, and size as well as color preferences. Such situations are good opportunities for shut-in performance marketers to get out and meet their customers, in the flesh.
Product and Niche Testing – Actual Selling Tests via Dry or Wet Tests
After all is said and done, the best research tool is an actual selling test. Tests can be conducted which are short of full promotional exposure and which will yield projectable results.
Testing via Your Customer Database
My favorite form of product selling test is a flyer inserted in product packages. Using this technique, as long as you have a large volume of items sold, you can:
- vary quantities,
- turn the test off if there are any quality, delivery or fulfillment problems,
- and roll it out to all packages and make a few dollars if it is working well.
The test is pretty inexpensive, often profitable if you choose to do it wet (i.e., you stock the product and actually sell it), and below the radar for your direct competitors who cannot see it via any ad spy tools or by signing up for your email newsletter, etc.
You may choose to use all the other media channels targeting customers in your database, i.e., by sending emails, SMS texts, robocalls, Facebook custom audiences and retargeting techniques, or perhaps direct mail – all driving traffic to your landing page. Expect a lower conversion rate though than with package inserts.
(There is also telemarketing, but it may be rather expensive and you would need a disciplined team of agents who would adhere to a given telemarketing script or you will not get any meaningful results to base your future predictions on.)
Objective
The objective here is to test, in a real-world buying situation, new product niches to your own customers. A secondary objective is to roll out winners and make some money.
Format and Test Design
I generally recommend long form landing pages or long copy inserts. This often means at least 2 pages worth of copy and pictures. This format let’s you present all your selling points – if you cannot make a sale this way, your product may be a loser.
On the other hand, your test format should imitate your normal promotional pieces.
If you are into short copy and video, then by all means do use short copy and video in your test format (this could mean an insert or SMS text driving traffic to a landing page with a video presentation.) Some of you may find it best to actually ‘self-plagiarize’ your winning promotional pieces and do as little as possible to squeeze your new product in a proven, winning copy.
Remember to establish benchmarks with proven sales history: one good seller and one average seller.
Online Display Advertising: Google Display Ads and Facebook Ads
Depending on your business model you may want to run online display ads tests before putting a product in your winning offer.
If you get most of the traffic this way and you are able to convert it for your usual products, do use online display ads. Watch out for your competition though – they will be using ad spy tools to discover your campaigns.
If you would like to see if your new products would recruit brand-new customers, i.e., such that may differ in psychographic terms from the ones you already have in your database, online display ads may be the way to go. Especially if you exclude the customers that are already in your database in your campaign settings.
As a matter of fact, depending on your niche and product testing objectives, you may want to have a standalone testing process using Google Display Ads or Facebook ads.
Occasionally, even if your business model does not employ generating traffic through the above methods, you may use them more for the sake of expediency than for any great desire to test this media channel.
For instance, if your product does not warrant an insert flyer / SMS text / email etc. test campaign, you simply have no time, or your customer database is not large enough, you can often rush it into an online display ad. This is a gamble since, to pull meaningful response, you need a fairly large traffic volume. My consulting clients had big successes and big flops doing this. Still, worth trying on your own.
Research for Forecasting the Sales of Your New Products
In addition to the product and niche studies described, I recommend doing extensive research to help forecast sales by product. As with all businesses, the profits are dramatically affected by the accuracy of our sales forecasts. This is especially true if you are also a manufacturer or create the products on your own. See the following system of early ecommerce sales forecasting research.
Objective
The objective is to obtain an accurate and very early reading of sales by product for an entire huge offering of an ecommerce business. For instance, let’s imagine you plan to launch a 6-month campaign offering 40 new products in a multitude of niches.
Format
You can direct a panel of your current customers to a ‘secret’ landing page or a ‘secret’ ecommerce platform before your launch. This effort could give you a reading, for instance, about two to four months before your normal results. While this would require additional budget to conduct, for instance more than $5,000, you could make that amount up by accurately identifying just two or three big winners or losers.
Test Design
Your 2,500 respondents panel receives an email, a phone call, an SMS text etc. asking for participation in a research project. Of these, about 1,300-1,500 actually will fill out the ‘order’ form.
Each participant receives his or her access and instructions (which indicate that this is not a real order); a $10 coupon good on the next order; a space for comments, and – optionally, if you do not have an accurate information in your database – a demographic questionnaire to ensure that your sample is representative of your whole customer database.
Name and address are optional, but most people do supply them, and you can use this information to obtain early testimonials.
You ask participants to respond to the test just as if they were making an actual order. This can be tricky, particularly if you are testing seasonal items well in advance of the time it is usually purchased.
This research gives mix percentage points only. Overall percent response cannot be projected from results because the participants are pre-screened and such a high number of them respond. However, you will get excellent information regarding units per order, and this is invaluable in generating a macro forecast. Rankings of products by research sales have usually better than a 90% correlation with actual sales.
One area in which correlation is lower is high-ticket items. Customers tend to order these more in research than in actual buying situations.
Additional Benefits of Early Research
Aside from providing critical forecasting information, I have found two other benefits from doing very early research.
- You can adjust pricing in response to a product’s research performance. Occasionally, if you have badly overbought or overproduced, you could slap a “sale” banner on a product and take your markdowns in your first showing.
- Collect early testimonials if you decide to send products free of charge or – if they are high-ticket items – you ask for their return after they have been reviewed for a testimonial.
Summary
- Obviously, not all your products should go through all the research stages described in this article. You should be most complete in testing products requiring capital equipment and those furthest afield from your proven product line and niches.
- Most often, you will put products in the offer that have undergone only one, or sometimes none, of the steps outlined. In an effort to stay with the marketplace, timing often dictates quicker and bolder moves than extensive research allows. Still, early product testing is vital to any product development program and will often save you from costly inventory and equipment errors.
- The testing described is not applicable to all businesses, as the economics and distinct business models add a dimension to every business’ situation. However, all businesses want to improve their new products’ batting average and the techniques in this article, in total or in part, may help achieve that goal.