Monday, November 28, 2011

Why Woolworths Homeshop Needs a CRM (And Maybe You Do Too)

For those of you from foreign shores Australia, basically, has two supermarket chains (Woolworths and Coles). Both offer online shopping and delivery (for a charge). I use Woolworth’s Homeshop and while, in my experience, the range is a little more limited, the costs are a little higher and the quality a little inferior, relative to the the physical stores, the benefits of time saving and convenience outweigh the negatives.
All was fine until we moved house a few weeks ago when the problems began.


Late Deliveries

Woolworths provides customers a range of windows for delivery. One Monday we picked 6am-8am so that the delivery was part of the usual morning drama of getting the children and ourselves ready for work. 8am came and went and nothing turned up. My wife called and asked what had happened. She was told the bad weather and traffic had delayed the delivery and it would be there soon. At 8:10am it turned up. A minor delay and a little disappointing but these things happen. I wrote it off as an outlier.

Eight days later it was time for the next shop. Again we picked the ‘morning madness’ slot. 7:45am and nothing had arrived. My wife’s spider senses began tingling so she called Homeshop. She was assured the driver was on his way and would be there by 8am. 8am rocked around and still no groceries. My wife called again and was told the driver was running late and he would be there at 8:10am. Nothing at 8:10am and she called back. She was told the delivery was delayed and that is the way it is. An apology is apparently not part of the standard customer service.

If you imagine the Hispanic wife from Modern Family with an Australian accent, this is my wife. Respect is a big thing for her. No apology when groceries are late and a cold attitude is not the way to go with her. She was told the next delivery will be free and asked if there anything else that can be done. My wife asked to speak to the warehouse manager so she can give him some selected thoughts on improving his operations management. After all, she has knowledge of operations management from her time at a Fortune 100. The customer service operator refused and suggested she get her manager to call later. This was accepted. Unfortunately, the customer service operator failed to organise this call.

With two late deliveries, being generous, I put this down to a coincidence.

Eliminating the excuse of traffic, we decided to opt for a Sunday delivery. Sure enough, the clock chimes 8am and no delivery. My wife, again, puts in the call and receives an apology from the operator who assures her it should not happen. Next delivery will be free. My wife tells the operator she values her time more than a nominal delivery fee, suggesting she will give the business one more chance before moving to Coles.

The groceries turn up as she puts down the phone. Unfortunately, no checklist is provided and some foods have been substituted for cheaper alternatives (despite us ticking the box ‘no substitutes’ on the online order and no discount given for the inferior product).

Three in a row is now a pattern and suggests something systemic is involved.

Where Are The Problems And How Can A CRM System Help?

To see where the problems are, let us look at the steps in the process:
  1. Customer orders online and selects a delivery window
  2. Goods are selected from a warehouse and loaded into a truck
  3. Driver takes goods to home
  4. Goods arrive late and customer phones up customer service, receiving mixed service
Let us also summarize the issues experienced in the three deliveries:
  • Goods were consistently late, despite a two-hour window
  • Customer service did not know where the trucks were
  • Customer experience from the call centre was mixed
  • No reliable escalation process
  • Cheaper substitutes provided despite the order specifying not to
So how can a CRM system help?

Traditionally, CRM systems were about sales automation; helping sales people track sales opportunities and converting them to sales but, these days, they cover a much wider range of processes than just sales.

In the case of Woolworths, a good CRM system would link the online ordering system with the warehouse and the customer service centre. It would also allow for better management of the process of delivering excellent customer service. Let us review the specific issues.

Late Goods

This one is probably beyond a CRM system. Given the consistent inability to deliver on time to our new address, relative to our old address, I can only conclude Woolworths have a predefined route for the area and we now fall towards the end of that route.

Where a good CRM system could help would be in reporting on complaints received and determining the cause. It is likely others, nearby, also have the same experience as us and in the effective capture of these complaints and consolidation of reporting through a business intelligence (BI) system this issue could be identified and remedied through a review of the route or through the hiring of more drivers.

A cynic would suggest that Woolworths are willing to endure some level of dissatisfaction in order to maximize profits but I am sure this is not the case, given the high level of competition with Coles.

Customer Service Not Knowing Where The Trucks Are

In these days of ubiquitous GPS systems, it would be a relatively simple procedure to link a GPS tracking system with a CRM system via wireless internet, allowing the call centre to know where the trucks are at all times. Failing this, another option would be electronic signature for deliveries, similar to what is experienced with many courier companies (Homeshop still has a paper-based signature). The electronic signature would be transmitted back to the CRM system and the call centre, presented with an ordered list of deliveries, would be able to see where the truck is up to and estimate a delivery time (or the CRM system could estimate for them).

Mixed Call Centre Experience

There is little excuse in today’s world for an inconsistent call centre experience. Most CRM system provide a scripting facility and the ability to automate the process of servicing the customer. In my opinion, it is clear the Woolworths Homeshop call centre do not have access to standard procedures for dealing with common complaints and, unfortunately, have to improvise to the detriment of the customer.

No Reliable Escalation Process

It is often the case that call centres are geographically separate from management so it is not practical or possible for the call centre to escalate a call directly to a manager. Therefore, it is sometimes necessary for a call centre operator to request a customer be called back by management to deal with an issue more appropriately. Where the system breaks down is if there is not a simple system to action this escalation. Most CRM systems have a workflow engine, allowing the automation of activity as the system is used. In this case, the call centre operator could tick a box in a CRM system and the system would take care of the rest; contacting the most appropriate senior member of staff, based on the type of complaint. Even then, if the issue has not be addressed within a fixed time period, the system automatically escalates the issue further. Most call centres, meeting international standards, have such systems in place as well as well-defined service level agreements (SLAs).

Cheaper Substitutes Despite Instructions Otherwise

In this case there appears to be a breakdown between the online ordering system and the warehouse picking slip. Given substitution, when it does happen, appears to be a cheaper product, the same cynic from before would suggest there is a conscious decision, on behalf of Woolworths, to ‘try it on’ and occasionally slip in a cheaper product to improve profit but, as before, I doubt this is the case because the cost in customer satisfaction would massively outweigh any benefit in a few pennies extra profit.

If the picking slip from the warehouse was generated from the same CRM system as captured the online order, it would be a simple case of ensuring the ‘no substitutes’ option was displayed in large font on the generated picking slip.

Taking it a step further, let us assume a barcode reader is used for inventory management, scanning the items as they are removed from the warehouse to update the ERP system. If the CRM system presented an on-screen picking slip, rather than printed out, it would also be possible to link the scanner to the CRM system so that, when a substitute is picked for a ‘no substitutes’ order, the warehouse operator received a warning so they could fix the mistake or automatically generate a credit, email to the customer etc.


Ultimately, people implement CRM systems to improve communication, either internally or externally. In the case of Woolworths Homeshop there appears to be communication issues between the online ordering system, the warehouse, the delivery drivers and the call centre. A consolidated CRM system has the potential to address these issues, improve customer satisfaction and provide a competitive advantage. Let us hope Woolworths starts reviewing their process before our next order, otherwise we will be going to Coles.

The challenge for the rest of us is to look at our own businesses and consider whether we also have communication issues, which affect our ability to deliver excellent service, and consider how technology can be used to improve the situation.

Saturday, November 19, 2011

Salesforce’s Third Quarter Results (and fixing up a small mistake)

Salesforce has just released their third quarter results. This gives me an opportunity to see how they are tracking financially but also to fix up an error in a previous post. Back in September I had suggested that salesforce was getting away from Microsoft in terms of subscriptions. It turns out my ability to combine tables was in error and thus the conclusion that salesforce was moving away from Microsoft was also in error. Let us delve into it shall we?

The Third Quarter Financial Results

For the details, you can go here.

My last review of the financials was back in September. At that time, salesforce had made a loss for the first time since the start of 2009. Unfortunately they have now made it two quarters in a row. The loss is smaller than before, but a loss regardless. If salesforce do not make a profit of more than $7.5m in the last quarter, they will make a net loss for the year. Marc should make some calls into the North Pole. They have a bunch of folk up there trying to track customers worldwide and delivering goods to them via overnight delivery. They might need some help in the next couple of months.

Here is the graph of the financials.


Those non-GAAP revenue costs appear to be tapering but those pesky operating costs keep growing. If only salesforce did not have to operate a business (improve the product, pay commissions, paint the world in adverts, run offices etc.) they would be making a fortune.

I again embraced the pleasure of listening to the earnings conference call and no one bothered drilling into the loss. Marc was certainly unconcerned, but this is no surprise. The fact that the analysts on the call were not overly concerned either was surprising. The analysts have noticed the slowdown in billings growth but that is about it. So, like a water balloon left on a running tap (faucet for our American readers), it is all still about growth and not about the long term consequences.

Subscription Numbers

Unfortunately, salesforce no longer report subscription numbers with their quarterly results. However, the revenue results can give us a clue. Here is the obligatory table of financial results.


2011 Q1

2011 Q2

2011 Q3

2011 Q4

2012 Q1

2012 Q2

2012 Q3









Revenue Cost








Operating Cost








Salesforce Income
























Revenue Per Subscriber








Revenue PUPM








Revenue Growth ($) yoy








Revenue Growth (%) yoy








The first four rows and the Revenue Growth ($) are in the thousands.

Numbers in red are my best guesses using the average company size as an indicator of subscriber numbers.

In this last quarter we have had to be a little smarter because the customer numbers were not released. In this case I have used the Revenue Per User Per Month which has tracked at around $50-51 for the last 18 months. Combining this with the Revenue numbers gives us a number for subscribers and, again, using the predicted average company size we can get a number for customers.

Assuming things are on track, salesforce is poised to make four million subscribers before the end of the next quarter.

Fixing Up My Market Maturity Numbers

As mentioned, back in September I made an error with the market maturity numbers which led to the conclusion salesforce were getting away from Microsoft. Time to fix that error. Here is the corrected table.

SFDC Subscribers

SFDC Customers

MSFT Subscribers

MSFT Customers

Subscriber Ratio

Customer Ratio

Difference in Subscribers

Total Subscribers

Customer Size Ratio
Mar-06 423,667 22,298 150,000 6,000 2.82 3.72 273,667 573,667 1.3



























































































Essentially, the error came because salesforce quote their financials by fiscal year. So the most recent results are Q3-2012 even though we are still in 2011. This led me to put the previous years subscription numbers in for salesforce in all but the last quarter. I also misunderstood when the salesforce quarters started. I had assumed fiscal year started on the first of January. It appears it actually starts on the first of February so this has also been adjusted. The above table, as far as I know, is now correct.

Again, the numbers in red are best guesses based on the information at hand.


The total number of subscribers for the two companies continues to grow unabated with it likely that the two products are already serving six million subscribers or more.


The graphs no longer show the jump seen previously. We now see that either the ratio of subscribers is flattening to just below two or the ratio is decreasing and the difference is flattening out. Time will tell which is correct. What is clear is that salesforce are not pulling away through the subscriber land grab acquisitions as previously speculated and Microsoft are still in the chase. Honestly, nothing would please me more than another drop in the subscriber ratio and a reduction in the subscriber difference. I cannot wait to see if Santa Marc will package that up for me in the next quarter results.

Google Trends


Here is the latest difference between “dynamics crm” and “” according to Google trends. In the past I have used a linear trend line but I do not feel this accurately represents the movement (with an r-squared value around 50% its presentation was unjustified). To this end I have changed the trend line to a moving average graph, akin to what is often used for stock prices. The advantage of this kind of trend line is it smooths out the fluctuations and still gives an indication of the general direction of movement. The graph indicates Dynamics CRM continues to gain mindshare amongst the Google search population.


My first conclusion is avoid crunching numbers in the wee hours (although I am writing this at 2am so I am not good at taking my own advice). Moreover, as any good theoretical physicist will tell you, if you do crunch numbers and a strange result pops out check and re-check to make sure you are not going to put something out there which is a nonsense.

In regards to the salesforce numbers, I predict that salesforce will make their four million subscribers by the end of the year and this year will result in a financial loss for the company. I often say “one is an outlier, two is a coincidence and three is a pattern”. Let us see if salesforce can turn their financial losses into a pattern next quarter.

As for accelerating away from Microsoft, this is far from the case. Microsoft continue to pursue salesforce like a dog chasing a car. Whether the car will stall or shift into second gear is yet to be seen.

Finally, Microsoft continue to gain public mindshare and this does not appear to be slowing. While it was close in July 2011, we are yet to see a day when more people search for Dynamics CRM than Appearances would suggest that day, however, will be soon.

Monday, November 7, 2011

Socialmention Battle Arena: Battle of the CRMs

I came across this tool called socialmention and thought I would give it a bit of a plug. Essentially it is a free social analysis tool covering quite a wide range of sources. Here are the results of the vanity search:


Most of it is pretty self-explanatory. The four measures in the top left are as follows:

  • Strength: Likelihood of keyword being mentioned socially (turns out I am not mentioned very much)
  • Sentiment: Ratio of positive to negative mentions (this is a pretty dodgy measure as it appears to use keyword matching to determine the sentiment of messages)
  • Passion: Measures how likely people are to repeatedly talk about the keyword. For example, if it is a small group of people that constantly mention ‘leontribe’ this will have a high passion score even if the relative Strength of the output is low.
  • Reach: A measure of influence

They also offer a service similar to Google Alerts, although I have not been able to get it working yet.

Battle of the CRMs

Using the CRM’s mentioned by Gartner we get the following results (I used all quadrants except the bottom left): Dynamics CRM Oracle CRM Siebel CRM SAP CRM SalesLogix
image image image image image image

The Winners

In terms of likelihood of being mentioned (Strength), SalesForce is the winner, closely followed by Dynamics CRM. Except for SAP CRM, the others are far behind.

For sentiment, SAP CRM can do no wrong but, as mentioned I am somewhat sceptical of this measure. Any system which gives Hitler and Osama Bin Laden a three positive to one negative ratio has got limitations.

For passion, look no further than SalesLogix. That small group of social networkers talk about SalesLogix the most frequently. Interestingly, SalesForce has the smallest passion score meaning the people mentioning it are not concentrated to a small group.

In terms of influence (reach), Siebel has the longest arms, closely followed by Oracle CRM and SalesForce. SalesLogix had the shortest arms so while the SalesLogix socialites talk often, they do not talk to a diverse group.

Review of the Keywords

The keywords of the different products also offer some insight. For SalesForce we see their focus on the cloud and for enterprise business to make use of social channels. We also see mention of Benioff, their CEO, speaking at the strong influence he has on their brand. They also mention Oracle, which may be because of the recent circus at OpenWorld.

For Dynamics CRM, the terms are a little more general but talk about office software and a customer focus.

The Oracle and Siebel keywords talk about business software with a strong salesforce automation and marketing focus.

The most interesting, in my opinion, are the SAP CRM keywords which are more about the people who work with the product and their career than about the product itself or the customer.

The rather strange keywords regarding SalesLogix are because of a recent press release put out by Sage regarding one of their customers.


If you are looking for a free social analysis tool, this is not a bad place to start and, while it is not as comprehensive as many of the non-free alternatives it is a good way to get an idea of what a brand is up to. If you are feeling bored, you can also play the Social Mention negative sentiment game where you (and as many of your friends as you like) try to find ANY keyword phrase which has a negative sentiment score i.e. more negative mentions than positive (it took a colleague and I about an hour to find one). If they are family-friendly rated, feel free to add them to the comments Winking smile