Skip to the content

2019 - Moving Forwards - Retail


Retail is in turmoil.

Online sales are increasing but at lower margins and higher cost demands.

Offline sales are decreasing at dramatically lower / pencil-thin margins and footfall is slowing.

Retailers are failing and (brexit induced) lack of UK investment is preventing a number of them from being saved and revitalised.

There is no magic answer but retailers need to start re-engaging customers and re-thinking their strategy.

Loyalty Programmes

When there were few programmes around and the rewards were greater loyalty / reward programmes worked.

In the last few years we've seen a proliferation of new schemes and with decreasing benefits amounting to a percent or two.  Barely of any interest to customers and leaving them feeling cheated when they've invested in the scheme (and therefore brand loyalty) only to be given reduced benefits.

Good loyalty schemes do a few things:

  • Are simple enough for a customer to understand
  • Give loyal customers a reason to return
  • Encourage customers to try out another services/product offering (e.g. food shoppers can be encouraged to try the clothing range)
  • Create a "random delight" by giving customers something they weren't expecting / different to others in the market
  • Increase physical or virtual footfall
  • Help customers to talk to their social network about you
  • Provide more revenue or more margin per customer.. or both
  • Move the discussion about purchase/utilisation from price to value
  • Gives you direct communications to the loyalty user base
  • Remind customers of the value of being a member

Over the past decade a number of "turn-key" loyalty programme packages have arrived but they're really a base to work upon.  Our most successful customers have loyalty programmes have bespoke elements to handle their unique USPs and offering. 

So, how to cure it?  It's not easy!  

First of all think carefully if you need a formal loyalty program; LIDL and ALDI both do fantastically well without a formal loyalty programme.  What's their loyalty based upon?  Price, quality and availability.  Take a look at the key criteria above, ALDI in particular handles most of the above without any formal programme.  By not having the costs (technology and human costs) they can decrease costs and therefore either increase margin or pass those savings to customers.

Secondly, why is a customer likely to shop with you?  For 1% return?  Forget it!  

Third, is your "reward" what the others already offer?  We've got customers who offer "free" delivery to their loyalty programme customers.  Amazing.. however their competitors offer free delivery so how does this increase loyalty?

Fourth, use your personas/segments to understand your customers properly.  No loyalty mechanism works for all so ensure it's a coherent holistic programme.

Fifth, although talk (amongst technologists) is the death of the mobile app it's demise is much over-reported.  Link loyalty to a mobile app to give users a reason to use your app/see your brand and get closer.

Sixth, don't force people into becoming loyalty customers.  Some sites force people into the scheme but a better approach is to allow the customer to complete their transaction and then messaging them about the benefits and offering to retrospectively apply their benefits.  

Seventh, introduce new mechanisms to know your customers. This could be a voucher (with unique code) they type into their app / your website to redeem (and therefore you can link their purchases) or just a  daily code per store (linking them in aggregate)

Business Intelligence / Data Analysis / Big Data / AI Driven Data

Much is made of how retailers should handle data and mine it for value.  However it's best to ignore the how and start with the why.

Why collect and mine data?  Simple because without data we are simply guessing on what's required/wanted.  Who are your best customers?  What age segments do you have and what do they do?  Who returns the most products?  What's the overall value of each customer?

A million vendors/providers will tell you their approach is best and you should do X, Y or Z.  The real answer is that if you're not already collecting data and analysing it then you should be.  

For most retailers (sub-1 billion transactions per year) a simple approach will suffice: a data warehouse (a database in a particular format optimised to collating and analysing data) with visualisation tools if fine.  These days there's no need for new hardware or buying software licenses, instead just use a cloud offering (we love Azure SQL) and a tool like PowerBI (desktop version is free).  You'll quickly see patterns that you haven't seen before.

For larger retailers a different approach is required where applications/tools are more complex and require specialists to provide the solution.  Again each of the clouds offers a different perspective on their solutions and approach.

What's important is to get the data into a form that it can be uniformly analysed and segmentations created (we'll write more on this in future!).

But, if you're using google analytics for your web traffic and you haven't added (effective) Extended Commerce tagging then you need to do that ASAP.  We've seen the pipeline optimisation from this make large differences to client's bottom line (and improvements to a customer's journey).

Marketing Automation and Targeting

A lot of retailers are still using "naive" / simplified marketing approaches: they attract customers/visitors to sign up for marketing emails and then they market to them.

Better retailers take data feeds from the website views/behaviours and plug that in.

Good retailers also pass their offline sales into the marketing pipeline.

Premier retailers have done all the above and included segmentation/persona and reward decisioning; and have included data around the communications sent/presented and the efficacy and efficiency of those mechanisms.

Tie your marketing to your data: feed in your data but pre-process it so you have control of the decisions/cohorts/segments and personas.  Only then can you understand how best to speak to a customer and their likelihood of response (and hopefully positive behaviours).

Marketing is often seen as a dirty word in IT.  However marketing's goal is to give each customer the right information to enable them to buy the right thing from you (and avoid showing them things they don't care about).

Offline Experience in Stores/Shops

Very few companies are effectively using their data to ensure that customers get the experience they wish.  Manufacturer/OEM/publishers are still dictating eyeballs and positioning of product and this is bolstered by a low data capture to optimise the approach.

Contrast this with online stores where customers have a personalised approach to put them within reach of products/offers they want to have.  Plus they use experimentation (A/B or multi-variate) to try different options to see which works best for each customer.

We've helped retailers analyse their data from tills and mapped this to customers and their demographics.  Mapping this to personas/segments quickly presents trends and customer wants.  The trick is how to do this effectively.

In a futuristic world we could re-create a shop around a customer: re-arranging products and layouts to their individual needs.  This is of course impossible (at present?).  Instead the layout is a collection of compromises to accommodate customers' wishes and wants.  However that layout needs moving based around new data entering the system.


Clienteling is another area where offline retailers can massively increase their relevance and footfall.

Often retailers are spending time on the question of how best to get a customer to be identified automatically when they enter a physical or online store.

Using Clienteling a customer could get personalised recommendations on their app/website.  For fashion we often see retailers presenting a flat picture of a product however creating 3D representations of the garment on a dummy/mannequin is possible and gives a better view (potentially reducing returns and associated costs too).  Even better a video/animation of a real person of a similar build.

Clienteling identification is harder: we need to identify the customer before they get to a till/POS and we need to decrease transaction costs (personalised shopping using sales assistants is amazing for the customer but very very expensive for the retailer to run.. and has scaling issues too).

One approach is to give clienteled customers a bespoke approach by discounting/giving other offers which allows us to understand that customer and what they've done (or not done!).  

Another is to lead them in-store using their app ("hey stuart, those trainers you wanted are in the green section of the men's department - go to the back of the Reading Broad Street store and you'll see them there")

Alternatively you could just offer to post the product to them to try (with the offer for free returns to their nearest store - thereby increasing footfall and hopefully related product sales).

Online to In-Store (offline) and vice-versa

Few retailers have managed to span the offline and online to best effect (John Lewis is a good example of success).

Where brands have unique products or a loyalty proposition that creates stickiness and retention then using physical stores as the viewing platform and coupling that with online is a great way to go.

Customers need an easy mechanism to establish what they want in-store (i.e. save their shopping list/wish list) and then get more information when they get home.

We're working with a client on a new system that allows their customers to save their in-store interests, it maps the customer's data against their insight system then notifies the customer when they get home that more data has been collected that's pertinent to them and allows them to purchase easily from their mobile.  The key is personalisation.  For example a young person isn't interested in the washability of a sofa but a family with young children definitely is..


Back end systems are often ignored as they're not customer facing.  However order replenishment, fulfilment, re-fulfilment, product management (including the "PIM" and "ERP") and lifecycle provide excellent opportunities to reducing the cost per transaction.

Most offline retailers have a hodge-podge of systems that have accreted over time.  Worse still they have a number of "experts" within the organisation who provide the intelligence and insight for ordering/fulfil/new product take-on.

Without measurement and experimentation we can't judge the efficacy of "experts".  Experts have often been with the retailer a long time and can give past experiences/lessons that denote that if X then Y.  However this doesn't mean they're right.

Ironically the expert's use of the systems/applications is the ideal starting point for optimisation.  They've spent years working out how to subvert these applications and how to make them follow their reasoning and analysis.  This allows you an opportunity to switch from expert-driven decisioning to data-driven decisioning.

We created a data warehouse ("Insight" - because that's what it provides!) for a client.  We fed in both offline and online purchase and viewing details; offer/advert presentations, view and clicks; customer's profiles and the product, ratings and store data.

Firstly we established that the meta-data against products was too poor (for example a lot of books had no genre, author, imprint/publisher, release dates etc) so that needed enriching.  Using that data we can quickly cluster customers according to their (general) behaviours and the lifecycle of various books (popular science books start slow before selling more, romance novels tend to sell more in the first week or two.. and for some authors they can buck these trends).

That simple data was used for prediction (e.g. we have a new romantic novel by a popular author - what will day 1 sales look like and how about over the next few weeks?) so we could define a re-fulfiment/stock shipping model.  Guess what?  We beat the expert for all but the most obvious associations..

Using that data, plus the expert view, we can put the data into the systems (just like the expert previously did based upon gut instinct) and use that data-driven approach to improve the lean/just-in-time fulfilment options and the return-to-manufacturer policy based upon individual products.  This reduction in cost-per-transaction saves less than a percent per product but across the estate it led to an ROI in less than a year (and a bedrock / foundation for future improvements).

Augmented Reality ("AR")

A lot of experts (including Gartner) believe 2019 is the year for AR to boom and get general usage.

As early adopters in AR and VR we're not so sure.

AR has the potential to reduce the friction for some products but for others it's become a worthless "must have" that doesn't increase customer loyalty, custom or retention.

One good use of AR technology is as (what we called in the old days of the web!) a "link magnet": it exists to get more people interested in your brand/stores/online shop and to communicate that to their social network.

However you need a brand where you can persuade a customer to get their phone out to try something and you must never force them to download something to progress.  Lastly you need to understand that while people are standing around in your store using your AR experience: they aren't buying anything (yet!).

Another good use is the augmentation potential.  Some details around products are hard to show physically in a store: ski wear/camping gear has a lot of data alongside it that's hard to condense near the product.  AR can explain that.


Both in-store and online retail has changed and will continue to change.

The successful retailers will follow the times and concentrate hard on the customer and their cost per transaction.

Most changes aren't massively expensive or onerous but can lead to tremendous ROIs (and potentially the chance to be around in 2020 and beyond).

As ever we welcome your comments and discussion!





About the author

Stuart Muckley

Stuart Muckley

I’ve been a programmer and IT enthusiast for 30 years (since the zx spectrum) and concentrated on AI (neural nets & genetic algorithms) at University. My principle skills are concentrated on Enterprise and Solution Architecture and managing effective developer teams.

I enjoy the mix between technical and business aspects; how technology enables and how that (hopefully) improves profit/EBITDA & reduces cost-per-transaction, the impact upon staff and how to remediate go-live and handover, and risk identification and mitigation. My guiding principle is “Occams Razor” that simplicity is almost always the best option by reducing complexity, time to build, organisational stress and longer term costs.

comments powered by Disqus

We're Honestly Here to Help!

Get in Touch

Got questions?  We're always happy to (at least) try and help.  Give us a call, email, tweet, FB post or just shout if you have questions.

Code Wizards worked with us a number of initiatives including enhancing our Giving Voucher proposition and fleshing out our Giving Street concept. They provide a good blend of both business and technology expertise. We are grateful for their contribution to TheGivingMachine allowing us to improve our services and enhance our social impact

Richard Morris, Founder and CEO TheGivingMachine

Code Wizards are a pleasure to work with.

We've challenged their team to integrate our cutting edge AI, AR and VR solutions and always come up trumps.

I love their approach and help on strategy.

Richard Corps, Serial Entrepreneur and Augmented Reality Expert

Working In Partnership

We're up for (almost) anything.

We believe that life is better in co-op (and if we can't help then we probably know somebody else that we can direct you to).  Get in touch.