How to do Social Selling Properly (for the benefit of Facebook, Google and Twitter...)


Download article:


Social commerce has been a bit of a struggle so far.

Lots of companies are trying lots of different models, but so far the market seems to be fragmenting into a number of niche segments. That is all OK, but you’d not be able to point at the social commerce space as it stands today – sites like Airbnb, Vayuable, RelayRides etc. – and use words  like ‘transformational’, ‘disruptive or ‘wow’.

There’s nothing super-interesting out there: there is no evidence of a new online service category emerging from all the experimentation. 

Over the last year, some companies have dropped out of the market, others have downsized and some have repositioned their models. Still, a healthy number are still actively developing their businesses.

Here is a summary of where we are right now:

Title:Social Commerce: Market Segmentation
Sub Title:Summary of the seven main types of social commerce
Companies:Getaround, RelayRides, WhipCar, Loosecubes, Airbnb, onefinestay, ezCater, TaskRabbit, Vayable, Venutastic, Sportaneous, Honest Buildings, Etsy, Houzz, Yardsellr, Flightfox, Zopa, Kiva, One Kings Lane,, ModCloth, Shoe Dazzle, Kickstarter, Quirky, Threadless, Wikets, Svpply, Copious,, Ruby Ribbon, Kitsy Lane
Markets:Social Commerce

There are currently 7 different types of social commerce business model: 

  1. Resource sharing marketplaces;
  2. Product & service marketplaces;
  3. Curated online retailers;
  4. Production communities;
  5. Product discovery communities;
  6. Directories & listings portals;
  7. Peer-to-peer selling platforms.

To us the last item on the list is by far the most interesting. The value of the online retail market in 2012 was USD 440 billion, and we project that this will increase to USD 668 billion by 2016.

Rather than try to create new segments - such as resource-sharing marketplaces – an alternative social commerce approach is to shake up the online retail market.

Online retail is pretty much the same as it was 10 years ago: the business model is established, the payment and product discovery infrastructure is in place and consumers understand and accept online retail as a major, sustainable service category.

So the question is whether online retail is mature? That is, is the user experience fixed, is it as good as it could be? Is there really no other business model?  Is this really is good as it gets?

I think that the answers to these questions are No.

For example, surely there must be a way to fuse online retail with social relationships so that one peer could sell a product to another peer? I am not thinking about selling used goods, like on eBay. I am thinking about selling new merchandise. So every person on the planet could become a retailer and, with wireless devices, every face to face interaction could become a point of sale.

If this were possible, then the online retail market would explode into billions of potential sales points and we would have an army of ‘social sellers’ who would be focussed on selling the specific products and services they knew about and were passionate about.

When we think about social selling, this is what we are thinking about.

If the market were to evolve in such a direction then in years to come we would look back at today’s online retail ‘shop’ model as being very crude.

Surprisingly, Google and Facebook do not appear to appreciate the potential of social selling:

Facebook seems unable or unwilling to implement any revenue model that does not involve selling advertising while Google is focussed on better enabling online retail in the wider sense, using products like Google Wallet and Google Shopping.

So how might social selling work?

How might social selling work if introduced on a site like Facebook, Twitter or Amazon?

The best place to start is with an example of where one individual sells something to another individual.

Example purchase: handheld GPS receiver

Imagine that someone wants to buy a handheld GPS receiver which will be used as a navigation aid for ski touring in poor visibility.

Such a person might type the following query into Google Shopping: “handheld gps receivers”. If you type this query into Google and select ‘Shop for handheld gps receivers on Google’ then this is what you will see:

Title:Goolge Shopping Search for Handheld GPS Receivers Screenshot
Markets:Retail, Online Retail/eCommerce

When I tried this on 10 October 2013, I was presented with 1,790 different results.

Here are some of the problems I can see with Google’s approach to online shopping (above):

  • The category options (left hand side of the above screenshot) are too broad and do not include the intended use (ski touring in poor visibility);
  • The buyer does not necessarily want the lowest cost device. Instead, he (that means me!) wants the device best suited to the job, so the ‘list in price order’ tab on the right hand side is not of much use.
  • Many of the reviews are posted anonymously. Although there are links to the sites that the original reviews were posted on, there is often no way to locate the original review post, or to learn more about the reviewer – for example, how this reviewer is rated by others.
  • We know that there is a lot of comment spam on the web, so how do you really know if the review you’re reading was written by a competitor or the vendor? Can you really trust a comment review by “LopseyDriver’’ Well, oftentimes you can’t.
  • Attempting to refine the shopping search by typing “handheld gps receivers ski touring” results in “Your search - handheld gps receivers ski touring - did not match any shopping results.”

From personal experience, this particular shopping requirement is quite complex and throws up a number of questions:

  • Temperature performance: The manufacturer’s recommended minimum operating temperature for the Garmin GPSMap 62ST is -15 degrees, but the application requires the unit to be serviceable at -25 degrees. What will really happen at this temperature?
  • Custom maps: The user wants to import their own map which has major avalanche zones highlighted. How will this look when viewed on the unit? Is it possible to view the user map and unit-supplied maps separately? How complex is the process for importing custom maps and what are the pitfalls?
  • Imported data: Is there a limit on file size for imported maps? If the user wants to draw his own route and import that into the unit before a tour then what are the limitations of using device-supplied software, vs. third party software to do that job?
  • Videos: Is there any further information or videos that show the product being used for my intended application?
  • Signal Coverage: What should I expect when using a GPS device in the high mountains? Are nearby mountains likely to affect the signal reception?

In this case, the manufacturer’s standard documentation (which is pretty good) is inadequate. User forums are not much better and the only way to find the answers to these questions is to buy the product and find out the hard way.

Even product comparison sites like, PocketGPSWorld, while good, do not provide the level of detail needed to properly answer the above questions. This particular site simply summarises manufacturer-supplied specifications in a tabular format and does not contain any original content of its own.

My purchase experience: sub optimal (understatement...)

The only way forward (I discovered) was to take the plunge and buy what I thought was the best unit for the job and try it out – but I was not really sure at the point of purchase.

In case you’re interested I ended up buying the Garmin GPS MAP 62st. For my requirements, this is an OK product but it has these flaws:

  • The software is terrible! This is a perfect example of a company that does a great job making radio electronics trying to do software. The menus on this device are really confusing and it’s a pain to use.
  • The custom map feature (which I really wanted) has an abysmally low memory capacity – like 5MB, in total. So when you scan in your chosen map segment then you will probably find it will not fit onto the device. So you then have to reduce the resolution using an iterative process so it does fit (there is no way of knowing what the available memory capacity is – you just get an error message if you try to upload an image that is too large). When you finally manage to load your custom map onto the device then you find the image resolution (combined with the poor screen resolution) means that you cannot see the the detail you can on your PC.
  • When you are viewing a custom map there is no way to view only the custom map OR the in-built map: you HAVE to view one overlaid on top of the other, which just means that the display is a mess. So custom maps did not work for me.
  • Because many of the features I really wanted do not work properly, I now only use my device for tracking routes so they are available as a back-up navigation aid.

So my question is this:  is there any way to improve this purchase experience?

How to improve this purchase experience

Option 1: Conventional retail concept: online shop (quite interesting)

I think that there would be an opportunity for a mountain guide or some other outdoor professional to create a really excellent social shop with very detailed application information on the various devices and their pros and cons for ski touring use in the mountains.

This would include one critical item of equipment – the handheld GPS.

Quite simply, nothing like this presently exists on the web, but for this particular use case, my experience suggests that it would be really valuable.

If such a person could then sell his favoured handheld GPS products directly from his online shop then we think that this could be interesting, for him and buyers.

We are talking here of a site that goes well beyond what one would expect to find on a review site or discussion forum: a social shop for handheld GPS receivers would be created by a genuine expert user who has very detailed information and insight into the product area and its real-world application.

Option 2: New retail concept: virtualised retail (very interesting)

Taking the ‘social selling’ idea to its logical conclusion would mean that sales would not be limited to random users finding this particular social seller on a social selling site: they would be possible anytime, anywhere.

For example, it would be possible for the mountain guide to sell to the group that has paid him to take them ski touring for a week.

Let’s imagine that this guide is taking a group ski touring trip and they are half-way into their ski tour having dinner together one night in a mountain hut.

It’s pretty likely that at some point the conversation would turn to ski equipment.

It is also pretty likely possible that some of the group members would be interested in the guide’s personal equipment, including his GPS.

A social selling model would allow one of the members of the group to use their smartphone to buy their own GPS device from the guide. All that would be required would be for the buyer to visit the guide’s social shop and buy the GPS from him.

The guide would be able to explain how the device is used – far better then any retail sales person – and the client would be able to see it being used in a mountain environment. Most important of all the guide could explain to the client the device’s strengths and weaknesses down to a level of detail that would be simply impossible to replicate in a retail store, let alone by a standard online retailer.

The advantages of a “social sale” such as this, compared with a traditional retail sale (offline or online), are that:

  • The client is assured of getting the best handheld GPS receiver for his needs;
  • The guide earns a retail margin on the sale.

It is clear that this selling concept could be extended to other items that the guide uses – rucksacks, storm jacket, climbing harness etc.

In this way, the mountain guide’s personal equipment and attire could become an ‘always open’ outdoor equipment retail store that he would carry with him on all his trips to the mountains: the guide would know that that his clients would be able to buy his favourite and most trusted equipment from his social shop at any time – using their smartphone, or his.

To reiterate, the incentive would be that the social seller (guide) would earn a retail commission and the buyer (ski touring clients) would get products that fully met their needs. 

And it is also clear that the same concept could be successfully applied to many, many merchandise categories.

Business Model

I see an opportunity for a new form of online selling model as shown below. The social selling ecosystem would comprise the following groups:

  • Market Maker: Acompany would need to create and operate an online marketplace connecting merchants, socials sellers and buyers. The offline equivalent would be a company which wants to build and operate a shopping mall.
  • Social Sellers: Individuals who act as value-added ‘enthusiastic boosters’ for a specific category of merchandise they know and love. In the offline example of a shopping mall, these would be those who would run the shops comprising the mall.
  • Merchants: Companies which have products and services they want to sell and already have an online presence which allows them to fulfil orders directly.
  • Buyers: Individuals who purchase products and services online.

Social sellers, merchants and buyers would operate within a curated ecosystem that would be facilitated by the market maker.

Title:Social Selling Model
Sub Title:Illustration of relationship between merchant, seller and market maker
Companies:Facebook, Twitter, Amazon
Markets:Social Commerce
Title:Social Selling Ecosystem
Sub Title:Summary of roles of main players
Markets:Retail, Social Commerce, Online Retail/eCommerce
Other:The social selling model described here is very different to established affiliate marketing programmes. Here are a few of the many differences: (i) The focus is on individuals not companies; (ii) the platform would be a branded service sought out by buyers; (iii) buyers could choose their preferred seller; and (iv) sellers set their own retail prices.

More detail on how the ecosystem could work

Market Maker

  • The Market Maker’s website would consist of a centralised e-commerce platform that could support many individual online shops;
  • This platform would have connections to a range of merchants who would list their products by publishing standard information over an API;
  • From the user’s perspective the site would resemble a shopping mall containing many separate shops, each focussing on a specific category of merchandise;
  • Each shop would have a built-in shopping cart functionality that would allow sellers to sell their products direct to their customers;
  • If the Market Maker was a start-up then a decision would need to be made to initially focus on a particular category of merchandise. If the venture was a success, then other categories could be added later. Amazon – currently the world’s largest online retailer – started out with printed books;
  • Later, the Market Maker could arrange for social sellers to expand their ‘retail footprint’ by adding a simple storefront to their profile pages on sharing sites like Facebook, Twitter, YouTube as well as their blog and website;
  • In return for setting up and running the social selling marketplace, the Market Maker would earn a commission which would be deducted from each transaction that passed through the marketplace.

Social Sellers

  • Individuals  would have to apply to become  ‘social sellers’;
  • Selection criteria would include aspects such as:
    • Real person (e.g. Facebook, LinkedIn profile etc.);
    • Not affiliated with any of the merchants that would be listed in the seller’s online shop;
    • Commitment to write knowledgably and passionately about products and services the seller has direct personal experience with and/or knowledge of;
    • Evidence of an existing online presence within the particular merchandise category that the seller wants to focus on.
  • If accepted the social seller would then use an online interface to select from a range of products in order to stock their online shop;
  • If the product they wanted to list were not available in the merchant’s catalogue then, assuming the merchant has included a suitable ‘Add’ button to their main internet website, the seller could add the relevant products manually;
  • The social seller would be able to use tools to customise their online shop;
  • The social seller would then create accurate, trusted value-added content that will draw customers into their social shop:
    • Detailed information on how the product is used in practice;
    • Aspects of the product’s performance - favourable and otherwise;
    • Interesting usage modes that might not at first be apparent;
    • Detailed pictures and videos showing the product in use;
    • Information about how to use the product in cases where the merchant’s documentation is not as clear as it could be;
    • How to deal with problems such as upgrades or use with other products;
    • For online services, information could be provided about how to use features that are not obvious;
    • For other services information could be provided about details of the terms and conditions.
  • This information, which could be thought of as the equivalent of a marketing brochure or product advertisement would be written by the users themselves who would have no formal relationship with participating merchants. This would mean that there would be no paid-blogging or product-plugging using shady intermediaries, as is common elsewhere on the web;
  • This content would then be indexed by search engines and shared using standard social tools;
  • Social sellers would earn a percentage of the transaction value for each sale;
  • It would be possible to put in place a tiered commission system, so that that “top rated” social sellers would earn the top commission – say 20% – but bad feedback could have them downgraded to say 5%;
  • We can see some users earning a living from becoming respected product experts in specific areas (e.g. baby products, garden supplies, interior design, kitchenware etc.). Importantly, user reviews would be subject to public critique, in the form of use ratings and would also be complemented with comments from other users. If someone purchased a product from a social seller but then found that the product’s benefits were not accurately described then that user/seller would be subject to adverse feedback;
  • With their smartphone or tablet in hand, a Facebook Social Seller could, at an opportune moment (perhaps at a dinner party where the discussion has turned to the merits of a particular baby swing), show a group of friends a review they had written. They could then close a sale in real time.


  • Merchants would view the social selling platform as a new online sales channel which would increase overall sales and which would work alongside their own, direct online retail shop;
  • Merchants would list their products using a standard API, similar to Google Shopping, except that listing would be free;
  • Merchants could choose to add an item of code to their product description pages, so their products could be manually added to the market maker’s site by social sellers who could click on an “Add” button;
  • The merchants would receive orders over a standard interface. The ordering information would be the same as that which would be collected by the merchant via their existing online shop (e.g. details on the product/s ordered, shipping address, customer details etc.);
  • Merchants would fulfil orders using the infrastructure that is already being used to fulfil orders that are received from the merchant’s existing online shop;
  • The social selling model could be operated on a local basis. Local merchants could create listings on a social selling site and then local people who had personal experience of those products could write reviews and provide comments about those products based on actual experience;
  • Merchants could receive the same proportion of the sale price as they do with their other online retail channels.


  • Consumers who were interested in buying a given product could visit the Market Maker’s website and search for that product;
  • Alternatively, prospective buyers could find a relevant social seller by typing a query into a search engine, or asking friends on their social networks for a recommendation;
  • The prospective buyer would then see a list of “social sellers” who have covered that product;
  • The prospective buyer could then browse their material presented by the different social sellers and ask questions;
  • In the event of a sale, the buyer would have the option of purchasing the product through their chosen social seller;
  • This would mean that the social seller would receive a percentage of the retail margin that would be shared between the social seller and the company which owns and operates that online marketplace.

But isn’t this how an affiliate marketing programme works?

No: the social selling model described here is very different to established affiliate marketing programmes. Here are a few of the many differences:

  • Individuals not companies: Many affiliate marketers are companies which re-sell products and services provided by others. The social selling model is focussed on individuals;
  • Branded service: The platform provider would become a recognised brand – like Facebook, Amazon or Twitter for example - and individuals could then use value-added search and recommendation features offered by that brand to find products and services. In affiliate marketing, the affiliate network sits in the background and is invisible to the buyer;
  • Buyers can choose their seller: Having found a product or service using the social selling platform, buyers could then choose from a range of social sellers and then ‘award’ the sale to the seller who the buyer felt had done the best job of explaining the product/service and its benefits. In affiliate marketing, buyers do not have the option to select;
  • Sellers set their own retail prices: The social seller could set their own price which would also allow some price competition in the market: the platform operator and the merchant would get the same price regardless of the price the seller chose. In affiliate marketing, the merchant sets the retail price and all affiliates then have to sell at that price.

Impact on Existing Retailers: Online and Offline

A full implementation of the social selling concept outlined in this report would create millions of individual sellers each of whom would act as a champion for the products they know and love.

Deep personal experience with their merchandise category, passion for their subject and a strong preference for particular products would result in a global network of natural sellers that would clearly challenge all retailers, both offline and online.

A successful social selling market would result in a structural, permanent reduction in the revenues and profitability of the entire online retailer sector.

If a social selling programme were implemented by Facebook, then it would bring Facebook into direct competition with all online retailers, and especially Amazon, eBay and their competitors.

  • Comparison with offline retail: We think that the typical individual within such a ’social selling’ network would far outperform the equivalent person who works in a bricks and mortar retail store. Most front line sales people lack detailed knowledge and impartiality: they are often remunerated on a commission basis and do not really care if they sell a product that they themselves do not regard very well. After all, sales people are trained first to sell. They are not paid to be totally open with prospective buyers, or to spend too much time explaining a product’s benefits;
  • Comparison with online retail: When we look at online retail today it is easy to see a number of challenges: product descriptions are often standardised across the web and lack detail. The online shops of many companies often resemble little more than an online product catalogue. In cases where user ratings can be viewed, the buyer is often unsure who has posted those reviews: who is the anonymous user “jakwwky3444”? Can the buyer really trust this person? Are they independent? Do they know what they are talking about? Has the merchant paid this person to post a favourable review?

In the current retail marketplace thousands of companies exist to sell products that are produced by others (merchants). Retailers like Comet in the UK focus on consumer electronics. Staples sells office supplies etc. With a Social Selling model the number of ‘sales points’ explodes into the 100s of millions – because every online could become a social seller.

If social selling became established, and we think it is only a matter of time before it does, then existing online retailers, like Amazon, would need to move quickly to catch up. And just as the transition from offline retail to online retail has resulted in the failure of some offline retailers we see the introduction of social selling resulting in the failure of a number of online retailers.


Social selling seems like a great idea: it is based on a user experience that could, for many product and service categories, be far more effective than what is offered by existing online and offline retail models.

In addition, the required ecosystem appears sound because it offers clear benefits to merchants, buyers and sellers.

The main problem is that this would be a model where it is not interesting until critical mass has been achieved, which is why start-ups like Copious have had to retrench and try something else.

But if a site that already had a higher user base, and which did not have an existing online retail business to worry about, decided to dive in then things could be very different, very quickly.

The obvious candidates are Facebook, Google and Twitter – as well as their equivalents in other markets.

When will one of these companies start thinking about transforming online retail?


Andrew Sheehy
Chief Analyst

Generator Research, Ltd.
Petitor House
Nicholson Road
Torquay, TQ2 7TD
United Kingdom

+44 (0)208 144 7073 (Direct)
+44 (0)7866 387 715 (Cell)
+44 (0)208 711 3065 (Fax)

Skype:      aaapps

Contact Andrew Sheehy at or follow him at