CASE STUDY: Center Parcs - The data plunge

Sometimes it takes an injection of fresh blood to bring out the best in a company. And while Center Parcs has always been a popular short break destination in the UK, when Nick Begy joined in January 2001 as direct marketing manager, the company had low occupancy at certain times of year coupled with over-demand at peak periods.

A firm believer in direct mail, Center Parcs already had a programme to try and plug these gaps, with major campaigns to its customer database every January and September supported by smaller mailings throughout the year. But still the problem persisted.

So when Begy joined, his aim was to fill up the periods of low occupancy while smoothing out the demand curve in the over-subscribed times of year.

Cutting wastage

It didn't take him long to realise that a change was needed. Looking at the results of just one September brochure mailing, it quickly became clear that the targeting was limited to those who had previously stayed in one of its three UK villages at a certain time of year. "We're talking huge wastage, says Begy. "We weren't identifying who we should be targeting and when. As a result, most of the bookings were actually coming from a small percentage of the total mailed.

The first move was to join forces with data solutions provider Talking Numbers to build models identifying who was coming at what time of year. The major objectives were to focus its targeting, to find the catchment areas for each village, and to establish who goes into which villas. With villas ranging from one to four beds, it would, says Begy, be pointless targeting a family of four with offers for a one-bed villa.

Previously, the holiday provider had identified when and where the gaps in occupancy were at short notice, then sent out postcards to its database to try and fill them. Begy identified a few errors in this approach. "We weren't looking at lifestyle, he says. "We were looking at people who'd booked in September before but we really needed to be looking at who'd booked at short notice before, recognising the stock we had left and identifying who out of these people would be likely to go into them. Basically, it's all about good old propensity modelling.

Working with Talking Numbers, Begy and his team took everything Center Parcs had done before in terms of targeting and turned it on its head, starting propensity modelling at the end of November last year. This has signaled a number of significant changes.

Sorting out the catchment areas has proved particularly useful, especially as April's fire at its Elveden site has left that village out of action for now. Rather than targeting all prospects in this area, Center Parcs can now decide which of Elveden's regulars should be targeted with information on which of the other two sites.

This year Center Parcs has also brought in Claritas's PRiZM Household tool to refine its targeting further by profiling customers and prospects.

The main component of this is Claritas's Lifestyle Database, combined with information such as census and subscription data.

Lifestyle components

To provide a clear picture of customers, information on lifestage and income is overlaid with lifestyle components like car ownership, readership, and cultural interests. This segments households into one of 52 distinct groups, such as 'most affluent empty nesters who like outdoor pursuits'.

Data, such as party composition and number of bookings, has also proved useful in identifying these targets, as David Dipple, technical director at Talking Numbers explains: "We look at the number of adults in a party who book at a certain time of year and use this information to predict what will happen this year and target them appropriately."

Having worked all this out, Center Parcs is now tailoring messages to particular groups such as empty nesters which it recognises are more likely to take advantage of offers to visit out of season, and are also good bets for one bed units.

The September brochure mailing now goes to around 50 disparate segments with each one receiving a different offer and letter.

Not only is modelling helping Center Parcs improve its targeting, but on a deeper level it enables the company to build up profiles of the visitors to each village. These very different profiles are also influencing the way people are now being targeted.

For example, visitors to the Longleat Forest village have a very different profile to other guests. They tend to be from a slightly higher affluence band, with more public school-educated children, which also means school holiday peaks tend to fall slightly differently than at the other villages.

"There are distinct groups of customers, says Dipple. "Elveden has people with big wads of cash, and they spend a lot there. Longleat is the old rich; there's a different atmosphere."

This insight has resulted in Dipple and Begy producing different profiles for each break at each village. These are then given to the villages, helping them to work out in advance if there's likely to be a lot of babies, male groups or teenagers at any one time. These are all factors that can lead to staffing issues. "We can now say that if A happens, then B will follow. It allows us to be more predictive, adds Begy.

This new confidence in targeting has led Center Parcs to give cold lists a go again - something that hadn't been particularly effective in the past. PRiZM Household is proving especially useful here. For this September's big brochure launch, it brought in large quantities of cold lists and overlaid them with the system, nailing prospects down to the exact postcodes Center Parcs wanted to target.

The promotion for the brochure launch is the new spa at Sherwood Forest.

This is something that wouldn't suit all customers so, using PRiZM Household, low affluency groups and people with young children were excluded from this particular offer.

It is still less than a year ago that Center Parcs and Talking Numbers started working together, yet in that time they have thrown up a wealth of data, including some unexpected findings. Apart from the differences in village profiles, the biggest learning so far, according to Dipple and Begy, is that promotions often stimulate more bookings even after the offer period is over.

Unexpected results

This was demonstrated after this year's January's campaign went out.

Postcards were targeted at different lifestyle groups with empty nesters receiving cards focusing on adult activities like the spa, while the pre-school market was sent pictures of babies in the pool or creche facilities.

It reaped rewards during the promotional period but the surprise came when they looked at the number of bookings made after the promotional period had ended. "Ten times as many people who were sent the promotion booked outside the promotional period as within it, so missing out on the discount, says an astounded Dipple. "We hadn't realised this before."

Even better, it's having a huge effect on profits. Although hesitant about actual figures, Begy admits that this particular campaign made 150 times the money spent on it in incremental terms.

This year has been a steep learning curve. "We're constantly refining what we do with every learning, says Begy. So far the DM is working very well, and it's all down to modelling. "By using models, expounds Dipple, "we can get the who, what and when right. So when we have one-bed apartments lying empty in September we know who to contact."

It doesn't stop there. In the pipeline are plans to marry up qualitative analysis with transactional and behavioural information to get an even clearer picture of behaviour and attitudes. But both Center Parcs and Talking Numbers are wary of jumping ahead of themselves too soon - tempting as it is with so much data now at their disposal. The important thing, says Dipple, is to remember what is useful. "We're like kids in a candy store, he says. "But we have to make sure what we're doing is always working properly."