Bol.com

Product research through keywords on bol (with the Niche Explorer)

Lars Hurkmans·Co-founder·28 January 2026·Reading time 13 minutes

Product research through keywords starts with consumer behaviour: keywords show intent, so what people search for on bol. But search volume does not prove sales. That is why we use the Niche Explorer, a filterable keyword database enriched with revenue, competition and seasonality, and filter per price segment from the table down to a concrete niche.

This is part 4 of our product research series. In part 3 we set up the framework for product research on bol, in which you choose between research through keywords or through categories. In this article we work out the keyword route in full with the Niche Explorer. The broader overview is in our piece on market research on bol, and in part 5 we look at whether there is room for a new player.

Why do you start product research on bol from keywords?

You start from keywords because they show consumer behaviour: what people search for on bol, and so what intent there is in the market. The limitation is that search volume only measures that intent and does not prove that products are actually sold.

People search on bol in a great many ways, and from those keywords we extract a lot of information about demand. The downside: a keyword mainly says something about general demand and the intent to buy something, not whether consumers actually buy it. For that you need other data, such as the real sales figures. That data is not accessible by default, so you need special software or a product research tool that makes it visible. In this article that is the Niche Explorer.

What is the Niche Explorer and what data does it contain?

The Niche Explorer is a filterable keyword database with the relevant keywords on bol, so the exact terms consumers search for. That way you see in one overview what people search for everywhere, and with it the opportunities that exist, purely from consumer behaviour.

Besides the keywords themselves, the Niche Explorer enriches every keyword with extra data: the revenue of the products behind that keyword, how many sellers there are, how those products are offered, how good those sellers are, how those figures change over time and at what prices products are sold. That information sits compactly in a table, so you can do research quickly. For that we use the Niche Explorer of MarktMentor.

The table contains the following columns. These are the basic KPIs with which you assess a keyword.

How do you assess demand with search volume, revenue and seasonality?

You assess demand with three columns together: search volume, total revenue and seasonality. Each column tells you something different, and only in combination do they give a reliable picture.

• Search volume. How many people search for the keyword. This says something about intent, but not about value.
• Total revenue. The hard numbers: how much money goes round in the products around the keyword. The caveat is that it concerns all products shown for the keyword, so there is correlation, not causation. Still, it often gives a good picture, especially if the keyword is specifically about a certain product type.
• Seasonality. Whether a keyword is evergreen, so stable throughout the year, or strongly dependent on a season such as winter, spring or autumn.

Next to the revenue and the search volume there is a trend indicator: a green or red arrow. Green means a positive trend, so more revenue or more searches than in the comparison period. You choose that comparison period yourself through the filters. By default it is set to 30 days, but you can switch to three months, a year or separate quarters, which is handy if you want to plan ahead. If you adjust the period, the data changes with it immediately.

Why do you research per price segment (high, mid and low ticket)?

You research per price segment because the selling price determines how much starting capital you need. A product with a high selling price usually requires more starting capital, while a cheaper product can often be done with less.

That is why the first split in this research is the price segment, and only then do you look at demand within that segment. In this article we go through three segments: high ticket (from 100 euros), mid ticket (the middle segment) and low ticket (the cheapest products). Those boundaries are our own choices to organise the research, not a fixed classification by bol. For each segment we fill in different filters and see which niches come out.

How do you filter a high ticket niche in the table?

You filter a high ticket niche by first narrowing down the average price and then restricting the revenue and the search volume. That way you reduce tens of thousands of keywords to a short list you can assess.

The steps from this article, with the filter values used as an example:

• Average price between 100 and 300 euros. The average price column looks at the products on the first search page, the most relevant page. After this filter, mostly brand-driven keywords remain, such as Samsung or e-readers, and those are not necessarily the products you want to sell yourself.
• Time period set to a year. By default the period is set to 30 days, but with a year of data you support your decision better.
• Total revenue between 1 and 10 million euros. An upper limit on revenue sounds counterintuitive, because you want revenue after all. But the keywords with the very highest revenue, the top of bol, often revolve around products where the name recognition of a brand plays a big role. As a private label seller those are harder to offer. With a lower limit (here 1 million euros) you also make sure there is enough market value in it.
• Search volume between 30,000 and 70,000 per year. The maximum again keeps out the overly specific brand keywords you cannot offer.

After these filters, 82 results remained in the example. By default they are sorted by relevance, which is mainly useful when you type in a specific search term. If you sort by revenue from high to low, you see at the top the keywords with the highest combined revenue, in this case the Philips Airfryer. The first keywords are strongly brand-related, so the next step is to look for keywords that describe a general product niche. You stay as general as possible and go deeper from there. In the example, pan sets and rotary washing lines stood out as more general niches.

What do the quality score, competition and dominant category tell you?

The quality score, the competition level and the dominant category together tell you how strong the supply is and whether the keyword represents the niche well. You read them in the table per keyword.

The quality score reflects the quality perception of consumers, based on the reviews. It is a score between 1 and 10, calculated from the review score and the number of reviews of all products behind the keyword. Many good reviews produce a high score, few or bad reviews a low one. Products without reviews get 5.5, the middle. Above 5.5 the scores are mostly good, below it relatively worse, and the closer to 10, the stronger the perception. For pan sets the score was just below 5.5, so there were probably products with lower scores among them.

The competition level for pan sets is high. That means the sellers are probably strong, think of bol offering the product itself or of big, well-known brands. If you want to offer that product, you compete with those brands, and the question is whether you want that. Do not only look at the number of competitors (around 3,000 products for pan sets), but also at how good they are and what types they are.

The dominant category is the category in which consumers mainly buy the products. That matters because the Niche Explorer is about keywords, so about intent, and not about the product people ultimately buy. For pan sets the dominant category is the same as the keyword, so that keyword is representative. For rotary washing lines the dominant category is drying racks, a different name from the keyword, which also explains why you see a difference in the search volume there.

The seasonality also differs between the two. Pan sets peak in summer, with a peak month of July that is very strongly seasonal, so that month deviates sharply from the rest. Rotary washing lines peak in spring, with March as the peak month, but moderately seasonal, so less pronounced than with pan sets. As for the type of demand, you preferably look for growth, and you do not base that on one keyword: for rotary washing lines the search volume dropped slightly, while revenue actually grew. So look at the revenue and the category behind the keyword, not at one single search term. How you read such a historical curve, so whether something is growing, seasonal or a hype, we work out in historical data and product choice on bol.

How do you go from the table to the niche details and the supply?

From the table you click through to the niche details, where four charts show demand more precisely, including the market value, the sales and the search volume. There you confirm or nuance the first impression the table gave.

For the rotary washing line from the example, you first set the view up conveniently: from lines to bars, the time period to the last year (preferably even two years), and a comparison with the previous period. The charts show all products behind the keyword by default, but around 70% falls in the drying racks category. If you filter on that dominant category, you are really looking at the niche. The market value and the sales then show growth with little seasonal effect, while the search volume follows a different pattern: there was one high peak in May 2024, which makes the trend come out as a slight decline, and search behaviour is somewhat seasonal (higher in the middle of the year). If you set the frequency to calendar month, you see that more clearly. The lesson: rotary washing lines are seasonal in terms of search behaviour, but not entirely based on the market value.

Then you assess the supply, at the top left of the details. For the rotary washing line it was more than 3,000 products, spread over 76 brands and 315 sellers. The number of brands is therefore much smaller than the number of products, and ultimately you compete with the brands and sellers. Around 8% of the products were unbranded, which is interesting: unbranded products have no name recognition, so with a good listing you already start with a head start there. In addition to unbranded, a single better-known top brand also came by, and bol itself was the seller of around 11% of the products. That points to a mix of a cheaper and a more premium market. Because there were still quite a few top brands in there, the conclusion in this article is to leave the rotary washing line for now and look at other niches first.

Back in the table, on the following pages, the foldable treadmill stood out. It showed strong growth, around 75%, higher than the rest of the page, with a growth trend in search volume too and a fairly high price. After the same analysis (top category treadmills, growth in both search volume and market value, slightly seasonal with a peak around January that may partly be a hype), the supply turned out to be a lot lighter: 218 products, 50 brands and 79 sellers, with unbranded products among them and bol far less prominent than with the rotary washing lines. Through "analyse products" you can then filter down per product until you end up with one product. The foldable treadmill therefore stays on the list as a high ticket candidate, with the caveat that high ticket requires a lot of starting capital.

How do you approach a mid ticket niche?

You approach a mid ticket niche the same way, with the filters one segment lower. You first reset all filters and build them up again, fitting a lower price and revenue level.

In the example: average price between 50 and 75 euros, time period set to a year again, total revenue between 500,000 and 1 million euros, and a maximum of 500 competitors. That last filter is new compared with the high ticket example: by filtering on the number of competitors up front, you do not have to stumble upon a niche with light supply by chance, but steer towards it directly. With an additional search volume filter (between 10,000 and 30,000), 34 results remained. If you filter too strictly, you become too specific and have to loosen the filters again a bit, so you play with that.

From this the outdoor wildlife camera emerged: revenue of around 700,000 euros, a selling price around 75 euros, a search volume of around 14,000, an average competition level and a quality score above 5.5 but below 7.5 (only above 7.5 is it really strong in quality). On the detail page the market value showed strong growth, with sales following, while the search volume was actually low in recent months. That means the keyword "outdoor wildlife camera" has become less relevant, while the market (the wildlife cameras category) is growing. The strong seasonality label also turned out to be caused by one month: viewed over two years it is less seasonal than the peak month suggests. That shows why you always check the labels from the table on the detail page. The supply was 419 products, 34 brands and 44 sellers, with many unbranded products and without bol as a seller, which indicates that the very biggest brands are absent here. Enough to remember the wildlife camera as a second candidate.

How do you approach a low ticket niche?

Low ticket on bol means the products below 20 euros, and we choose that boundary because the commission works differently below it. Above 20 euros a more general rate applies that is the same across all categories (a fixed amount plus a stable percentage), while below it the rate can vary.

You could split low ticket further into very low (below 10 euros) and medium-low (between 10 and 20 euros), but in the example we look broadly below 20 euros. After the usual year of data came the filters: total revenue between 100,000 and 500,000 euros (lower than in the previous segments, because the prices are lower), a maximum number of competitors, and two new filters. We set the seasonality filter to evergreen, so you only see niches that stay stable all year round, without strong fluctuations. With a search volume between 10,000 and 20,000, 115 results remained, with many functional keywords.

From this "blue light glass" stood out, which turned out to have computer glasses as its dominant category. Here a column comes up that we had not used yet: the average winning bid. That indicates roughly how expensive advertising will be. In this example it says 1.34 euros, so you pay on average 1.34 euros per click you get through that keyword, around 7% of the selling price. Whether that is high or low depends on your conversion: with a high conversion you need few clicks per sale and it is manageable, with a low conversion it turns out very differently. On the detail page the market value grew (sales slightly less fast), but the search volume showed a big difference: the keyword "blue light glasses" had a peak around two years ago that did not return afterwards, so that term seems to have gone somewhat out of fashion, while the market for computer glasses remains in demand. People apparently search for it differently. The seasonality was nuanced: the search volume looks seasonal around the start of the year, the market value more around December, although the label said evergreen. Among the sellers it stood out that one brand (Gunnar) also appeared as a seller, which indicates that it sells its own products. This niche too is interesting enough to remember as a low ticket candidate.

What do you research next?

After the Niche Explorer you have a number of candidate niches and a list of follow-up questions. With three ways of filtering you found three very different products in this article: a foldable treadmill (high ticket), an outdoor wildlife camera (mid ticket) and a blue light glass, in other words computer glasses (low ticket).

Those three niches and all the figures mentioned are illustrative examples from this article to show the method, not recommendations and not results to expect. For each candidate you want to know more next: how exactly the prices are distributed (so you know whether you can sit higher or lower than the average price), what kind of brands there are (top brands or private label), whether they are big or small products (that determines the shipping costs) and how expensive advertising really is. After that comes the question of whether you can enter as a new player and on what you can differentiate. We cover that in part 5, in which we look at whether there is room for a new player.

Summary: from keyword to niche

• Keywords show intent, not sales. Start with consumer behaviour, but back up demand with revenue and market value, because search volume alone does not prove that people buy.
• The Niche Explorer is an enriched keyword database. The table shows per keyword the revenue, average price, search volume, competition, dominant category, average winning bid and seasonality.
• Assess demand on three columns together. Search volume (intent), total revenue (value) and seasonality, with the trend arrows and a comparison period of at least a year.
• Filter per price segment. High ticket, mid ticket and low ticket each require different starting capital; narrow down the price and then restrict revenue, search volume and the number of competitors.
• Read the supply, not just the numbers. Look at the share of unbranded products, the types of brands and the share of bol; that determines how heavy the competition really is.
• Check the labels on the detail page. One peak or one keyword can give a misleading growth or seasonality label; the four charts and the dominant category nuance that.