When a retailer or brand decides it's time to replatform their ecommerce website, many will opt to embark on an ITT process (Invitation to Tender).
This can involve two or three distinct phases: RFI (Request for Information) and RFP (Request for Proposal).
The RFP is often huge: I've waded through several that have made my eyes bleed due to their complexity, all the time thinking that the poor souls that have slaved over the incredibly detailed document have somewhat missed the point.
My question is, therefore: Is the RFP an effective mechanism that best serves the involved parties?
I've felt for a long time that perhaps it is not, and in 2010 Forrester research produced a paper entitled ‘A Scenario Based Ecommerce Technology Evaluation Process’ on the subject suggesting there might just be a better way of going about things.
Over the course of the last year or so I've had conversations with several ecommerce agency owners who have talked to me about issues they have with the majority of RFP's they receive, namely:
- Overly prescriptive nature: 'the system must work exactly like this…"
- Homogenisation of technology: 'the current system works in this way, the new system must replicate the way in which the current system works'.
- Overly detailed, rigid descriptions of every single feature.
- Conversely, not enough information on others.
- Little or poor attempt to describe the business problem, therefore doesn't allow a respondent to show how they can address it.
In an interview with The Register, Ian Birdsey of Pinsent Masons, the law firm behind Out-Law.com, said that organisations often make the mistake of trying to replicate outdated or inefficient processes and work methods.
He says organisations that are overly prescriptive in their briefs to suppliers risk not getting their desired outcomes when updating IT systems.
On a lot of levels it does not make sense to replicate old methods and processes with a new system. IT customers should challenge existing processes to determine whether they are best for them, otherwise the benefit of changing IT systems can often be diluted or lost altogether. Smart organisations use transformational IT projects to introduce more efficient and effective business process change.
Not everyone dislikes, however. Bob Curran of Buy4Now tells me he generally likes them and has historically done well via this format. In one instance recently Buy4Now managed to pip one of the 'big three' and solve a monster problem for a large retailer.
In light of this, I think a distinction needs to be made between a good and a bad RFP.
- Kitchen sink approach.
- Overly detailed descriptions of features.
- Not enough information on certain aspects.
- Online tender.
- Fixed format for responses.
- Inability to engage and ask questions.
- Poor understanding or communication of business issues.
- Asks the 'wrong' questions.
- Not fixated on rigid processes or features.
- Focused on achieving outcomes.
- Open format for responses.
- Ability for respondent to formulate and ask questions.
- Asks well framed questions.
- Good understanding and communication of business issues.
Prescriptive and homogenisation aspects can be viewed together. Whilst a business may have business issues that are causing it to replatform it may wish for the bulk of its operations to remain the same and so require any new technology to facilitate certain processes.
However, specifying that something must work in a very specific way might rule out another very good option, or mean costs are escalated unnecessarily through customisation work.
Specifying every single requirement in detail is a monstrous task and a dangerous one too. If a vendor or ecommerce agency follows this to the letter, which is surely the intention, then nothing can be omitted or incorrectly specified. There is no room for error or interpretation with this ‘everything but the kitchen sink’ approach.
Those compiling the RFP may be unaware how different ecommerce platforms operate and it's very difficult for them to fit the platform to the process (or vice-versa). The more complex the business, the harder this becomes. Sometimes the only way to describe a desired feature or process is by referencing the way things work currently.
Because a retailer (and most consultants) most likely won't have perfect knowledge of every platform, vendor, or agency, it's incredibly difficult for them to say what's going to be best for them, and it is, therefore, nigh on impossible for them to know what to ask for – this is a really key point.
It's very possible for the client to omit features from a spec, which can cause major relationship problems down the line.
Maxwell Lamb, co-founder of ecommerce agency blubolt says:
This is a harbinger of doom - the number of times we've seen RFPs with 2000 one-line rows in an excel spreadsheet with "Yes" and "No" with indication of importance in columns, with a single row that just says B2B.
A better way?
Firstly, replicating a feature should not be confused with facilitating an outcome or a process - there is a huge difference.
Forrester recommends the formulation of 3 to 4 well planned scenarios that encompass all touch points of a business.
Personally, during my time as an intermediary between brands and agencies I found a non-prescriptive approach worked very well for my SME and midmarket clients.
My favoured process took a broad, outline approach to initial scoping, helping the client to take a step back and think about the customer, the administrator, the marketer and the customer service people, for example.
I would encourage them to come up with a short scenario that encapsulates the day to day workings of each touch point. All subsequent agency demo's and conversations with the retailer or brand are then firmly focused on the key issues, whilst retaining individuality and allowing creativity.
One well thought out scenario can encompass a large number of processes and required features, and allow freedom of expression in each response. This approach allows the retailer to gain a much better insight into an agency or platform vendor too, especially valuable in instances where agencies or SI's all use the same platform.
Here's a great example of a fictional multichannel scenario I really like, adapted from the HBR:
Amy, age 28, needs resort wear for a Caribbean vacation. She starts shopping from her couch by emailing customer services at Danella, the retailer where she bought two outfits the previous month. The representative recommends several items, emailing a link to superimposed photos of them on Amy’s avatar in the member’s area. Amy buys one item from Danella online and then drives to the Danella store near her to collect her order and try on the other in-stock items she likes.
As Amy enters Danella, a sales associate walks her to a dressing room stocked with her online selections—plus some matching shoes and a cocktail dress. She likes the shoes, the dress is daring and expensive, so Amy sends a video to three stylish friends, asking for their opinion. The responses come quickly: three thumbs down. She collects the items she wants, and checks out with her smartphone.
As she heads for the door, a life-size screen recognizes her and shows a special offer on an irresistible summer-weight top. Amy quickly checks her budget online, smiles, and uses her phone to scan the customized Quick Response code on the screen. The item will be shipped from another Danella store to her home tomorrow.
Can lessons be learned from the persona model pioneered by the Eisenberg brothers to understand and influence motivation in a purchasing situation? I think a similar approach will prove invaluable here.
I think written scenarios can work, although Max suggests otherwise:
I would actually argue that the assertion in the Forrester report that stipulates ‘Communicate these scenarios to the vendors in written form’ is wrong. I think it would be better if vendors were only told general areas in advance of an interactive session - far more effective if you put them on the spot and see how well they understand their own tools, and how to use them in context.
To wrap up…
Storytelling is the future.
More and more agencies and vendors are deciding not to respond to hugely time consuming, prescriptive RFP documents and so retailers, brands and any consultants brought in to help need to show an acute awareness of the marketplace so as not to rule out perfectly suited options.
There’s a place for tiny detail, but leave this for the functional spec.
A more open, flexible approach is better for all concerned and will add tons of value. It's far more inclusive, prevents homogenisation of technology, and allows technology and service remit to serve the business needs as opposed to mimicking an incumbent setup that's being replaced.
The playing field should not be levelled, and the concept of comparing like for like should be banished. Rather than the technology per se it's more often the capabilities of a provider and after-sales service that are the deciding factors where selection is concerned. These should be allowed to be fully demonstrated and not stifled.
Detail should be arrived at, not started with – we have BDD (Behavioural Driven Development), let’s start utilising BDS (Behavioural Driven Scoping).