20 Years. An agency systems perspective.

Chris Lever, Founder, Agency DNA

My first UK marcoms role was in 1995 as agency FD of Bernard Hodes, an Omnicom/DAS group company. Obliged to conform to the rigour by which DAS managed its portfolio and deliver rapid improvement in their two key pertinent drivers, profit margin and the working capital ratio, I was aghast at the way DDS, the group imposed agency media booking/billing, job management and accounting system, made getting these metrics a major exercise. A single minded Kiwi Financial Controller who I’d hired to assist me had words I couldn’t even begin to pronounce for a system which offered not only opportunity for one-sided entry, but produced a balance sheet by downloading disparate ledgers and combining them in Excel.

chris.jpg

Fast forward 20 years, DDS rebranded as Media Ocean, rebuilt its database, and yet seems it still can’t handle supplier invoices with more than six characters, or allow a major network create a new field to hold the PO references of Google and Facebook, who quite rightly have their own booking systems. If a network with two billion pounds passing through its pay-as-you-go ledgers doesn’t get what it wants, what hope might a smaller agency have? Excel is a strong supplement.

Step back fifteen years and I’m on a team with other senior players from BMB&B, Leo Burnett and  BBH collectively giving up weeks of our lives to work with Maconomy, guiding them on the development of their agency solution on top of what was admittedly a complex, yet robust financial system.

Youthful and optimistic this was heralding a golden age where a Scandinavian vendor understood that agency complexity is a state of mind, we then waited for the release of a system that would look as good as the Power-Point slides suggested, give clear and instant visibility over the inner workings of our agencies, enhance our staff’s lives and make the coffee. Strange I have since felt a strong compulsion to apologise to the many Maconomy users who in those early days found themselves short of utopia. It was more like the Underworld.

Talking it through afterwards with Morton, the Danish Senior Developer, trying to understand what happened between stating our reasonable requirements and the end product, it seems back in Copenhagen they re-thought our needs, and built as only they knew how. Something got lost in translation.

Undaunted, this was the start of the new millennium and things would only get better.

The thing with Maconomy is you’re buying a big box of tricks, and you need to know what you’re doing, have a very detailed spec, patience and a frothy budget.

A range of separate and colourful workflow systems have been invented to fill the many gaps left by Maconomy, Deltek even brought one themselves, but the reality is Maconomy for many ends up an expensive and under-utilised general ledger and accounting system.  Excel reigns supreme.

In all this time, like a long-life battery powered toy soldier, Rebus, now Paprika has quietly and defiantly marched on to a steady beat. A Cobol-based solution developed almost 30 years ago, it was the go to system for smaller agencies and despite attempts by SAGE, QuickBooks and latterly, Xero, the NZ originated cloud system, in many ways it still is.

Paprika has never made a big show of itself, yet through the passage of time and the reluctance of agencies to change systems, it has built a tidy market share of around 1000 agencies, a handful who are global. Despite being clunky and not so easy to use, it’s seen as a safe bet.   

A major drawback though is it is a one currency database system. What this means is as soon as you open your first offshore office, you need two databases. Imagine you have 20 offices in 14 countries. That’s a lot of databases, and even more Excel.  

Today the large networks still angst over systems, systems like Maconomy were seen as a great leveller. Mid-tier platforms for standardising multiple agencies, before moving upwards onto something that allowed greater flexibility in global client billing and reporting hierarchies, and migration to some form of shared service. Indeed this was my own strategy in deploying it into WCRS, now The Engine Group, following our MBO in 2004.

The problem is upwards moves to say Microsoft Dynamics AX or SAP come with their own monumental challenges, and matching price tags. A litany of issues prevail in the network attempts with these systems.

For SME’s, the great hope is the cloud.  

New agile systems, built from the ground up overcome many of the limitations sitting in systems made in the past. Agencies can implement without major cash outlay, retain flexibility and enjoy inbuilt scalability. Cloud usually means seamless and automatic upgrades, allowing users to benefit from continuous system improvements. Never again be threatened with an unsupported version.  

The better cloud solutions have open API’s, allowing true connectivity with other business tools, CRM, CPM, and BI, fully embracing the notion that whilst they may do more, they also do it better. The really good ones integrate all the features of job management, accounting and resourcing – the benefits of which are staggering.

There is still no single panacea for the world of marcoms business management and reporting intelligence, and in a tight margin and competitive digital world, agencies need to find their own differentiators.

Assessing what is important to agencies, the information needs at all levels, the degree of flexibility and autonomy management wants to allow, requires a matching to increased number of suitable systems of today which can offer exactly that. Even Excel now competes with Google Docs.

Which all means it is a really interesting time for agencies.

Overall in the past 20 years agencies have not demanded more, and have been fearful of change, even at risk of continuously adding overhead to get basic information circulated. Accepting compromise has unfortunately ingrained vendor complacency.

Yet for many of the new breed of agency, client demands, the general increase in the variety, scope and speed required of work, and the associated massive increase in the volume of data, compromise and complacency can no longer be tolerated.

The good news is there is now choice. The cloud has brought options that are fully featured, continuously improving, intuitive, easy to use, cost effective and quick to implement.

Agencies fear making decisions on systems, but change should be embraced. As a famous person once said, “Not to change, is to die.”  

So the future for agency systems is indeed bright and can be whatever colour you choose for your management dashboard.

 

How Long Should Your ERP Implementation Take

By Chris Lever

I've been advising agencies on their system selection for many years now, and have led a number of full cycle implementations - that is the design, development, testing and training, go-live and support.

By default, an Enterprise Resource Planning ("ERP") system is for managing agencies, or any people-based enterprise, quintessentially to capture and track time against quoted fees to ensure against over-delivery, hold brought-in third party costs against specific job (or project) records for visibility over what cash has come in from clients, to what's going out to suppliers, and be the secure store of all accounting transactions associated with this and the other activities of the business.  

For many agencies the prospect of upgrading or changing from an existing set-up is fraught with challenges from the cultural resistance to change, to the unknown quantity of what will the vendor actually deliver, in what time frame and at what eventual cost.

It's probably accurate to say a majority of agencies are happy to live with compromise (and spreadsheets) until some driving factor such as the arrival of an enlightened FD with a mandate for change, a vendor contract expiring, or the system version in place is no longer supported, that the difficult questions of a new ERP implementation or upgrade have to be answered. 

So how long should it take?

Well, that does depend on many factors, but here are a couple of examples I can attest to:

UK creative agency, 25 people, new company, so minimal data migration.Duration: 6 hours (4 on-site); System: AccountAbility, a cloud ERP; 1 external consultant.

UK digital agency, 90 people, full data migration, including 1,000 live-jobs.Duration: 3 weeks; System: AccountAbility; 1 external consultant.

UK independent group, 120 people, partial data migration. Duration: 3 months; System: Deltek/Maconomy, a server based system; 2 external consultants.

UK independent group, 650 people (14 offices globally), full development & data migration. Duration: 2 years (and ongoing); System: Deltek/Maconomy; 3-4 external consultants.

UK mid-cap plc, 700 people (20 offices globally), full development & data migration.Duration: 15 months; System: AccountAbility; 1 external consultant.

Of course every agency is governed by their specific sets of circumstances which contribute to making each implementation unique, but at the end of the day the"processes" agencies follow are fundamentally the same.

The dependencies too are all pretty generic, and there's plenty of better documentation than I can provide on the core principles of good ERP project management, but here's a couple of practical tips:

1 System-owner engagement. A strong internal driving force who remains involved and interested is a big factor in helping meet objectives on implementation timing. Someone who understands the system's inner workings, can lead, communicate, gain buy-in and mentor the business all come within this mandate.

2 Experienced project team & competent vendor (or partner) consultants. It helps greatly if the people who work together on the detail and the day-to-day driving forward of the project, know what they're doing and are able to operate within clear guidelines set by the system-owner. 

3 Keeping it simple. Agencies may try and ensure every instance they can think of is covered by the new system, and vendors and consultants will not deter them for complex projects turn into long running projects, with associated fees generated. With few caveats though, I've always endorsed a view of not embarking on substantial customisation, to go simple and "vanilla" first, as any good system will lend itself to modification when it's identified as needed through use.

4 Data in good shape. ERP systems are structured depositories for your business data, and so migrating from a spreadsheet heavy environment will require collating and reformatting of data to fit in these new structures. Agreeing what these structures are can be fraught, and is one of the biggest areas where vendors must be prepared to guide project teams adequately. Most vendors will provide templates for easy upload of data into their system, although likely a bigger challenge may actually be extracting data from your existing system. Tip: Validate the import/export capabilities of your new system. 

Of course if all these remain equal, straight run comparisons on implementation time will inevitably come down to the system itself. 

Server-based systems, like Maconomy, will always take longer. If I can offer an analogy - it's like a large kit (house, car, etc) that comes in a box for which it then has to be put together, piece by piece, each time from scratch. These also need the involvement of IT to agree the hardware & networking specification on which Maconomy will be housed (excuse the pun) and accessed. 

Cloud ERP systems, like AccountAbility, however come pre-defined with best-practice structures, workflows and reporting as determined from experience, which immediately lends itself to rapid implementation. A couple of focused workshops quickly realises a workspace build that meets the agencies need. With browser access, the only involvement of IT is for them to satisfy themselves on the security aspect, then make sure a good broadband connection is available. 

Managing a systems implementation consultancy may suggest we'd prefer long running and complex projects. Well, we're very good at doing them, and like them, sure - to a point.

Our real motivation however comes from getting agencies onto better platforms for growth in much faster time frames, to get their data joined up and them benefiting from the visibility on how their business is performing, and able to make the adjustments they need to be more profitable. This for us means happy clients, and stronger, longer term relationships. 

Final point then, balls on the line - how long should an ERP implementation take?

My rule of thumb. A 100-person agency, going AccountAbility, will be live within one month. If you're going Maconomy, I'd have to have a look and give you a quote.